PROMOTING MATHEMATICAL THINKING IN

PHYSICS: SOME THEORETICAL EXPLORATIONS

IN MECHANICS AND THEIR USE IN CONCEPTUAL
LEARNING

BERNARD RICARDO WIDJAJA

NATIONAL INSTITUTE OF EDUCATION

NANYANG TECHNOLOGICAL UNIVERSITY

2020
 Promoting Mathematical Thinking in Physics: Some
Theoretical Explorations in Mechanics and Their Use in
Conceptual Learning

Bernard Ricardo Widjaja

A thesis submitted to the

National Institute of Education,
Nanyang Technological University
in partial fulfillment of the requirement for the degree of
Doctor of Philosophy (PhD)

2020
 STATEMENT OF ORIGINALITY

I hereby certify that the work embodied in this thesis is the result of original research and has

not been submitted for a higher degree to any other University or Institution. In addition, I

declare that to the best of my knowledge, this thesis is free of plagiarism, and contains no

material previously published or written by another person, except where due reference has

been made in the text.

6 April 2020
................. ...........................
Date Bernard Ricardo Widjaja

i
 SUPERVISOR AND CO-SUPERVISOR DECLARATION STATEMENT

We have reviewed the content and presentation style of this thesis, and we declare that it is

free of plagiarism and of sufficient grammatical clarity to be examined. To the best of my

knowledge, the research and writing are those of the candidate except as acknowledged in

the text and/or the Author Attribution Statement. We confirm that the investigations were

conducted in accord with the ethics policies and integrity standards of Nanyang

Technological University and that the research data are presented honestly and without

prejudice.

6 April 2020
................. ...........................
Date A/P Lee Choon Keat, Paul

6 April 2020
................. ...........................
Date A/P Subramaniam Ramanathan

ii
 AUTHORSHIP ATTRIBUTION STATEMENT

This thesis contains material from four papers published in the following peer-reviewed
journal(s) / from papers accepted at conferences in which I am listed as an author.

Chapter 2 is published as Ricardo, B., & Lee, P. (2015). Maximizing kinetic energy transfer
in one-dimensional many-body collisions. European Journal of Physics, 36(2), 025013.
DOI: 10.1088/0143-0807/36/2/025013.

The contributions of the co-authors are as follows:

• I provided the initial idea, worked out all the theoretical backgrounds and formulas.
• A/P Lee Choon Keat, Paul provided guidance in the way the manuscript should be
written.
• I prepared the draft of the manuscript.
• The manuscript was revised by A/P Lee Choon Keat, Paul.
• I did the final revision of the manuscript.

Chapter 3 is published as Ricardo, B. (2016). Comprehensive Analysis of the Failure of

Intuition in Elementary Rigid Body Dynamics. Applied Physics Research, 8(1), 125.
DOI:10.5539/apr.v8n1p125.

The contributions of the co-authors are as follows:

• I provided the initial idea, worked out all the theoretical backgrounds and formulas.
• I prepared the draft of the manuscript.
• I did the final revision of the manuscript.

Chapter 8 Section 8.2.1.3 is published as Ricardo, B. (2015). Using scaling to compute

moments of inertia of symmetric objects. European Journal of Physics, 36(5), 055003. DOI:
10.1088/0143-0807/36/5/055003.

iii
The contributions of the co-authors are as follows:
• I provided the initial idea, worked out all the theoretical backgrounds and formulas.
• I prepared the draft of the manuscript.
• I did the final revision of the manuscript.

Chapter 8 Section 8.2.1.4 is published as Wang, J., & Ricardo, B. (2019). Squashing Method
for Moment of Inertia Calculations. The Physics Teacher, 57(8), 551-554. DOI:
10.1119/1.5131123

The contributions of the co-authors are as follows:

• Wang Jinhui provided the initial idea, worked out the theoretical backgrounds and
formulas.
• I provided guidance in the formulas and the way the manuscript should be written.
• Wang Jinhui prepared the draft of the manuscript.
• The manuscript was revised by me.
• Wang Jinhui did the final revision of the manuscript.

6 April 2020
................. ...........................
Date Bernard Ricardo Widjaja

iv
 ACKNOWLEDGEMENTS

I would like to thank these people for their significant role during my PhD journey:

My PhD supervisor, A/P Lee Choon Keat Paul, for supervising the theoretical physics

part of this study, and my PhD co-supervisor, A/P Subramaniam Ramanathan, for

supervising the educational research part of this study. I am thankful for the opportunity to

pursue this research under their guidance. I have received continuous guidance, valuable

inputs to improve on the quality of research and the thesis content and exceptional

encouragement. It has been a privilege to learn from gifted teachers and mentors like them.

The Thesis Advisory Committee (TAC), Dr Darren Wong and Dr Yeo Ye. Thank you

for spending time proof-reading my thesis and for valuable inputs to improve the quality of

my thesis significantly.

The Principal of NUS High School of Mathematics and Science, Ms Soh Lai Leng

Magdalen, for the support and scholarship given for me to pursue my PhD degree and

opportunity to conduct my research study in the school.

My former school Principals, Dr Hang Kim Hoo and Mrs Lee Bee Yann. Dr Hang

was the one who gave me the biggest motivation to pursue my PhD in this area. As a

mathematician and scientist, he never gives up on his passion, his resilience has become my

constant inspiration. Mrs Lee inspired me to be persistent in learning and discovering new

knowledge, a perfect example of growth mindset. She gave me the drive to complete my

PhD.

v
 My former superintendent, Mr Barnabas Tan Yii Hsien, who gave me constant

encouragement throughout my PhD journey. He was a great supporter to me, especially in

the beginning stage of this research.

The Head of Physics and Engineering Department in NUS High School of

Mathematics and Science as well as my current superintendent, Mr Poh Boon Hor, and all

my colleagues in the Department of Physics and Engineering of NUS High School of

Mathematics and Science, for their tremendous support and interest in the research topic that

I have been pursuing.

My ex-student and collaborator, Wang Jinhui, for his significant contribution to one

of my published papers on the Squashing Method for Moment of Inertia Calculations.

Throughout this PhD journey, we also published two volume book series titled Competitive

Physics, a perfect reflection of our passion for physics.

My friend and colleague, Andre Jusuf, for helping me validate the test instruments

and survey instruments used in this research and to help organise the data collection in our

school. As a long-time friend of mine, it was indeed a privilege to have collaborated with

him for a poster presentation titled Building an Effective Assessment to Test the Applications

of Physics Concepts in Higher Order Thinking Questions in the International Science

Education Conference (ISEC) 2018.

The Assistant Head of Physics and Engineering Department in NUS High School of

Mathematics and Science, Ms Lim Jia Hui, and my ex-colleague, Mr Seow Yongli, for

helping me validate the test instruments used in this research.

vi
 My lovely wife, Dr Yoanna Arlina, thank you for your incessant prayer and support.

As a researcher herself, she has given me important feedback on the research design and

statistical approaches used in the study.

My daughter, Evangeline Scarlett Ricardo, and my son, Reinard Wyatt Ricardo, thank

you for being the sources of my joy and laughter.

My parents, the late Yahya Widjaja and Julianti Widjaja, my elder sister and brother,

Marisca Widjaja and Albert Ricardo. They have been my constant source of encouragement

and support throughout my PhD journey, especially during the births of my children.

Last but most importantly, my Lord and Saviour, Jesus Christ. It is only by His Grace

that I could start and complete this research work, and it is all for His Glory. Soli Deo Gloria.

vii
 TABLE OF CONTENTS

STATEMENT OF ORIGINALITY ................................................................................. i

SUPERVISOR AND CO-SUPERVISOR DECLARATION STATEMENT ................ ii
AUTHORSHIP ATTRIBUTION STATEMENT .......................................................... iii
ACKNOWLEDGEMENTS ............................................................................................. v
TABLE OF CONTENTS .............................................................................................. viii
LIST OF FIGURES ...................................................................................................... xiv
LIST OF TABLES ....................................................................................................... xvii
SUMMARY .................................................................................................................... xx

CHAPTER 1: INTRODUCTION.................................................................................. 1
1.1. Research Background .......................................................................................... 1
1.1.1. Mathematical Thinking in Physics............................................................... 1
1.1.2. Researcher’s Perspectives and Theoretical Stance ...................................... 4
1.2. Significance of Study .......................................................................................... 11
1.3. Research Questions and Objectives ................................................................... 17
1.4. Glossary of Terms .............................................................................................. 20
1.5. Organisation of Report....................................................................................... 24

PART I: THEORETICAL EXPLORATIONS OF SELECTED TOPICS IN

MECHANICS ...................................................................................... 26

CHAPTER 2: THEORETICAL EXPLORATIONS IN MOMENTUM/COLLISION:

ONE-DIMENSIONAL CHAIN-COLLISION .................................... 27
2.1. Introduction ........................................................................................................ 30
2.1.1. Theory ...................................................................................................... 33
2.1.2. Two-Body Collision .................................................................................. 36
2.1.3. Three-Body Chain-Collision ..................................................................... 40
2.1.4. Four-Body Chain-Collision....................................................................... 44
2.2. Generalisation: Many-Body Chain-Collision .................................................... 47

viii
2.3. Chapter Summary .............................................................................................. 49

CHAPTER 3: THEORETICAL EXPLORATIONS IN ROTATIONAL

MECHANICS: RIGID BODY COLLISION ...................................... 51
3.1. Introduction ........................................................................................................ 51
3.1.1. Theory ...................................................................................................... 54
3.2. Generalisations ................................................................................................... 59
3.2.1. First Generalisation: Unevenness (Non-uniformity) of Rod ....................... 59
3.2.2. Second Generalisation: Elasticity of Collision .......................................... 64
3.2.3. Third Generalisation: Existence of Pivot................................................... 67
3.3. Chapter Summary .............................................................................................. 72

PART II: EDUCATIONAL STUDIES ON MATHEMATICAL THINKING IN

PHYSICS.............................................................................................. 80

CHAPTER 4: LITERATURE REVIEW .................................................................... 81

4.1. What is Conceptual Understanding? ................................................................. 82
4.2. Representational Fluency ................................................................................... 86
4.3. Alternative Conceptions ..................................................................................... 88
4.4. Conceptual Change ............................................................................................ 89
4.5. Mathematical Thinking in Physics .................................................................... 93
4.6. Relevant Studies in Physics Education on Momentum/Collision and Rotational
Mechanics ......................................................................................................... 101
4.7. Relevant Studies in Physics Education Involving High-Performing Students 104
4.8. Formats of Diagnostic Instruments for Use in Assessing Students’
Understanding and Identifying Alternative Conceptions ............................... 106
4.9. Mathematical Thinking as a Teaching Intervention Strategy ........................ 115
4.10. Gaps in the Literature ...................................................................................... 118

CHAPTER 5: METHODOLOGY ............................................................................ 121

5.1. Evaluating Suitability of Sample for Current Study....................................... 122

ix
 5.1.1. Materials ................................................................................................ 123
5.1.2. Participants ............................................................................................ 124
5.2. Study on Momentum/Collision ........................................................................ 124
5.2.1. Development of Initial Version of Instrument .......................................... 125
5.2.2. Pilot Study .............................................................................................. 133
5.2.3. Instrument Validation: Chain-Collision Assessment ................................ 134
5.2.4. Main Study Participants.......................................................................... 143
5.3. Study on Rotational Mechanics ....................................................................... 143
5.3.1. Development of Initial Version of Instrument .......................................... 144
5.3.2. Pilot Study .............................................................................................. 152
5.3.3. Instrument Validation: Rigid Body Collision Assessment ........................ 152
5.3.4. Main Study Participants.......................................................................... 160
5.4. Interviews ......................................................................................................... 160
5.5. Teaching Interventions in Promoting Mathematical Thinking ...................... 162
5.5.1. Pedagogical Framework ......................................................................... 163
5.5.2. Examples in Conceptual Teaching and Problem-Solving Questions ........ 167
5.5.3. Main Studies ........................................................................................... 178
5.5.4. Instrument Validation: Survey Form ....................................................... 184
5.6. Data Analyses ................................................................................................... 188
5.6.1. Evaluating Suitability of Sample for Current Study ................................. 188
5.6.2. Excerpting Students’ Mathematical Thinking in Main Studies ................. 189
5.6.3. Evaluating Effectiveness of Teaching Intervention .................................. 193

CHAPTER 6: RESULTS ........................................................................................... 195

6.1. Evaluating Suitability of Sample for Current Study....................................... 195
6.1.1. Test Scores and Comparison between Batches for FCI, MBT and CSEM 195
6.1.2. Comparing between Subject Pool from NUS High School of Mathematics and
Science and Students from Other Schools ................................................ 199
6.1.3. Interview Findings .................................................................................. 201
6.2. Study on Momentum/Collision ........................................................................ 206

x
 6.2.1. Pilot Study .............................................................................................. 207
6.2.2. Main Study.............................................................................................. 209
6.2.2.1. Analysis of Test Statistics ............................................................... 209
6.2.2.2. Analysis of Test Scores ................................................................... 211
6.2.2.3. Analysis of Correlations ................................................................. 212
6.2.2.4. Analysis of Confidence Ratings ...................................................... 213
6.2.2.5. Analysis of Mathematical Expression Tier ...................................... 217
6.2.2.6. Analysis of Interviews..................................................................... 229
6.2.2.7. Analysis of Alternative Conceptions in Momentum/Collision .......... 236
6.3. Study on Rotational Mechanics ....................................................................... 244
6.3.1. Pilot Study .............................................................................................. 244
6.3.2. Main Study.............................................................................................. 247
6.3.2.1. Analysis of Test Statistics ............................................................... 247
6.3.2.2. Analysis of Test Scores ................................................................... 249
6.3.2.3. Analysis of Confidence Ratings ...................................................... 250
6.3.2.4. Analysis of Mathematical Expression Tier ...................................... 254
6.3.2.5. Analysis of Interviews..................................................................... 267
6.3.2.6. Analysis of Correlations ................................................................. 272
6.3.2.7. Analysis of Alternative Conceptions in Rotational Mechanics......... 273
6.4. Teaching Intervention in Promoting Mathematical Thinking........................ 280
6.4.1. Students’ Reflections on Lesson Intervention ........................................... 281
6.4.2. Survey Analysis ....................................................................................... 286
6.4.2.1. Study on Momentum/Collision ........................................................ 286
6.4.2.2. Study on Rotational Mechanics ...................................................... 291
6.4.3. Post-Test Analysis ................................................................................... 295
6.4.3.1. Study on Momentum/Collision ........................................................ 295
6.4.3.2. Study on Rotational Mechanics ...................................................... 308
6.5. Chapter Summary ............................................................................................ 320

CHAPTER 7: DISCUSSION ..................................................................................... 321

xi
7.1. Development of Five-Tier Instrument to Excerpt Students’ Mathematical
Thinking in Physics .......................................................................................... 323
7.2. Use of Mathematical Thinking in Physics ....................................................... 327
7.3. Use of Mathematical Thinking in Physics Conceptual Learning in Improving
Students’ Performance in Higher Order Thinking Questions ........................ 334

CHAPTER 8: CONCLUSION .................................................................................. 342

8.1. Study Limitations ............................................................................................. 343
8.2. Future Works ................................................................................................... 346
8.2.1. Theoretical Exploration on Moment of Inertia ........................................ 346
8.2.1.1. Introduction ................................................................................... 346
8.2.1.2. Theory ............................................................................................ 350
8.2.1.3. Scaling Method for Moment of Inertia Computation ....................... 356
8.2.1.4. Squashing Method for Moment of Inertia Computation .................. 376
8.2.2. Educational Studies on Mathematical Thinking in Physics ...................... 392
8.3. Study Implications............................................................................................ 394
8.4. Contributions of This Study to New Knowledge ............................................. 397

REFERENCES ............................................................................................................ 399

APPENDIX 1: VALIDATION OF CHAIN-COLLISION ASSESSMENT .............. 427

APPENDIX 2: VALIDATION OF RIGID BODY COLLISION ASSESSMENT ... 462
APPENDIX 3: CHAIN-COLLISION ASSESSMENT .............................................. 499
APPENDIX 4: RIGID BODY COLLISION ASSESSMENT.................................... 532
APPENDIX 5: INTERVIEW PROTOCOLS FOR STUDY ON
MOMENTUM/COLLISION ............................................................. 560
APPENDIX 6: INTERVIEW PROTOCOLS FOR STUDY ON ROTATIONAL
MECHANICS .................................................................................... 562
APPENDIX 7: SLIDES USED FOR TEACHING INTERVENTION ON
MOMENTUM/COLLISION ............................................................. 563

xii
APPENDIX 8: SLIDES USED FOR TEACHING INTERVENTION ON
ROTATIONAL MECHANICS ......................................................... 571
APPENDIX 9: VALIDATION OF SURVEY INSTRUMENT ................................. 579
APPENDIX 10:SURVEY INSTRUMENT ON TEACHING INTERVENTION ON
MOMENTUM/COLLISION ............................................................. 583
APPENDIX 11:SURVEY INSTRUMENT ON TEACHING INTERVENTION ON
ROTATIONAL MECHANICS ......................................................... 585
APPENDIX 12:STUDENTS’ RESPONSES IN SURVEY ON LESSON
INTERVENTION ON THE TOPIC OF MOMENTUM/COLLISION
............................................................................................................ 587
APPENDIX 13:STUDENTS’ RESPONSES IN SURVEY ON LESSON
INTERVENTION ON THE TOPIC OF ROTATIONAL
MECHANICS .................................................................................... 588
APPENDIX 14:STUDENTS’ REFLECTIONS ON LESSON INTERVENTION .... 589

xiii
 LIST OF FIGURES

Figure 1.1 People’s Conceptions of the Relationships among Terms for Mental Activities
...................................................................................................................... 3
Figure 1.2 NUS High School Physics and Engineering Department Curriculum
Framework .................................................................................................... 5
Figure 1.3 Revised Bloom’s Taxonomy .......................................................................... 8
Figure 2.1 Newton’s Cradle .......................................................................................... 27
Figure 2.2 Chain-Collision Involving 𝑛 Masses in between 𝑚1 and 𝑚2 ........................ 31
Figure 2.3 One-Dimensional Two-Body Motion before Collision ................................. 34
Figure 2.4 One-Dimensional Two-Body Motion after Collision .................................... 35
(1+𝑒)2 𝑥
Figure 2.5 Three-Dimensional Plot of 𝜂 = (1+𝑥)2
........................................................ 39

Figure 2.6 Three-Body Chain Collision ........................................................................ 41

Figure 2.7 Four-Body Chain-Collision .......................................................................... 45
Figure 2.8 Many-Body Chain-Collision ........................................................................ 47
Figure 3.1 Collision between Projectile and Stationary Rigid Rod ................................ 51
Figure 3.2 Collision between Projectile and Moving Rigid Rod .................................... 57
Figure 3.3 Generalisation to Projectile-Rod Collision: Unevenness (Non-Uniformity) of
the Rod ........................................................................................................ 60
Figure 3.4 Generalisation to Projectile-Rod Collision: Elasticity of Collision ............... 64
Figure 3.5 Generalisation to Projectile-Rod Collision: Existence of Pivot ..................... 68
ωℓ a (𝑚+𝑀)𝐼CM (𝑚+𝑀)𝐼CM
Figure 3.6 Graph of 2v vs ℓ for: (a) √ = 0.8ℓ, x = 0.1ℓ ; (b) √ =
0 𝑀𝑚 𝑀𝑚

1.3ℓ, 𝑥 = 0.7ℓ ............................................................................................. 74

𝜔ℓ 𝑎 𝑀 𝑀
Figure 3.7 Graph of 3(1+𝜀)𝑣 vs ℓ for: (a) 𝑚 = .47 ; (b) 𝑚 = 5.75 .................................. 76
0

𝜔ℓ 𝑎
Figure 3.8 Graph of 2𝑣 vs ℓ for: (a) 𝑚 = 𝑀, 𝑥 = .2ℓ ; (b) 𝑚 = .5𝑀, 𝑥 = .5ℓ ; (c) 𝑚 =
0

.1𝑀, 𝑥 = .5ℓ ................................................................................................ 78

xiv
Figure 4.1 Illustration of Multiple Representations Used in Physics to Depict Real World
Situation (Redish, 2003) .............................................................................. 87
Figure 4.2 Rigid Body Collision ................................................................................... 99
Figure 4.3 Example of 2TMC Question (Chang et al., 2007)....................................... 109
Figure 4.4 Example of 3TMC Question (Caleon & Subramaniam, 2010a) .................. 113
Figure 4.5 Example of 4TMC Question (Caleon & Subramaniam, 2010b) .................. 114
Figure 5.1 Sample Question in the Chain-Collision Assessment ................................. 142
Figure 5.2 Elastic Collision between Point Mass and Uniform Rod............................. 146
Figure 5.3 Elastic Collision between Point Mass and Uniform Rod............................. 149
Figure 5.4 Sample Question in the Rigid Body Collision Assessment ......................... 159
Figure 5.5 Framework of the Newly-Explored Teaching Intervention Strategy ........... 163
Figure 5.6 Two Particles Moving with Constant Velocities ......................................... 169
Figure 5.7 The Motion of Particle 1 as Viewed from Particle 2 ................................... 170
Figure 5.8 Sample Slides Used in Teaching Intervention on Momentum/Collision ..... 180
Figure 5.9 Sample Slides Used in Teaching Intervention on Rotational Mechanics ..... 183
Figure 6.1 Distributions of Participants’ FCI, MBT and CSEM Test Scores ............... 197
Figure 6.2 FCI, MBT and CSEM Score Comparisons across Schools ......................... 200
Figure 6.3 Scores Distribution across Questions – FCI, MBT, CSEM ......................... 202
Figure 6.4 Distribution of Students’ Performance in the Pilot Study on Rotational
Mechanics ................................................................................................. 246
Figure 8.1 Point Mass Undergoing Circular Motion about Fixed Axis ........................ 351
Figure 8.2 Rigid Body Rotating about 𝑧-Axis ............................................................. 353
Figure 8.3 Centre of Mass Position of Rigid Body ...................................................... 354
Figure 8.4 Two-Dimensional Rigid Body on 𝑥𝑦-Plane ............................................... 355
Figure 8.5 (a) One-Dimensional and (b) Two-Dimensional Uniform Objects Scaled
Proportionally by Linear Factor 𝑛 .............................................................. 358
Figure 8.6 Uniform Rod about Axis Passing Perpendicularly through Centre of Mass 361
Figure 8.7 Shorter Rod about Axis Passing Perpendicularly through Centre of Mass .. 361
Figure 8.8 Longer Rod about Axis Passing Perpendicularly through Centre of Mass .. 362

xv
Figure 8.9 Solid Rectangle about Axis Passing Perpendicularly through Centre of Mass
.................................................................................................................. 363
Figure 8.10 Smaller Rectangle about Axis Passing Perpendicularly through Centre of
Mass .......................................................................................................... 363
Figure 8.11 Bigger Rectangle about Axis Passing Perpendicularly through Centre of Mass
.................................................................................................................. 365
Figure 8.12 Solid Square about Axis Parallel to Plane Passing through Centre of Mass and
Tilted from Symmetry Axis ....................................................................... 366
Figure 8.13 Smaller Square about Axis Parallel to Plane Passing through Centre of Mass
and Tilted from Symmetry Axis ................................................................. 367
Figure 8.14 Solid Square on 𝑥𝑦-Plane .......................................................................... 368
Figure 8.15 Solid Right-Angled Triangle about Axis Passing Perpendicularly through
Centre of Mass........................................................................................... 369
Figure 8.16 Solid Rectangle Constructed from Two Right-Angled Triangles ................ 370
Figure 8.17 Equilateral Triangle about Axis Passing Perpendicularly through Centre of
Mass .......................................................................................................... 371
Figure 8.18 Smaller Equilateral Triangle about Axis Passing Perpendicularly through
Centre of Mass........................................................................................... 371
Figure 8.19 Slicing Hollow Sphere into Horizontal Infinitesimal Rings ........................ 373
Figure 8.20 Slicing Solid Sphere into Horizontal Infinitesimal Disks ............................ 374
Figure 8.21 Squashing Operation along 𝑧-Direction...................................................... 377
Figure 8.22 Squashing Cylinder along 𝑧-Direction ....................................................... 385
Figure 8.23 Squashing Plane along 𝑦-Direction ............................................................ 386
Figure 8.24 Squashing Uniform Triangle along 𝑧-Axis ................................................. 387
Figure 8.25 Squashing Hollow Cone along 𝑧-Direction ................................................ 389

xvi
 LIST OF TABLES

Table 1.1 Topic, Concept Inventory, and Reference .................................................... 14

Table 1.2 Glossary of Terms ....................................................................................... 21
Table 4.1 Studies on Students’ Alternative Conceptions and Conceptual Change on the
Topic of Momentum/Collision ................................................................... 101
Table 4.2 Studies on Students’ Alternative Conceptions and Conceptual Change on the
Topic of Rotational Mechanics .................................................................. 104
Table 4.3 Level of Internal Consistency based on Cronbach’s Alpha ......................... 107
Table 4.4 2TMC Test Instruments in Physics ............................................................ 109
Table 4.5 3TMC Test Instruments in Physics ............................................................ 111
Table 4.6 4TMC Test Instruments in Physics ............................................................ 112
Table 4.7 Conceptual Change Strategies in Physics ................................................... 116
Table 5.1 Checklist for Validation of Chain-Collision Assessment by Physics Teachers
(with Responses) ....................................................................................... 135
Table 5.2 Comments by Validators and Actions Taken for Refinement of Chain-
Collision Assessment ................................................................................. 136
Table 5.3 Checklist for Validation of Rigid Body Collision Assessment by Physics
Teachers (with Responses)......................................................................... 154
Table 5.4 Comments by Validators and Actions Taken for Refinement of Rigid Body
Collision Assessment ................................................................................. 155
Table 5.5 Examples of Qualitative vs. Mathematical Guiding Inquiry Questions in
Physics Conceptual Teaching..................................................................... 167
Table 5.6 Comments by Validators and Actions Taken for Refinement of Rigid Body
Collision Assessment ................................................................................. 186
Table 5.7 Marking scheme for Main Study Assessments ........................................... 189
Table 5.8 Summary of Formulas Used for Data Analyses .......................................... 191
Table 6.1 Comparison of Test Scores across Batches of Students .............................. 198
Table 6.2 Reliability Statistics for Chain-Collision Assessment ................................. 210

xvii
Table 6.3 Cognitive Scores for Chain-Collision Assessment...................................... 211
Table 6.4 Analysis of Cognitive and Confidence Measures per Item (Chain-Collision
Assessment)............................................................................................... 216
Table 6.5 Students’ Performance in Mathematical Expression Tier across Questions
(Chain-Collision Assessment) .................................................................... 217
Table 6.6 Possible Students’ Responses, Score Assignments and Teacher’s Commentary
for Mathematical Expression Tier of Each Question in Chain-Collision
Assessment ................................................................................................ 219
Table 6.7 Alternative Conceptions on Momentum/Collision ...................................... 237
Table 6.8 Reliability Statistics for Rigid Body Collision Assessment ........................ 249
Table 6.9 Cognitive scores for Rigid Body Collision Assessment .............................. 250
Table 6.10 Analysis of Cognitive and Confidence Measures per Item (Rigid Body
Collision Assessment) ............................................................................... 253
Table 6.11 Students’ Performance in Mathematical Expression Tier across Questions
(Rigid Body Collision Assessment) ........................................................... 254
Table 6.12 Failure in Representing Mathematical Thinking in terms of Conceptual
Reasoning – Answer is Correct but “Other Reason” was Selected (Rigid Body
Collision Assessment) ............................................................................... 256
Table 6.13 Possible Students’ Responses, Score Assignments and Teacher’s Commentary
for Mathematical Expression Tier of Each Question in Rigid Body Collision
Assessment ................................................................................................ 257
Table 6.14 Alternative Conceptions on Rotational Mechanics ..................................... 273
Table 6.15 Excerpts from Students’ Reflections on Teaching Intervention (𝑁 = 11)... 283
Table 6.16 Responses to Survey Statements on the Topic of Momentum/Collision (𝑁 =
37) ............................................................................................................ 287
Table 6.17 Responses to Survey Statements on the Topic of Rotational Mechanics (𝑁 =
43) ............................................................................................................ 292
Table 6.18 Cognitive Scores for Chain-Collision Assessment (Post-Test) ................... 298
Table 6.19 Analysis of Cognitive and Confidence Measures per Item (Chain-Collision
Assessment, Control Group Post-Test) ....................................................... 301

xviii
Table 6.20 Analysis of Cognitive and Confidence Measures per Item (Chain-Collision
Assessment, Experimental Group Post-Test) .............................................. 302
Table 6.21 Students’ Performance in Mathematical Expression Tier across Questions
(Chain-Collision Assessment, Control Group vs. Experimental Group Post-
Test) .......................................................................................................... 304
Table 6.22 Alternative Conceptions on Momentum/Collision (Chain-Collision
Assessment Pre-Test vs. Post-Test) ............................................................ 307
Table 6.23 Cognitive Scores for Rigid Body Collision Assessment (Post-Test) ........... 310
Table 6.24 Analysis of Cognitive and Confidence Measures per Item (Rigid Body
Collision Assessment, Control Group Post-Test)........................................ 313
Table 6.25 Analysis of Cognitive and Confidence Measures per Item (Rigid Body
Collision Assessment, Experimental Group Post-Test) ............................... 314
Table 6.26 Students’ Performance in Mathematical Expression Tier across Questions
(Rigid Body Collision Assessment, Control Group vs. Experimental Group
Post-Test) .................................................................................................. 316
Table 6.27 Alternative Conceptions on Rotational Mechanics (Rigid Body Collision
Assessment Pre-Test vs. Post-Test) ............................................................ 319

xix
 SUMMARY

Mathematical thinking is a crucial aspect in the study of physics. In this thesis, we

explored the extent of use of mathematical thinking in a few high school physics topics. The

unique feature of this thesis is the blending of theoretical physics and physics education

research. This thesis consists of two parts: theoretical explorations of selected topics in

mechanics as well as educational studies on mathematical thinking in these topics. The

explorations on the topics of momentum/collision and rotational motion were crucial in the

construction of higher order thinking questions. These questions were implemented in a new

5-tier format, which seeks to excerpt students’ mathematical thinking in association with

their conceptual understanding and certainty of response. The test instruments were also used

in a pre/post-test design to measure the effectiveness of teaching interventions in the selected

topics, where mathematical thinking was strongly emphasised and promoted. Due to the

higher order thinking nature of the questions, we focused on high-performers in this

educational study.

The findings in the educational aspect of this study showed that the students’

articulation of mathematical thinking was limited, though they performed well in the answer

and reason tiers. This limitation was significantly overcome in the intervention lessons,

where students were encouraged to develop the habit of performing mathematical thinking

when approaching physics problems. The strong positive correlation between the extent of

students’ mathematical thinking and their conceptual understanding was also shown in the

interventions, suggesting the importance of early exposure to mathematical thinking in

physics.

xx
 CHAPTER 1

INTRODUCTION

1.1. Research Background

1.1.1. Mathematical Thinking in Physics

Scientific theories, laws and principles are often expressed in mathematical language.

Even observation and data mining in experimental research often require mathematics to

model the behavior of a system. Without mathematics, it would be impossible to formulate

using computational algorithms, with the observations providing initial data for the model

and serving as a check on the accuracy of the model. Physics, more specifically, is a subject

which comprises both qualitative and mathematical concepts; they both rely strongly on each

other. The mathematics justifies the physical phenomena while the qualitative explanations

often gives a picture of what truly occurs in nature.

Mathematics and Physics have a long and close relationship that is of crucial and

growing importance for both. Many well-known physicists such as Isaac Newton,

Archimedes, Richard Feynman, James Clerk Maxwell and Albert Einstein are also

mathematicians themselves. Mathematics is an intrinsic component of science and is part of

its fabric. It is also a universal language and an indispensable source of intellectual tools.

Reciprocally, science inspires and stimulates mathematics – for example, posing new

questions, engendering new ways of thinking, and ultimately conditioning the value system

of mathematics. Any progress in physics and in other sciences, in general, requires close

involvement and strengthening of the mathematical enterprise. Very often, new physics and

new mathematics go hand in hand.

1
 British mathematician and philosopher, Alfred North Whitehead, wrote in one of his

books,

“Mathematics is the most powerful technique for the understanding of pattern, and

for the analysis of the relationships of patterns… Having regard to the immensity of

its subject-matter mathematics, even modern mathematics, is a science in its

babyhood. If civilization continues to advance, in the next two thousand years the

overwhelming novelty of human thought will be the dominance of mathematical

understanding.” (Whitehead, 1942)

His philosophy signifies the importance of “mathematical thinking” or “to think

mathematically” in understanding nature, which is governed by scientific laws and

principles. This element of “mathematical thinking” is one aspect that is often missing or

overlooked in high school physics. Mathematics, unfortunately, is often avoided in the

conceptual teaching of high school physics, as it is perceived to be a killing factor in respect

of students’ interest (Politis et al., 2007; Ma’Moon, 2005).

Adopting the definition by Holyoak and Morrison (2005), “thinking” is the systematic

transformation of mental representations of knowledge to characterise actual or possible

states of the world, often in service of goals. It is a mental activity (mental assessment or

judgment) that eventually leads to one taking or making an action. The ability to think is a

cognitive process. Rips and Conrad (1989) elicited judgments from college students about

how various mentalistic terms relate to each other. Figure 1.1 is a summary of these

relationships. The first school of thought (Figure 1.1(a)) describes that planning is a kind of

deciding, reasoning, conceptualising, and thinking. The second school of thought (Figure

1.1(b)) treats thinking as part of conceptualising, which includes remembering, reasoning,

2
and so on. In these two schools of thought, most strikingly, “thinking” is the most general

term in both orderings – the grand superordinate of mental activities, which permeates all the

others. In contrast, Holyoak and Morrison (2005) describes “reasoning” as a process of

drawing inferences (conclusions) from some initial information (premises). In other words,

in performing reasoning, one scrutinises his/her thoughts and it involves a deeper process.

The habit of thinking deeply needs to be developed before one can perform reasoning in any

specific context. In view of these definitions, the word “mathematical thinking” would be

preferred in this thesis rather than “mathematical reasoning”.

Figure 1.1

People’s Conceptions of the Relationships among Terms for Mental Activities

This thesis study explores the use of mathematical thinking in the subject of Physics.

Unfortunately, the definitions of mathematical thinking vary across different fields and

3
educational levels (Lutfiyya, 1998). Isoda (2006) stated that mathematical thinking requires

higher order thinking skills with a blend of mathematics to understand the ideas, to identify

and prove the relationships between the differences in ideas and connect these differences

together to solve an existing problem. Sam and Yong (2006) defined mathematical thinking

as a mental operation used in solving problems affirmed by mathematical knowledge and

dispositions. The most recent literature by Uyangör (2019) defined mathematical thinking as

a cognitive activity using mathematical techniques, concepts, and methods, directly or

indirectly, in the problem-solving process. Mathematical thinking allows one to explore a

physical situation and perform a deep exploration, resulting in various situations that require

multiple layers of analyses.

1.1.2. Researcher’s Perspectives and Theoretical Stance

The joy of teaching science is expressed when teaching becomes more than just

transferring knowledge to students but also stretching their capacity to develop good

understanding and build on existing concepts in order to expand and create new knowledge.

In this way, students are not simply seen as consumers of knowledge but also, together with

their teachers, the co-producers in developing ideas.

The research presented in this thesis was motivated by the attempt to make teaching

of physics more effective as well as my professional background as a teacher and Consultant

in NUS High School of Mathematics and Science1, Singapore. I started teaching in this

school in 2008 and was promoted to Consultant in the Physics and Engineering Department

1
Refer to www.nushigh.edu.sg for the details of the school profile.

4
in 2013. My role as a Consultant has allowed me to be deeply involved in the department’s

curriculum planning and design, especially the curriculum for higher grades. This allows me

to understand in depth the department’s curriculum framework used for teaching and

learning.

Figure 1.2

NUS High School Physics and Engineering Department Curriculum Framework

The approach taken by the Physics and Engineering Department in NUS High School

focuses on three aspects: content, pedagogy and assessment, as summarised in the curriculum

framework (Figure 1.2). As a school specialising in mathematics and science, in terms of

content, the school has designed very rigorous curricula pegged for students with passion for

5
the subjects. On top of that, practical skills, attitudes and values towards science are also

emphasised. In terms of assessment, the department focuses on balancing between formative

and summative assessments. The main distinction between formative and summative

assessment relates to purpose and effect, not just pertaining to timing (Sadler, 1989).

Formative assessment is concerned with how judgments about the quality of student

responses can be used as feedbacks to improve the student’s competence by minimising

randomness and inefficiency of trial-and-error approaches. Thus, the formative assessment

for all physics modules in NUS High School is usually not counted towards the students’

final grades. On the other hand, summative assessment is concerned with summing up or

summarising the achievement status of each student, and is geared towards reporting at the

end of a course of study. Across all the physics modules in this school, the summative

assessment that is counted towards the students’ final grades usually consists of three

different types of questions: conceptual questions, application questions and higher order

thinking questions. Higher order thinking questions are questions that require skills involving

analysis, evaluation, and synthesis (creation of new knowledge), which are the highest three

levels of skills according to the revised Bloom’s Taxonomy (Figure 1.3). In terms of

pedagogy, as a relatively young school, we have implemented a lot of different approaches

for teaching and learning. Pedagogies ranging from as traditional as didactic teaching to

inquiry approaches, flipped classroom and modelling, even including the ones involving

Information Communication Technology (ICT) tools, whiteboarding and lightboarding, have

been explored over the years, and many of these are indeed effective in teaching basic physics

concepts. However, observations show that only a few students are able to translate the

concepts learnt in solving higher order thinking questions in the assessment. Many students

6
tend to use either incorrect reasoning or false intuition when it comes to solving higher order

thinking questions, rather than to correctly apply the concepts they have previously learnt.

As such, there is a need to explore a deeper variable related to students’ thinking pattern that

will have impact on content (what needs to be emphasised), pedagogy (how we teach it) as

well as assessment (what we want to assess).

Although there is a concern that students tend to use false scientific intuition when it

comes to solving higher order thinking questions, this does not nullify the importance of

having a strong scientific intuition in the study of science. Hersh (1997) provides a list of

definitions of the word “intuition” and describes his conviction on the importance of the use

of intuition, especially in the study of Mathematics. Unfortunately in all these usages,

intuition is vaguely defined and the definition varies from one usage to another. Burton

(1999), in his research, however, noted that there was agreement that intuition alone is not

always enough, as it only reinforces the feel that you know how to approach. One respondent

in his interview offered a caution: “intuitions can come close to prejudice, which can be a

very blinding thing.” Although the use of intuition, be it qualitative or quantitative, may not

necessarily be wrong, it may not encapsulate the whole knowledge students have learnt, and

usually comes from their common sense beliefs, daily experiences or what they have heard

before, regardless of whether it is scientifically backed up. It is even questionable whether or

not the process of thinking indeed takes place as students provide an intuitive answer to a

question. This produces a potential problem and concurs with the observation I have

encountered as a physics teacher. One main observation is the declining performance among

physics students whenever the test questions involve higher order thinking. Especially for

the upper year students, where physics concepts and assessment questions become less and

7
less intuitive, many have dropped Physics and lost interest as they start seeing the concepts

as contents that they have to memorise, thus concurring with Kadir et al. (2009). On top of

that, being exam-oriented, students have a tendency to think that memorising key words,

formulas, solving past-year examination questions and scoring well in exams can make them

grasp good conceptual understanding in the subject. As a snowball effect over the years,

many find Physics is not an interesting subject to pursue for higher degree. Unfortunately,

the underlying problem that these students face is not their inability to understand the

concepts, but their inability to perform the thinking deeply, i.e. translating the concepts into

articulation of thoughts.

Figure 1.3

Revised Bloom’s Taxonomy

CREATE
produce new or original work

EVALUATE
justify a stand or decision

ANALYSE
draw connections among ideas

APPLY
use information in new situations

UNDERSTAND
explain ideas or concepts

REMEMBER
recall facts and basic concepts

8
 Following Kant’s philosophy, even the formation of a scientific concept is a task that

requires mathematics (Kant, 2004). Supporting the positivism paradigm of science,

mathematics provides a special avenue for helping to ensure the ‘truth’ that science contains

by providing a model of a codified scientific discipline, the possibility of which is clear and,

moreover, guaranteed by its own achievement of cognition that is both synthetic and a priori,

i.e. power of reasoning based on self-evident truths. In other words, this provides an

explanation of how synthetic a priori judgments are affirmed in mathematical contexts,

together with the resulting and related explanation of how a systematic body of demonstrable

knowledge comprises such judgments, allowing mathematical truths to be invoked as a

paradigm of the substantive, yet necessary, universal truths that metaphysics hopes to

achieve.

My participation as a former student in the Asian and International Physics

Olympiads as well as my involvement as a leader and observer in the competitions over many

years have allowed me to observe whether students’ conceptual understanding in physics

correlates with their mathematical thinking, thus concurring with the study by Meltzer

(2002). These general observations have encouraged me to explore this link and see whether

students’ understanding in physics can be uplifted further by focusing on mathematical

thinking. Prior to this thesis study, I had explored the use of mathematical thinking in physics

for a small group of students undergoing Physics Olympiad training in NUS High School,

blending conceptual learning and mathematical thinking – similar to a study done by Kuo et

al. (2013). Even for basic physics concepts such as Newton’s law, the use of mathematical

thinking had resulted in significant improvement in their conceptual understanding as well

9
as test performance. As the robustness and quality of this approach were not known due to

the very small number of students and the lack of any analysis into it, this research study was

done with the purpose of confirming the benefit of this teaching strategy in a wider pool of

high-performing students.

With the above descriptions of the research background, it was mandatory that this

research study be practitioner-based, that is, the researcher was the same person doing the

teaching intervention. Also, the nature of the research was such that only the candidate was

best equipped to conduct a study of this nature in the specified school. An overlying

interpretive theoretical perspective (Crotty, 1998) has enabled us to explore and articulate

many of the assumptions underlying the languages, inter-relationships and communities

within which the teaching intervention was conducted. Prior to this study, there have actually

been a number of reports on the involvement of school science teachers in practitioner

research projects (for example, Bennett (2002) and Tobin (1999)), making the practitioner-

based research a defensible theoretical stance, especially in a study of this nature.

The main limitation of practitioner-based research, as adopted in this thesis study, is

its potential for bias in the findings, such as students being influenced in their responses to

the test questions, teacher teaching to the test during the intervention, as well as the approach

being not relevant for students from other schools and for other teachers doing the teaching

intervention. However, some measures were consciously taken to minimise this bias. As the

researcher in this thesis study as well as the teacher doing the teaching intervention, I

consciously followed all the ethical protocols in the conduct of this study. The slides for the

teaching intervention, including the concepts taught and practice questions used were also

attached in the appendices, allowing other researchers to replicate these for similar studies.

10
Furthermore, the nature of the study is indeed student-centric; there was very little way to

influence students on how to respond to the questions in this study owing to the difficulty

level and leveraging on aspects of theoretical physics, which is not an easy ball game for high

school students. On top of these, the evolutionary nature of the initiative encouraged the

adoption of the iterative strategies of planned action that Carr & Kemmis (1986) put forward

as key components of educational action research; that is, implementation, observation,

reflection and change. An action research-like methodology offered a mechanism for

implementing change and improving student learning (through test construction and teaching

intervention), analysing in a systematic way the impact of the innovation (through survey

analysis and students’ reflection), and then allowing us to adapt it to the needs of our school

(Krockover et al., 2001). This largely forms the design of the educational study in this thesis.

Although most articles in the literature only relate mathematical thinking to the aspect

of problem-solving, this thesis study explores three aspects of mathematical thinking in

physics: creation of higher order thinking problems, problem-solving and conceptual

learning. It was hoped that the excerpts of students’ mathematical thinking from the way they

approach physical situations would provide valuable insights on an effective teaching

intervention for the students. Other aspects such as explorations of mathematical methods

and techniques in physics are left for future studies.

1.2. Significance of Study

Some good theoretical explorations requiring mathematical thinking have been done

in a number of physics topics previously. For instance, explorations on Bertrand’s theorem

(Ricardo & Ho, 2019; Santos et al., 2009; Salas-Brito et al., 1997; Ballesteros et al., 2008;

11
Tikochinsky, 1988), Carnot engine (Bender et al., 2000; Martínez et al., 2016; Esposito et

al., 2010), Fluids Dynamics (McDonough, 2009; Crank, 1979, Howison, 2005), and some

more. However, commonly these explorations involve deeper concepts with more advanced

mathematics. Being the most basic topic in physics, mechanics has received maximum

attention from educational researchers, as its phenomena can be found everywhere around

us, and its analysis is relatively easier as compared to other topics. However, we could not at

all find in the literature the kind of explorations of high school level mechanics using

mathematical thinking. If mathematical thinking in physics is to be promoted from early

years, theoretical explorations using mathematical thinking have to be done for the most basic

physics topics, i.e. mechanics, using basic mathematics (e.g. algebra, graphs, simple

calculus). Currently in most high school physics classes, mathematics is only introduced as

a tool for problem solving. In this manner, students are only required to know, memorise and

understand when to use the formulas required for a specific physics question. At the same

time, they also get to know the algorithmic approaches that are needed for the solving of

standard physics problems involving calculations. What is missing here is the thinking. In

our theoretical explorations in this thesis study, mathematics is used not simply as a means

to solve problems, but also as a thinking tool for deep theoretical explorations in order to

create higher order thinking problems. For this reason, this thesis study focuses on two topics

in mechanics, i.e. momentum/collision and rotational mechanics. The latter requires more

abstract concepts, such as those targeted for higher grade students. In the last chapter, the

topic of moment of inertia will also be discussed as a display of the use of mathematical

thinking in another aspect, i.e. development of new mathematical methods for computation

of physical quantities. This provides an insight for future studies.

12
 Educational studies on mathematical thinking have been done previously in the

subject of Mathematics (Dubinsky, 2002; Dreyfus, 2002; Jacobs et al., 2010; Tall, 2002) as

well as in relation with creativity tests (Ervynck, 2002; Haylock, 1997; Leikin, 2007),

computer science (Wing, 2008; Bar & Stephenson, 2011; Henderson et al., 2001; Gries et al.,

2001), psychology (Resnick & Ford, 2012; Hannula, 2006), neuroscience (Dehaene et al.,

1999; Sohn et al., 2004; Goswami, 2004) and many others. However, very limited literature

was found on educational studies involving mathematical thinking in physics. One of them

is a study by Long and Jiar (2014). This study showed that each of the mathematical thinking

skills (intellectual skills, verbal information, cognitive strategy, and attitudes) had significant

relationship with physics achievement of the students, thus signifying the importance of

mathematical thinking as the bridge that connects mathematics and physics. Tuminaro (2004)

provided a theoretical framework that offers educators and researchers a vocabulary

(ontological classification of cognitive structures) and grammar (relationship between the

cognitive structures) for understanding the nature and origin of students’ use of mathematics

in the context of physics. However, none of the studies have focused on excerpting students’

mathematical thinking from a physics test as well as not providing a framework for a teaching

intervention involving the use of mathematical thinking in physics.

Another significance of this study is to encourage high school teachers to promote

mathematical thinking in teaching physics concepts, even incorporating it in the guiding

inquiry questions, as students’ performance in higher order thinking assessment will be

limited otherwise. Although equations and formulas would have already been introduced in

high school physics syllabus, the activation of mathematical thinking in conceptual learning

and problem-solving is usually lacking. The study by Liu and Liu (2011) revealed that many

13
teachers did not realise the organic interaction between physics and mathematics. This

incomplete understanding had driven them to not promote mathematical thinking in physics

to their future students, thus hindering their interest in learning physics holistically. We hope

that teachers, through the findings presented in this study, would understand the benefits of

a mathematical thinking approach, especially in helping high-performing students stretch

their cognitive ability further in the study of physics.

Many readily-available instruments to test students’ basic conceptual understanding

in high school physics are available in the literature (Table 1.1). However, as this thesis study

requires assessments with higher order thinking problems, all the listed assessments failed to

meet this requirement. To the best of our knowledge, no available assessments involving

higher order thinking questions that invoke mathematical thinking are available in the

literature.

Table 1.1

Topic, Concept Inventory, and Reference

References Name of Assessment Topic

Hestenes et al. (1992) Force Concept Inventory (FCI) Motion

The Force and Motion

Thornton & Sokoloff (1998) Newton’s Laws of Motion
Conceptual Evaluation (FMCE)

Rotational and Rolling Motion

Rimoldini & Singh (2005) Rotational Mechanics
Conceptual Survey (RRMCS)

Test of Understanding Graphs in

Beichner (1994) Kinematics Graph
Kinematics (TUG-K)

The Energy Concept Assessment

Ding (2007) Energy
(ECA)

14
 The Energy and Momentum
Singh & Rosengrant (2003) Energy and Momentum
Conceptual Survey (EMCS)

Thermodynamics Concept
Midkiff et al. (2001) Thermodynamics
Inventory (TCI)

Introductory Thermal Concept

Yeo & Zadnik (2001) Thermodynamics
Evaluation (ITCE)

The Heat and Energy Concept Temperature, Heat Transfer,

Prince et al. (2012)
Inventory (HECI) Thermal Radiation

Heat Transfer Concept Inventory

Jacobi et al. (2004) Heat Transfer
(HTCI)

Heat Transfer, Fluid

Thermal and Transport Concept
Streveler et al. (2011) Mechanics,
Inventory (TTCI)
Thermodynamics

Roedel et al. (1998) Visualisation, Mathematical

Waves Concept Inventory (WCI) Depiction and Definitions of
Thoads & Roedel (1999) Waves

Electromagnetics Concept
Notaros (2002) Fields and Waves
Inventory (EMCI)

Quantum Mechanics Survey

Zhu & Singh (2012) Quantum Mechanics
(QMS)

Quantum Mechanics Conceptual

McKagan et al. (2010) Quantum Mechanics
Survey (QMCS)

Quantum Mechanics Assessment

Goldhaber et al. (2009) Quantum Mechanics
Tool (QMAT)

Newtonian Gravity Concept

Williamson et al. (2013) Newtonian Gravity
Inventory (NGCI)

In investigating students’ mathematical thinking in physics, the most appropriate

technique would be free-response pencil-and-paper tests. This method allows the flexibility

of analysing in-depth explorations of the manifestation of students’ mathematical thinking in

relation to the physics concepts used. However, the data analysis for free-response tests is

15
time-consuming and may be biased if thorough marking schemes are not provided. The use

of multiple-choice tests is also viable to diagnose students’ conceptual understanding,

considering that this type of test is simple, efficient, valid, reliable and ready to use.

Moreover, multi-tier tests have been used by educational researchers as instruments to test

students’ conceptual understanding – the different tiers provide more information for in-

depth analyses. For example, 2-tier multiple-choice (2TMC) tests (Tsui & Treagust, 2010;

Treagust, 1988; Chang et al., 2007; Haslam & Treagust, 1987; Tan et al., 2002; Tsai & Chou,

2002; Voska & Heikkinen, 2000) that evaluate students’ descriptive knowledge (answer tier)

and explanatory knowledge (reason tier) have been developed. However, some other

researchers feel that another tier, which evaluates students’ confidence rating for their

responses in the answer and reason tiers, has to be included, thus creating a 3-tier multiple-

choice (3TMC) test. This tier is useful to provide more information on whether students’

correct responses are due to adequate understanding or simply guessing, and whether

students’ mistakes are due to lack of knowledge or existence of alternative conceptions

(Hasan, Bagayoko & Kelley, 1999). When the confidence ratings for the answer and reason

tiers are separated, this creates a 4-tier multiple choice (4TMC) test, and this format has been

adopted by some researchers (Caleon & Subramaniam, 2010; Yang & Lin, 2015;

Sreenivasulu & Subramaniam, 2013; Kaltakci-Gurel et al., 2016, 2017) to uncover students’

alternative conceptions in different topics in science.

How to excerpt students’ mathematical thinking and determine its association with

their conceptual understanding then? If we were to use multi-tier diagnostic formats, it is

very difficult to provide a few expressions of mathematical thinking as the options for the

answer and reason tiers. Forcing it may cause students to just choose correct formula based

16
on their memory or expressions they are familiar with, without fully understanding the

correct reasoning. If the answer and reason tiers cannot assess students’ mathematical

thinking, then even the 4TMC test would still not be sufficient to evaluate the association

between students’ conceptual understanding and mathematical thinking. As such, there is a

need for an additional tier to be constructed for such purpose and this will serve to excerpt

students’ mathematical thinking in relation to their conceptual understanding. The newly-

constructed 5-tier assessment is hoped to be sufficient, reliable and useful for this purpose.

To sum up, this thesis study requires both theoretical explorations in a few physics

topics for the purpose of constructing higher order thinking questions as well as educational

studies to excerpt students’ mathematical thinking. In this context, instrument development

and teaching interventions for high-performing students are also needed. To the best of our

knowledge, this is the first thesis in the university to straddle both content and pedagogy in

physics – that is, theoretical physics and physics education research. The research findings

are expected to contribute to both areas.

1.3. Research Questions and Objectives

The main research questions for the educational aspects of my PhD thesis are:

(1) How can a five-tier diagnostic instrument that builds on previous formats of

two-tier, three-tier and four-tier instruments be developed for the specific

purpose of this study, and what are the issues in its development?

As we would like to see the use of mathematical thinking in high school physics, two

topics were chosen (i.e. momentum/collision and rotational mechanics). The use of

17
mathematical thinking will come strongly in the development of higher order thinking

problems, as mere intuition is foreseen to fail in the analysis of the problems. Development

of suitable assessments for this study should assess the degree of students’ conceptual

understanding in the topics assessed and its association with their ability to perform

mathematical thinking. This aspect of students’ mathematical thinking is the shortcoming of

existing 2-tier, 3-tier and 4-tier test instruments, as neither of the tiers in these instruments

could excerpt students’ mathematical thinking nor be easily expressed using mathematical

formalism. This prompted the development of a 5-tier test format for this thesis study. This

study will still be focusing on two topics of mechanics (i.e. momentum/collision and

rotational mechanics) as extensions to the theoretical explorations on these topics.

(2) To what extent do high-performing students demonstrate mathematical

thinking when solving higher order thinking questions in selected topics in

physics?

The second research question demands a thorough analysis of students’ mathematical

responses in the constructed assessments in relation to their physics conceptual

understanding. Using the constructed 5-tier instruments, students’ responses in each tier

would need to be carefully analysed. The answer and reason tiers focus on elucidating

students’ conceptual understanding while the respective confidence ratings indicate the

certainty of students’ responses for these tiers. The fifth tier is for them to articulate their

approaches mathematically – that is, to focus more on mathematical thinking. As the novelty

of these instruments lies in the fifth tier, a thorough examination had to be performed for this

tier, including the analysis of correlations with respect to other tiers. Although Tuminaro

18
(2004) noted that students with sufficient knowledge of mathematics might not approach a

physics question with the right mathematics, we also foresaw that even students with strong

conceptual understanding in physics might have difficulties in expressing their mathematical

thinking. Furthermore, in solving higher order thinking problems, many of the participants

may still overly rely on their scientific intuition rather than mathematical thinking.

(3) How do high-performing students respond to teaching interventions that seek to

promote mathematical thinking on higher order thinking questions in the

selected topics in physics?

The third research question requires an educational study involving teaching

interventions. The teaching interventions should revise on the concepts tested but with a

different approach emphasising the use of mathematical thinking. As elaborated above, this

aspect of mathematical thinking in physics conceptual learning has never been explored in

any educational study. It would be interesting to see how this intervention enhanced students’

ability in performing mathematical thinking at the same time, while the goal of teaching

intervention in most educational studies focuses only on conceptual change. The participants

would be divided into two groups: control group and experimental group. The teaching

intervention would be administered to the experimental group, while the control group

participants would not undergo any meaningful interventions. Both groups would then be

asked to take the same assessments one more time. Students’ performance prior to the

intervention would be taken as pre-test scores and their performance after the intervention

would be taken as post-test scores. The analysis of the post-test might reveal students’

conceptual changes and improvement in their mathematical thinking ability.

19
 The main objectives of this study are:

(1) to use mathematical thinking in the topics of momentum/collision and rotational

mechanics in order to construct higher order thinking questions and develop

diagnostic instruments on those topics;

(2) to examine the association between students’ conceptual understanding and their

mathematical thinking in the topics of momentum/collision and rotational mechanics;

(3) to develop teaching packages that can promote mathematical thinking in the topics of

momentum/collision and rotational mechanics; and

(4) to expose students to teaching interventions on the topics of momentum/collisions

and rotational mechanics.

More importantly, since the study involves the conceptualisation, construction,

administration and grading of assessments where higher order thinking questions are

involved as well as teaching interventions targeted at high-performing students, it would be

appropriate for a person with a strong grounding in physics to carry out this research.

1.4. Glossary of Terms

Table 1.2 provides readers with the definitions or descriptions of terms commonly

used in this thesis. The definitions provided were formulated specifically for this thesis study

and may not be applicable to other research contexts. Readers may find more details about

the terms used in the main text of this thesis.

20
Table 1.2

Glossary of Terms

Term Definition

A test where each question comprises a stem and two tiers

requiring students’ responses (answer and reason), each
2-tier multiple choice (2TMC) test
tier assesses different aspects of students’ cognition or
metacognition

A test where each question comprises a stem and three

tiers requiring students’ responses (answer, reason,
3-tier multiple choice (3TMC) test
confidence for answer and reason), each tier assesses
different aspects of students’ cognition or metacognition

A test where each question comprises a stem and four tiers

requiring students’ responses (answer, confidence for
4-tier multiple choice (4TMC) test
answer, reason, confidence for reason), each tier assesses
different aspects of students’ cognition or metacognition

A test where each question comprises a stem and five tiers

requiring students’ responses (answer, confidence for
5-tier test answer, reason, confidence for reason, explanation), each
tier assesses different aspects of students’ cognition or
metacognition

Students’ conceptions that differ from, or are not

alternative conceptions consistent with those commonly accepted by the scientific
community (Sanger & Greenbowe, 2000)

A point representing the mean position of the matter in a

centre of mass body or system; the unique point where the weighted
relative position of the distributed mass sums to zero

A frame moving together with the centre of mass of the

centre of mass frame
system

A series of collisions between masses (Ricardo & Lee,

chain-collision
2015)

The absolute value of the ratio between the relative

coefficient of restitution velocities after and before a collision (Ricardo & Lee,
2015)

The mental action or process of acquiring knowledge and

cognition
understanding through thought, experience, and the senses

21
 Disagreement and dissatisfaction with one’s existing
cognitive dissonance conception, before they would abandon the conception
and replace it with the correct one

Packages of meaning that capture regularities (similarities

and differences), patterns, or relationships among objects,
concept events, and other concepts that oftentimes represent
different ways of organising the nature or the world (West
& Pines, 1985)

A process where existing conceptions are reorganised

(Tyson et al., 1997) or replaced so as to accommodate new
conceptual change ideas (Smith et al., 1993), which are scientifically
acceptable in students’ conceptual framework (Tekkaya,
2003)

Deep cognitive engagement to facilitate conceptual

conceptual change intervention
change

One’s ability to make a science concept their own such that

they can think with it, apply it in broader areas, state it in
conceptual understanding their own words, find a metaphor or an analogy for it, or
build a mental or physical model of it (Konicek-Moran &
Keeley, 2015)

confidence discrimination quotient An indication of whether the students can distinguish

(CDQ) between what they know and what they do not know

A measure of the self-awareness of students of their own

confidence bias (CB)
performance

A student’s own judgment, or “internal, estimated belief”

confidence rating
(Renner & Renner, 2001) of his/her answer

A curve (resembling a series of arches) traced by a point

cycloid
on a circle being rolled along a straight line

A measure of how well a question can distinguish between

discrimination index (DI)
high and low performers in the test

An encounter between two bodies in which the total

elastic collision kinetic energy of the two bodies after the encounter is
equal to their total kinetic energy before the encounter

A measure of the difficulty level of a question; the

facility index (FI) proportion of students who have chosen the correct
response for an item

22
 Questions that require skills involving analysis,
evaluation, and synthesis (creation of new knowledge),
higher order thinking questions
which are the highest three levels of skills according to
Bloom’s Taxonomy

Students with strong conceptual understanding, indicated

high-performing students
by their high performance in conceptual surveys

An encounter between two bodies in which the total

inelastic (dissipative) collision kinetic energy of the two bodies after the encounter is less
than their total kinetic energy before the encounter

A measure based on the correlations between different

items on the same test (or the same subscale on a larger
internal consistency
test), it measures whether several items that propose to
measure the same general construct produce similar scores

The ability to understand something immediately, without

intuition
the need for conscious reasoning

Kinetic energy that is transferred from one body to another

kinetic energy transfer
through an encounter

A mental operation used in solving problems affirmed by

mathematical thinking the mathematical knowledge and dispositions (Sam &
Yong, 2006)

A process by which an individual is aware of his or her

metacognition
own brain processes that occur during learning

Meta-level thinking processes that are acting on students’

metaconceptual conceptions (Yürük, 2007; Mason & Boscolo, 2000;
Vosniadou, 1994, 2002, 2003; Wiser & Amin, 2001)

An encounter between two bodies in which maximum

amount of kinetic energy has been lost during the
perfectly-inelastic collision
encounter, resulting zero relative velocity after the
collision

Scrutinising thoughts (product of thinking, ideas and

reasoning beliefs) that one has thought about, with facts or data,
logic, etc.

rigid body collision A collision involving at least one rigid body

Kinetic energy of a rigid body associated with its

rotational kinetic energy rotational motion (motion with respect to its centre of
mass)

23
 A bouncing ball that is able to bounce higher than the
superball original height it is dropped from, resulting in a condition
that seemingly violates the conservation of energy

A cognitive process which describes the systematic

transformation of mental representations of knowledge to
thinking
characterise actual or possible states of the world
(Holyoak & Morrison, 2005)

Kinetic energy of a rigid body associated with its

translational kinetic energy
translational motion (motion of its centre of mass)

A technique to indicate that two (or more) methods are

triangulation used in a study in order to check the results of one and the
same subject

1.5. Organisation of Report

This report consists of eight chapters. The background study of this thesis, which

includes the significance of study, research questions and objectives, are presented in Chapter

1. This thesis study consists of two major parts: the first part being on the theoretical

explorations of selected topics in mechanics, which will be elaborated in Chapters 2 and 3,

and the second part being the elaboration of educational studies on mathematical thinking in

physics, which will be elaborated in Chapters 4-6. Chapter 4 will focus on the literature

review on conceptual understanding, representational fluency, alternative conceptions,

conceptual change, mathematical thinking in physics, relevant studies in physics education,

diagnostic instruments, teaching interventions as well as the gaps in the literature that

motivate this thesis study. Chapter 5 will focus on the elaboration of the research

methodology: this involves evaluating suitability of sample for current study, development

of new test instruments, pilot and main studies, interviews, teaching interventions and data

analysis. Chapter 6 will document all the results from the studies, including the pilot and

24
main studies. The general discussion and conclusion of the overall research findings will be

discussed in Chapters 7 and 8. These include future works, study limitations and implications

as well as contributions of this study to new knowledge.

25
 PART I

THEORETICAL EXPLORATIONS OF

SELECTED TOPICS IN MECHANICS

26
 CHAPTER 2

THEORETICAL EXPLORATIONS IN MOMENTUM/COLLISION:

ONE-DIMENSIONAL CHAIN-COLLISION

The first topic that we explored in this thesis study was a topic in mechanics,

momentum/collision. The underlying concepts of momentum/collision are very basic and

fundamental, and they are thoroughly taught in Year 3 (Grade 9) in NUS High School of

Mathematics and Science, Singapore. The concepts involved can be easily found in many

physics toys, e.g. Newton’s cradle (Figure 2.1).

Figure 2.1

Newton’s Cradle

Past research has also observed that if one drops a small “Superball” immediately

after a large one, one directly above the other, the small ball, on rebounding from the large

one, appears to be shot into the air (Mellen, 1968). This demonstration is quite interesting to

both physics teachers and students, as the phenomenon shows an “apparent violation” to the

conservation of energy. This velocity amplification effect was described theoretically using

three different theoretical models by Harter (1971). An extension of the problem to multiple

bodies was described by Kerwin (1972). These phenomena with surprising and

27
counterintuitive results are used regularly in many forms as classroom teaching

demonstrations and novel methods of analysing the problem of chain-collisions, and are still

of interest to teachers and students (Santos et al., 2012).

A few alternative conceptions that happened during the teaching and learning process

in the topic of one-dimensional collision will be elaborated in this chapter. They have been

repeatedly observed even after the qualitative concepts and conceptual formulations were

taught in class and after going through several rounds of problem-solving. Here are some

provoking questions that require one to use a series of mathematical thinking and eventually

lead to the construction of higher order thinking questions in the diagnostic test:

(1) Is conservation of momentum sufficient to describe how a system evolves during a

collision?

(2) Assuming no energy loss, how to maximise the kinetic energy transfer from one block

to another?

(3) Will inserting a stationary block in between the two blocks increase the kinetic energy

transferred from the first block to the last one?

The first question is crucial for a simple one-dimensional collision. After learning the

concept of momentum/collision, one might have a strong tendency to think that if a question

could be solved by the conservation of momentum alone, then there would be no points to

do any further analysis. While conservation of momentum is necessary to be satisfied due to

Newton’s second and third laws, this only constitutes the first layer of thinking. The second

layer would be to analyse the system’s initial and final kinetic energies. If there is no energy

28
released into the system, a collision would only be possible if the final kinetic energy of the

system is less than the initial kinetic energy of the system, or in the best case scenario, equal

to the initial kinetic energy of the system. The former refers to inelastic collision, also known

as dissipative collision, and the latter refers to elastic collision. This fact can also be analysed

from the centre of mass frame of the two bodies. In this frame, the two bodies approach each

other before the collision and move away from each other after the collision. For an elastic

collision, each body has to bounce back with speed equal to its initial speed, as this is the

only way to conserve the system’s total kinetic energy. This implies that the magnitude of

the relative velocity of the two bodies is preserved before and after the collision in both

frames, as relative velocity is invariant across inertial frames. For an inelastic collision, the

magnitude of the relative velocity decreases after the collision. It is impossible for the

magnitude of the relative velocity to increase after the collision unless some energy is

released, such as in the case of an explosion.

For the second question, it is natural that one would expect that the first block should

come to a stop as it transfers all of its momentum or its kinetic energy to the second one. This

is only the first layer of thinking and unfortunately, without further analysis, it ends up as

wrong application of the concept. Without explosion or energy release case, transferring all

momentum of a body to another body is only possible if the system’s final kinetic energy is

less than or equal to its initial kinetic energy. In contrast, transferring all kinetic energy of a

body to another body is only possible if the two bodies have equal mass, for otherwise, the

conservation of momentum would be violated. An indication of the existence of this

alternative conception has been observed during the classroom teaching and learning process.

From the centre of mass frame, it is clear that the best case scenario for kinetic energy transfer

29
occurs when the collision is elastic. This is because, in an elastic collision, no kinetic energy

is lost, and the second body would “bounce” with the greatest possible speed in this frame,

which implies the greatest possible speed in the original frame too.

Many students are naturally expected to answer no to the third question as higher

number of blocks would intuitively imply more momentum loss or more energy loss. This

intuition is unfortunately a false one; in fact, if all collisions are elastic, then there is no energy

loss. Indeed the fact is counterintuitive: there is a possibility in increasing the kinetic energy

transfer by inserting a block and the process could be done recursively. This is valid

especially as the elasticity of the collision is bigger than a certain value depending on the

masses of the first and last blocks.

Surprisingly, all the required analyses to the abovementioned questions do not require

advanced mathematics and only involve simple algebra. Hence, the mathematical analyses

presented in this chapter are still suitable for high school students. But as described earlier,

even students with good mathematical skills may not be able to perform the required

mathematical thinking, especially when they are not accustomed to doing it.

2.1. Introduction

One-dimensional chain-collision is a series of collisions between masses that are

arranged in a straight line, as shown in Figure 2.2 below. The key feature in a chain-collision

is that the collisions happen sequentially while multiple collisions involving the same masses

are ignored. In the case of multi-vehicle collision involving cars of different masses, for

example, the cars are initially far apart. In the “Superball” demonstration, however, the

separations between the balls are negligible, yet the collisions can still be assumed to occur

30
sequentially as the energy wave takes time to travel through the balls. As such, the time taken

between any two collisions is not important as long as each collision is finished before the

next starts or each collision is assumed to occur instantaneously (Muller & Pöschel, 2011).

For simplicity, we will only discuss the case where the first mass has a fixed initial velocity

and the rest of the masses are initially at rest. Nonetheless, the analysis is still valid for

Superball where all masses have initial velocity, as the analysis for each collision can always

be done in the rest frame of one ball. As a natural extension to the Superball demonstration,

the question “What would happen if more than one Superball is used?” often arises.

Figure 2.2

Chain-Collision Involving 𝒏 Masses in between 𝒎𝟏 and 𝒎𝟐

…
𝑣1

𝑚1 𝜇1 𝜇2 𝜇3 𝜇𝑛 𝑚2

In our analysis, 𝑚1 indicates the mass of the first body, 𝑣1 indicates the velocity of

the first body, 𝑚2 indicates the mass of the last body, 𝜇𝑖 indicates the 𝑖-th mass that is put in

between 𝑚1 and 𝑚2 .

The following discussions would show the effect of inserting mass(es) in between

two masses to the maximum kinetic energy transfer, and how the coefficient of restitution

affects it in a chain-collision, while at the same time addressing a few alternative conceptions

that arise from incorrect qualitative reasoning or false intuition. From a simplistic viewpoint,

one would have thought that increasing the number of collisions would give rise to lower

31
energy transfer as each collision increases the amount of energy loss, and the maximum

energy transfer occurs when the first mass becomes stationary after collision. However, as it

will be shown, the final situation after each collision seems counterintuitive as the maximum

kinetic energy transfer can actually be improved by putting the right mass(es) in between and

this maximum energy transfer does not happen when the first mass loses all of its kinetic

energy, as the coefficient of restitution limits the amount of the energy transfer.

The coefficient of restitution, which is basically the absolute value of the ratio

between the relative velocities after and before a collision, as it is going to be explained in

the theory, is a very important concept to understand the behaviour of two masses in a one-

dimensional collision. The coefficient of restitution of a collision between two certain

materials is assumed to be constant, and it is independent of the initial velocities of both

masses. This assumption can be understood as one could just move to the rest frame of one

mass and perform exactly the same analysis – in accordance to one of Einstein’s postulates

on the principle of relativity (Einstein & Davis, 2013). A common alternative conception that

may occur is that, the ratio of kinetic energy loss to the initial kinetic energy, instead of the

ratio of relative velocities, is taken to be constant for a collision. This cannot be true as energy

is not invariant across inertial frames. As such, the percentage of energy loss varies when

analysed in different inertial frames of reference. Thus in general, the coefficient of

restitution can be taken as the parameter which characterises a collision. Some exceptions

occur for viscoelastic particles in which the coefficient of restitution depends on the masses

of the bodies as well as their relative velocity (Ramírez et al., 1999). Although this problem

has been analysed rigorously and more generally for constant and non-constant coefficients

32
of restitution by Pöschel and Brilliantov (2001), we will present a simplified and more

specific case of chain-collisions in great detail for teaching and learning purposes.

2.1.1. Theory

Newton’s second law states that the total external force acting on a system equals to

the rate of change of the system’s momentum. This implies that when there is no net external

force acting on the system, the momentum of the system remains constant. This is the

fundamental principle of the conservation of momentum. If instead we consider two separate

systems interacting with each other, the force of interaction acting on one system would equal

to the rate of change of the momentum of that particular system. From Newton’s third law

we understand that the force acting on the second system will be equal in magnitude but

opposite to the direction of the force acting on the first system. This is known as an action-

reaction pair. Hence, in a given amount of time, whatever amount of increase (or decrease)

in the momentum of the first system will be compensated by the amount of decrease (or

increase) in the momentum of the second system. This perspective allows us to arrive at the

same concept: the total momentum of the combined system remains constant.

Interactions between two bodies can happen through contact forces such as normal

force and friction force, as well as non-contact forces such as gravitational force and magnetic

force. Collision is the simplest example of mechanical interactions. Although the normal

force between the two bodies varies with time during the short duration of collision, the

conservation of momentum still holds. For elastic collisions, which cannot happen in a real

macroscopic experiment, the total kinetic energy of the system is conserved, as we assume

33
no energy is lost. For dissipative collisions, some kinetic energy is lost, mostly to the internal

degrees of freedom as phonons, and some to the surrounding (mainly as sound).

The analysis of relative velocity – which arises from the analysis in the centre of mass

frame – gives rise to the definition of the coefficient of restitution, i.e. the ratio of the relative

velocity of the two bodies after collision to their relative velocity before collision. Since we

only deal with one-dimensional collisions, the use of vector notations in expressing velocities

is redundant. Instead we will use positive sign to indicate rightward velocity and negative

sign to indicate leftward velocity.

In our discussions, the following notations will be used: 𝑚 for mass, 𝑣 for velocity,

𝐾 for kinetic energy, and prime (′) for physical quantity after the collision. Index 1 refers to

the first body and index 2 refers to the second body.

Figure 2.3

One-Dimensional Two-Body Motion before Collision

𝑣1 𝑣2
𝑚1 𝑚2

In Figure 2.3, 𝑣1 > 𝑣2 is necessary for a collision to occur. After the collision, 𝑣1′ ≤

𝑣2′ is implied, as described in Figure 2.4. If suppose 𝑣1′ > 𝑣2′ , the interaction force between

the two bodies still exists to change the momenta of both bodies. The collision ends when at

least the equality holds, which happens in a perfectly inelastic collision.

34
Figure 2.4

One-Dimensional Two-Body Motion after Collision

𝑣1′ 𝑣2′

𝑚1 𝑚2

Conservation of momentum yields

𝑚1 𝑣1 + 𝑚2 𝑣2 = 𝑚1 𝑣1′ + 𝑚2 𝑣2′

𝑚1 (𝑣1 − 𝑣1′ ) = 𝑚2 (𝑣2′ − 𝑣2 )

(2.1)

For a perfectly elastic collision, conservation of kinetic energy yields

1 1 1 2 1 2
𝑚1 𝑣1 2 + 𝑚2 𝑣2 2 = 𝑚1 𝑣1′ + 𝑚2 𝑣2′
2 2 2 2

𝑚1 (𝑣1 − 𝑣1′ )(𝑣1 + 𝑣1′ ) = 𝑚2 (𝑣2′ − 𝑣2 )(𝑣2′ + 𝑣2 )

(2.2)

Combining equations (2.1) and (2.2) yields

𝑣1 + 𝑣1′ = 𝑣2 + 𝑣2′

𝑣1 − 𝑣2 = 𝑣2′ − 𝑣1′

(2.3)

Since 𝑣1 > 𝑣2 , it implies 𝑣1′ < 𝑣2′ for both sides to yield positive values. This result

is usually taught in high school physics class: the absolute value of the relative velocity of

the two bodies remains constant, or is preserved, after a perfectly elastic collision. More

details can be found in any physics textbooks, such as by Young et al. (2012).

Using the definition of the coefficient of restitution (𝑒), we have

35
 𝑣2′ − 𝑣1′
𝑒=−
𝑣2 − 𝑣1

(2.4)

It is obvious that the value of 𝑒 equals one for an elastic collision and zero for a perfectly-

inelastic collision (as 𝑣1′ = 𝑣2′ ). This suggests to us that for any dissipative collision, the

coefficient of restitution must be between these two values, i.e. 𝑒 ∈ [0,1].

In the next section, mathematical thinking will be invoked to show that it is

impossible to transfer the whole momentum or the whole kinetic energy of a moving body

to a stationary body via an elastic collision, unless they have identical mass. Otherwise,

elastic collision is still the best case scenario for energy transfer. Should the collision be

dissipative, the maximum kinetic energy transfer still occurs when their masses are identical,

independent of the coefficient of restitution. Furthermore, the case of three-body and four-

body chain-collisions will also be elaborated as an extension of the analysis, and generalised

for many-body chain-collisions. Lastly, the whole analysis will also show a counterintuitive

fact: inserting intermediary bodies in between two colliding bodies would increase the kinetic

energy transferred, as long as the coefficient of restitution falls within a certain range.

2.1.2. Two-Body Collision

In general, for two-body collisions with coefficient of restitution 𝑒 ∈ [0,1], the

velocities of the two bodies after the collision can be obtained from equations (2.1) and (2.4),

𝑚1 − 𝑒𝑚2 𝑚2 ( 1 + 𝑒 )
𝑣1′ = 𝑣1 + 𝑣
𝑚1 + 𝑚2 𝑚1 + 𝑚2 2

(2.5)

36
 𝑚1 (1 + 𝑒) 𝑚2 − 𝑒𝑚1
𝑣2′ = 𝑣1 + 𝑣
𝑚1 + 𝑚2 𝑚1 + 𝑚2 2

(2.6)

In matrix form,

𝑣′ 1 𝑚 − 𝑒𝑚2 𝑚2 (1 + 𝑒) 𝑣1 1 𝑥−𝑒 1 + 𝑒 𝑣1
( 1′ ) = ( 1 ) (𝑣 ) = ( ( )( )
𝑣2 𝑚1 + 𝑚2 𝑚1 (1 + 𝑒) 𝑚2 − 𝑒𝑚1 2 𝑥 + 1 𝑥 1 + 𝑒) 1 − 𝑒𝑥 𝑣2
𝑚
where 𝑥 = 𝑚1 . Let the initial velocity of the second body, 𝑣2 , be zero.
2

𝑣′ 1 𝑥−𝑒 1 + 𝑒 𝑣1 𝑣1 𝑥−𝑒
( 1′ ) = ( ( )( ) = (𝑥 (1 + 𝑒))
𝑣2 𝑥 + 1 𝑥 1 + 𝑒) 1 − 𝑒𝑥 0 𝑥+1

(2.7)

It is indeed quite tempting to think that the maximum energy transfer should happen

when the first body comes to rest after the collision, so as to give 100% of its momentum, or

100% of its energy. Even students with good conceptual understanding may fall into this

trap. However, equation (2.7) suggest to us that the transfer of 100% momentum from the

first to the second body is impossible. As we can see from the expression for 𝑣1′ , the only

possibility for 100% transfer to happen is when 𝑚1 = 𝑒𝑚2 , or 𝑥 = 𝑒, that is, the type of

collision affects the ratio of masses involved. With this analysis, for an elastic collision, the

two masses have to be identical; and for a perfectly-inelastic collision, the first mass has to

be zero, that is, the first body does not even have a momentum in the first place.

Now, let us analyse the final kinetic energies of both masses.

1 2 (𝑚1 − 𝑒𝑚2 )2 1 (𝑚1 − 𝑒𝑚2 )2

𝐾1′ = 𝑚1 𝑣1′ = ( 𝑚 𝑣 2
) = 𝐾
2 (𝑚1 + 𝑚2 )2 2 1 1 (𝑚1 + 𝑚2 )2 1

1 ′2
(1 + 𝑒)2 𝑚1 𝑚2 1 (1 + 𝑒)2 𝑚1 𝑚2
𝐾2′ = 𝑚2 𝑣2 = 2
( 𝑚 𝑣 )= 𝐾
2 (𝑚1 + 𝑚2 )2 2 1 1 (𝑚1 + 𝑚2 )2 1

37
As mentioned, the fact that the first body stops moving when 𝑚1 = 𝑒𝑚2 can be clearly seen,

but it will imply 𝐾2′ = 𝑒𝐾1 , which is clearly less than unity. Hence, giving up all the kinetic

energy of the first body does not imply transferring the maximum possible amount of kinetic

energy to the second body; it actually contributes to the amount of energy loss. Thus,

unfortunately, this intuition fails.

The fraction of the kinetic energy transferred is

𝐾2′ (1 + 𝑒)2 𝑚1 𝑚2 (1 + 𝑒)2 𝑥

𝜂= = =
𝐾1 (𝑚1 + 𝑚2 )2 (1 + 𝑥 )2

(2.8)
𝑚
where 𝑥 = 𝑚1 . Firstly, it is obvious that for fixed masses, setting 𝑒 = 1 maximises this kinetic
2

energy transfer and it yields

4𝑚1 𝑚2 4𝑥
𝜂elastic = 2
=
(𝑚1 + 𝑚2 ) (1 + 𝑥 )2

(2.9)

Secondly, for a fixed value of 𝑒 < 1, to maximise 𝜂 we have to set the first derivative of 𝜂 to

be zero and to ensure the second derivative of 𝜂 is negative.

𝑑𝜂 (1 + 𝑥 )2 − 2𝑥(1 + 𝑥 ) 1−𝑥
= (1 + 𝑒 )2 ( 4
) = (1 + 𝑒 )2
𝑑𝑥 (1 + 𝑥 ) (1 + 𝑥 )3

𝑑2𝜂 2
−(1 + 𝑥 )3 − 3(1 − 𝑥 )(1 + 𝑥 )2 𝑥−2
2
= ( 1 + 𝑒 ) 6
= 2( 1 + 𝑒 )2
𝑑𝑥 (1 + 𝑥 ) (1 + 𝑥 )4

𝑑𝜂 𝑑2 𝜂
= 0 gives us 𝑥 = 1 and < 0 at 𝑥 = 1, i.e. maximum 𝜂 occurs when the masses are
𝑑𝑥 𝑑𝑥 2

identical. Interestingly, this result has nothing to do with the value of 𝑒. This tells us that the

38
type of collision does not really affect the mass ratio that maximises 𝜂, contrary to the

previous incorrect analysis.

1
𝜂max = (1 + 𝑒 )2
4

(2.10)

The same analysis can also be done by drawing a three-dimensional plot of 𝜂 versus

𝑒 and 𝑥, in equation (2.8). From Figure 2.5, it is obvious that the global maximum transfer

of energy occurs at 𝑥 = 1 and 𝑒 = 1, that is, perfectly elastic case with equal masses. For an

arbitrary value of 𝑥, maximum energy transfer occurs at 𝑒 = 1. And for an arbitrary value of

𝑒, maximum transfer of energy occurs at 𝑥 = 1.

Figure 2.5

(𝟏+𝒆) 𝟐𝒙
Three-Dimensional Plot of 𝜼 = (𝟏+𝒙)𝟐

𝑥
𝑒

39
 So, is it even possible to transfer 100% of energy from the first to the second body?

From equation (2.8), setting 𝜂 = 1,

( 1 + 𝑒 )2 𝑥
=1
(1 + 𝑥 )2

𝑥 − ( 1 + 𝑒 ) √𝑥 + 1 = 0

1 + 𝑒 ± √(𝑒 + 3)(𝑒 − 1)
√𝑥 =
2

(2.11)

There is only one possible solution as the root must be a real number, that is, 𝑒 = 1. This

implies 𝑥 = 1, or 𝑚1 = 𝑚2 , which reproduces the previous result: 100% transfer of kinetic

energy can only occur in an elastic collision involving identical masses.

Contrary to common intuition, it has been shown that in general, maximum kinetic

energy transfer does not happen when the first body becomes stationary after the collision.

This is because giving up all the kinetic energy of the first body would contribute to more

energy loss rather than being transferred to the second body. Additionally, mathematical

analysis has shown that for a fixed type of collision, the maximum energy transfer occurs if

the masses are identical; and for fixed masses, the maximum energy transfer occurs if the

collision is elastic.

2.1.3. Three-Body Chain-Collision

Suppose 𝑚1 and 𝑚2 are fixed; 𝑚2 is initially stationary, as in the previous case. What

if we insert a third stationary mass 𝜇 in the middle?

40
Figure 2.6

Three-Body Chain Collision

𝑣1

𝑚1 𝜇 𝑚2

Using the same analogy and from equation (2.7), after the first collision,

𝑣′ 1 𝑚 − 𝑒1 𝜇 𝜇 (1 + 𝑒1 ) 𝑣1 𝑣1 𝑚1 − 𝑒1 𝜇
( 1′ ) = ( 1 )( ) = (𝑚 (1 + 𝑒 ))
𝑣𝜇 𝑚1 + 𝜇 𝑚1 (1 + 𝑒1 ) 𝜇 − 𝑒1 𝑚1 0 𝑚1 + 𝜇 1 1

and after the second collision (indicated by double-prime) between 𝜇 and 𝑚2 ,

𝑣𝜇′′ 1 𝜇 − 𝑒2 𝑚2 𝑚2 (1 + 𝑒2 ) 𝑣𝜇′ 𝑚1 (1 + 𝑒1 )𝑣1 𝜇 − 𝑒2 𝑚2

( ′′ ) = ( )( ) = ( 𝜇 (1 + 𝑒 ))
𝑣2 𝜇 + 𝑚2 𝜇 (1 + 𝑒2 ) 𝑚2 − 𝑒2 𝜇 0 (𝑚1 + 𝜇)(𝜇 + 𝑚2 ) 2

since 𝑚2 and 𝜇 are initially stationary. This yields

𝑚1 𝜇 (1 + 𝑒1 )(1 + 𝑒2 )
𝑣2′′ = 𝑣
(𝑚1 + 𝜇)(𝜇 + 𝑚2 ) 1

(2.12)
2
1 2 1 𝑚1 𝜇 (1 + 𝑒1 )(1 + 𝑒2 )
𝐾2′′ = 𝑚2 𝑣2′′ = 𝑚2 [ ] 𝑣1 2
2 2 (𝑚1 + 𝜇)(𝜇 + 𝑚2 )

(1 + 𝑒1 )2 (1 + 𝑒2 )2 𝑚1 𝑚2 𝜇2 1
= ( 𝑚1 𝑣1 2 )
(𝑚1 + 𝜇)2 (𝜇 + 𝑚2 )2 2

The fraction of the initial kinetic energy of the first body that is transferred to the second

body is given by

𝐾2′′ (1 + 𝑒1 )2 (1 + 𝑒2 )2 𝑚1 𝑚2 𝜇2
𝜂= =
𝐾1 (𝑚1 + 𝜇)2 (𝜇 + 𝑚2 )2

(2.13)

41
From equation (2.13), it is again obvious that varying the third mass to maximise 𝜂 will yield

a result that is independent of 𝑒1 and 𝑒2 . When this condition is satisfied, neither 𝑚1 nor 𝜇 is

required to be stationary after the collisions, as derived in detail in the earlier section.

The first and second derivatives of 𝜂 are

𝑑𝜂 (𝑚1 + 𝜇)(𝜇 + 𝑚2 ) − 𝜇 (𝑚1 + 𝑚2 + 2𝜇)

= 2(1 + 𝑒1 )2 (1 + 𝑒2 )2 𝑚1 𝑚2 𝜇
𝑑𝑥 (𝑚1 + 𝜇)3 (𝜇 + 𝑚2 )3

𝑑𝜂 𝑚1 𝑚2 − 𝜇2
= 𝑘1 𝜇
𝑑𝑥 (𝑚1 + 𝜇)3 (𝜇 + 𝑚2 )3

where 𝑘1 = 2(1 + 𝑒1 )2 (1 + 𝑒2 )2 𝑚1 𝑚2 .

𝑑2𝜂 (𝑚1 𝑚2 − 3𝜇2 )(𝑚1 + 𝜇)(𝜇 + 𝑚2 ) − 3𝜇(𝑚1 𝑚2 − 𝜇2 )(𝑚1 + 𝑚2 + 2𝜇)

= 𝑘1
𝑑𝑥 2 (𝑚1 + 𝜇)4 (𝜇 + 𝑚2 )4

𝑑𝜂 𝑑2 𝜂
= 0 gives us 𝜇 = √𝑚1 𝑚2 and < 0 when 𝜇 = √𝑚1 𝑚2 , i.e. maximum 𝜂. Thus the
𝑑𝑥 𝑑𝑥 2

maximum fraction of the kinetic energy transferred is

2
(1 + 𝑒1 )(1 + 𝑒2 )𝑚1 𝑚2 (1 + 𝑒1 )2 (1 + 𝑒2 )2 𝑚1 𝑚2
𝜂max =[ ] =
(𝑚1 + √𝑚1 𝑚2 )(𝑚2 + √𝑚1 𝑚2 ) (√𝑚1 + √𝑚2 )4

(2.14)

This result shall now be compared with the 𝜂max for a two-body collision. Firstly, let

us consider two extreme cases: elastic and perfectly-inelastic collisions. For an elastic

collision, we have for a two-body collision (equation (2.9)),

4𝑚1 𝑚2
𝜂2-body =
(𝑚1 + 𝑚2 )2

and for a three-body chain-collision (substituting the maximum values for the coefficients of

restitution, i.e. 𝑒1 = 𝑒2 = 1, to equation (2.14)),

42
 16𝑚1 𝑚2
𝜂3-body =
(√𝑚1 + √𝑚2 )4

Checking the ratio between these two,

𝜂3-body 4(𝑚1 + 𝑚2 )2
=
𝜂2-body (√𝑚1 + √𝑚2 )4

2
𝜂3-body 2(𝑚1 + 𝑚2 ) √𝑚1 − √𝑚2
√ −1 = − 1 = ( ) ≥0
𝜂2-body (√𝑚1 + √𝑚2 )2 √𝑚1 + √𝑚2

It is then obvious that 𝜂3-body ≥ 𝜂2-body , implying that for elastic case, inserting a mass in

between two colliding masses improves the kinetic energy transferred!

For perfectly-inelastic collisions, equations (2.8) and (2.14) read

𝑚1 𝑚2
𝜂2-body =
(𝑚1 + 𝑚2 )2

𝑚1 𝑚2
𝜂3-body =
(√𝑚1 + √𝑚2 )4

Checking the ratio between these two,

𝜂3-body (𝑚1 + 𝑚2 )2
=
𝜂2-body (√𝑚1 + √𝑚2 )4

𝜂3-body 𝑚1 + 𝑚2 2√𝑚1 𝑚2
√ −1 = 2
−1=− ≤0
𝜂2-body (√𝑚1 + √𝑚2 ) (√𝑚1 + √𝑚2 )2

It is then obvious that 𝜂3-body ≤ 𝜂2-body , implying that for perfectly inelastic case, inserting a

mass in between does not improve the kinetic energy transfer!

These two opposite results suggest to us another layer of thinking: a general type of

collision needs to be considered for more complete analysis and accurate analysis. From

equations (2.8) and (2.14),

43
 𝜂3-body (1 + 𝑒1 )2 (1 + 𝑒2 )2 (𝑚1 + 𝑚2 )2
=
𝜂2-body (1 + 𝑒)2 (√𝑚1 + √𝑚2 )4

𝜂
Setting (√𝜂3-body − 1) to be non-negative for an improvement in the energy transferred,
2-body

2
(1 + 𝑒1 )(1 + 𝑒2 )(𝑚1 + 𝑚2 ) ≥ (1 + 𝑒)(√𝑚1 + √𝑚2 )

Considering a special case where 𝑒1 = 𝑒1 = 𝑒, and since 𝑒 is upper-bounded by 1,

1⁄ 1⁄ 2
(𝑚1 2 + 𝑚2 2)
−1≤𝑒 ≤1
𝑚1 + 𝑚2

(2.15)

Invoking mathematical thinking to analyse this theoretical scenario gives us an idea

that inserting a third mass in between two colliding bodies increases the energy transferred

1 1 2
(𝑚1 ⁄2 +𝑚2 ⁄2 )
if the coefficient of restitution is bigger than − 1, and fails to increase the
𝑚1 +𝑚2

energy transferred otherwise. An intuitive and trivial case occurs when all three masses are
1
identical. In such case, 𝜂3-body = 1 if all collisions are elastic, as expected; and 𝜂3-body = 16

if all collisions are perfectly inelastic. Note that for the latter, the first mass is assumed to not

be involved in the second collision, it needs to be detached from the intermediary mass, as

chain-collision assumes that all collisions happen sequentially.

2.1.4. Four-Body Chain-Collision

Since we now have an idea that inserting a third body in between two colliding bodies

improves the kinetic energy transferred as long as the coefficient of restitution falls within a

44
certain range of values, the idea can always be extended to a four-body chain-collision.

Suppose 𝑚1 and 𝑚2 are fixed; 𝑚2 , 𝜇1 , and 𝜇2 are initially stationary.

Figure 2.7

Four-Body Chain-Collision

𝑣1

𝑚1 𝜇1 𝜇2 𝑚2

Analogous to the previous result for a three-body chain-collision, we should have

𝜇1 = √𝑚1 𝜇2 and 𝜇2 = √𝜇1 𝑚2 for the maximum kinetic energy transferred, regardless of the

2⁄ 1 1⁄ 2
type of each collision. It is equivalent to 𝜇1 = 𝑚1 3 𝑚2 ⁄3 and 𝜇2 = 𝑚1 3 𝑚2 ⁄3 . These

results can also be obtained using an analogy to the previous expressions,

(1 + 𝑒1 )2 𝑚1 𝜇1 (1 + 𝑒2 )2 𝜇1 𝜇2 (1 + 𝑒3 )2 𝜇2 𝑚2
𝜂=
(𝑚1 + 𝜇1 )2 (𝜇1 + 𝜇2 )2 (𝜇2 + 𝑚2 )2

(2.16)

𝜕𝜂
= 0 yields
𝜕𝜇1

𝑘2 𝜇2 (𝑚1 + 𝜇1 )(𝜇1 + 𝜇2 ) − 𝜇1 (𝑚1 + 2𝜇1 + 𝜇2 )

2
[ ]=0
(𝜇2 + 𝑚2 ) (𝑚1 + 𝜇1 )3 (𝜇1 + 𝜇2 )3

where 𝑘2 = (1 + 𝑒1 )2 (1 + 𝑒2 )2 (1 + 𝑒3 )2 𝑚1 𝑚2 𝜇1 𝜇2 .

𝑚1 𝜇1 + 𝑚1 𝜇2 + 𝜇1 2 + 𝜇1 𝜇2 = 𝑚1 𝜇1 + 2𝜇1 2 + 𝜇1 𝜇2

𝜇1 = √𝑚1 𝜇2

𝜕𝜂
and 𝜕𝜇 = 0 yields
2

45
 𝑘2 𝜇1 (𝜇1 + 𝜇2 )(𝜇2 + 𝑚2 ) − 𝜇2 (𝜇1 + 2𝜇2 + 𝑚2 )
[ ]=0
(𝑚1 + 𝜇1 )2 (𝜇1 + 𝜇2 )3 (𝜇2 + 𝑚2 )3

𝜇1 𝜇2 + 𝜇1 𝑚2 + 𝜇2 2 + 𝜇2 𝑚2 = 𝜇1 𝜇2 + 2𝜇2 2 + 𝜇2 𝑚2

𝜇2 = √𝜇1 𝑚2

as expected. The maximum fraction of the kinetic energy transferred here is

(1 + 𝑒1 )2 (1 + 𝑒2 )2 (1 + 𝑒3 )2 𝑚1 𝑚2
𝜂4-body =
1⁄ 1⁄ 6
(𝑚1 3 + 𝑚2 3)

(2.17)

To check whether the kinetic energy transferred has increased, we have to check their ratio.

We shall now consider the case where 𝑒1 = 𝑒2 = 𝑒3 = 𝑒. From equations (2.14) and (2.17),

𝜂4-body (1 + 𝑒)2 (√𝑚1 + √𝑚2 )4

= 6
𝜂3-body 1 1
(𝑚1 ⁄3 + 𝑚2 ⁄3 )

𝜂
For an improvement to happen, we set √𝜂3-body − 1 to be non-negative.
2-body

2 1⁄ 1⁄ 3
(1 + 𝑒)(√𝑚1 + √𝑚2 ) ≥ (𝑚1 3 + 𝑚2 3)

The condition for improvement is then

1⁄ 1⁄ 3
(𝑚1 3 + 𝑚2 3)
−1≤𝑒 ≤1
1⁄ 1⁄ 2
(𝑚1 2 + 𝑚2 2)

(2.18)

When the coefficient of restitution falls within this range, putting two bodies in

between the original two colliding bodies can increase the kinetic energy transferred. This

conclusion can also be obtained by inquiring in the following way: as the existence of 𝜇1

46
increases the kinetic energy transferred from 𝑚1 to 𝑚2 , in the same way the existence of 𝜇2

should increase the kinetic energy transferred from 𝜇2 to 𝑚2 , as it has been proven for a

three-body chain-collision. Hence, the overall kinetic energy transferred can be improved as

long as the above conditions are satisfied. And when these conditions are satisfied, none of

the masses are required to be stationary after the collision, as concluded before.

Now, comparing the lower bound for the coefficient of restitution for a three-body

chain-collision with the one for a two-body collision,

1⁄ 1⁄ 3
(𝑚1 3 + 𝑚2 3)
[ ]
1⁄ 1⁄ 2 1⁄ 1⁄ 3 1⁄ 3
(𝑚1 2 + 𝑚2 2) (𝑚1 3 + 𝑚2 3 ) (𝑚1 + 𝑚2 ) (𝑥 3 + 1) (𝑥 + 1)
2 = 4 = 4
1 1 1 1 1⁄
(𝑚1 ⁄2 + 𝑚2 ⁄2 ) (𝑚1 ⁄2 + 𝑚2 ⁄2 ) (𝑥 2 + 1)
[ ]
𝑚1 + 𝑚2

1 1 3
(𝑚1 ⁄3 +𝑚2 ⁄3)
This function has a global minimum at 𝑥 = 1. Hence, 1 1 2 − 1 ≤ 𝑒 ≤ 1 is a stricter
(𝑚1 ⁄2 +𝑚2 ⁄2)

1 1 2
(𝑚1 ⁄2+𝑚2 ⁄2)
requirement as compared to − 1 ≤ 𝑒 ≤ 1.
𝑚1+𝑚2

2.2. Generalisation: Many-Body Chain-Collision

Figure 2.8

Many-Body Chain-Collision

…
𝑣1

𝑚1 𝜇1 𝜇2 𝜇3 𝜇𝑛 𝑚2

47
 At this stage, mathematical thinking can be invoked by applying the idea obtained in

the previous sections, recursively: “Putting 𝑛 bodies in between colliding 𝑚1 and 𝑚2 would

possibly improve the kinetic energy transferred,” where the intermediary masses are given

by {𝜇𝑖 |1 ≤ 𝑖 ≤ 𝑛}, so that there would be (𝑛 + 1) collisions involving a total of (𝑛 + 2)

bodies. Assuming the coefficient of restitution for all collisions is identical, its value should

be within the range of

1⁄ 1⁄ 𝑛+1
(𝑚1 𝑛+1 + 𝑚2 𝑛+1 )
𝑛 −1 ≤𝑒 ≤1
(𝑚1 1⁄𝑛 + 𝑚2 1⁄𝑛 )

(2.19)

Moreover, for maximum improvement to the kinetic energy transferred, the inserted masses

should satisfy

√𝑚1 𝜇2 for 𝑖 = 1
𝜇𝑖 = {√𝜇𝑖−1 𝜇𝑖+1 for 1 < 𝑖 < 𝑛
√𝜇𝑛−1 𝑚2 for 𝑖 = 𝑛

(2.20)

This condition is equivalent to

𝑚1 𝜇𝑖 𝜇𝑛
= = , ∀ 𝑖 ∈ [1, 𝑛 − 1]
𝜇1 𝜇𝑖+1 𝑚2

(2.21)

If this constant mass-ratio is defined as 𝑥, we obtain the final velocity of each mass in between

𝑚1 and 𝑚2 after collisions to be

[𝑥(1 + 𝑒)]𝑖
𝑣𝜇𝑖 = (𝑥 − 𝑒)𝑣1
(𝑥 + 1)𝑖+1

(2.22)

48
and the final velocity of 𝑚1 to be

𝑥−𝑒
𝑣
𝑥+1 1

(2.23)

and the final velocity of 𝑚2 to be

𝑛+1
𝑥 (1 + 𝑒 )
[ ] 𝑣1
𝑥+1

(2.24)

And the maximum fraction of the kinetic energy of 𝑚1 that is transferred to 𝑚2 is

(1 + 𝑒)2(𝑛+1) 𝑚1 𝑚2
𝜂(n+2)−body = 2(𝑛+1)
1⁄ 1⁄
(𝑚1 𝑛+1 + 𝑚2 𝑛+1 )

(2.25)

2.3. Chapter Summary

We can conclude that the kinetic energy transferred from one body (mass 𝑚1 ) to an

initially-stationary body (mass 𝑚2 ) can be improved by inserting a mass in between the two

of them. This idea can then be applied recursively to inserting 𝑛 masses in between the

original two masses, as long as the coefficient of restitution for each collision is bigger than

1 1 𝑛+1
(𝑚1 ⁄𝑛+1+𝑚2 ⁄𝑛+1 )
1 1 𝑛 − 1 and the mass of each body is chosen appropriately so that the mass
(𝑚1 ⁄𝑛 +𝑚2 ⁄𝑛)

ratio between two adjacent masses is identical throughout the chain. Moreover, this mass

ratio does not depend at all on the coefficient of restitution of each collision.

49
 Both physics students and teachers with strong basic conceptual understanding on

momentum/collision could easily perform inaccurate analysis in attempting to answer the

questions. Firstly, to transfer the energy of the first mass to the second mass as much as

possible, the first mass is not required to come to rest. This is because the transfer of 100%

momentum or 100% energy only occurs if the collision is elastic and the masses are identical.

Should the first mass come to a stop after a dissipative collision, there would then be some

energy loss. Instead, for fixed masses, elastic collision is still the best case scenario for the

energy transfer. Secondly, putting more masses in between could potentially increase the

amount of energy transfer, even in a dissipative collision. These results rectify a few possible

alternative conceptions due to overreliance in one’s scientific intuition.

50
 CHAPTER 3

THEORETICAL EXPLORATIONS IN ROTATIONAL MECHANICS:

RIGID BODY COLLISION

In this chapter, the concepts involved in rigid body collision are explored. These

concepts are taught as early as Year 3 (Grade 9) and they are further developed in Years 5

and 6 (Grades 11 and 12) in NUS High School of Mathematics and Science, Singapore.

3.1. Introduction

The questions constructed for the first study assessment were inspired by an

elementary problem in rotational mechanics, which is discussed in an article written by

Lemos (2008) with at least one aspect that does not conform to what qualitative reasoning or

intuition suggests. It involves a collision shown in Figure 3.1, where a projectile hits a rod

perpendicularly on a smooth horizontal table.

Figure 3.1

Collision between Projectile and Stationary Rigid Rod

51
 This particular problem was originally taken from an undergraduate physics textbook

(Resnick & Kenneth, 2001) and the analysis presented in this chapter is suitable for high-

performing high school students. Some questions that would provoke one’s mathematical

thinking are:

(1) Is conservation of linear momentum still applicable in a rigid body collision?

(2) Is it possible to find a reference point where the angular momentum of the rod (the

projectile is excluded) is conserved?

(3) At which point should the rod be hit by the projectile to acquire maximum angular

speed? Will the masses of the rod and projectile affect the required point of collision

for maximum angular speed of the rod to be attained?

(4) Should the projectile transfer all of its kinetic energy to the rod to create maximum

angular speed?

The first question may seem obvious, the linear momentum of a system is conserved

as long as the system is isolated, i.e. no net external force acts on the system, or when the net

external force acting on the system equals to zero. However, based on teaching experience,

a lot of students would actually think that some of the linear momentum could be transformed

into angular momentum as the body rotates. This is indeed one obvious alternative

conception as linear momentum and angular momentum are two different physical quantities

of two different dimensions and units, they cannot be transformed into one another. For the

second question, a lot of physics students would naturally think that it is impossible to isolate

52
the rod from the projectile and yet having its angular momentum conserved. A glance at

Figure 3.1 would also easily tell us that the rod is initially at rest but is finally moving after

the collision. However, taking the reference point at the point of collision would yield zero

torque no matter how large the force exerted by the projectile, hence, angular momentum of

the rod about this point is conserved. The third question would also go against scientific

intuition, as one would naturally expect that larger 𝑎 implies larger 𝜔, with 𝑎 = ±ℓ giving

the largest angular speed. This happens as our qualitative reasoning tells us incorrectly that

the largest angular speed is achieved when the force is applied as far as possible from the

rotational axis, or the fulcrum, in this case, the centre of mass of the rod. And this fact is

supported by our everyday experience with doors and levers, which has become false

intuition to many physics students. This false intuition would potentially lead students to

think that the masses of the rod and projectile should not affect the required location of the

point of collision for maximum angular speed of the rod to be attained. For the fourth

question, intuitively the projectile is expected to transfer all its kinetic energy to the rotational

kinetic energy of the rod, resulting in the projectile coming to rest after the collision.

Unfortunately, this is also another false intuition. In the centre of mass frame of the projectile

and rod, they approach each other with equal momentum before the collision and move away

from each other with equal momentum after the collision. Maximising the rotational kinetic

energy of the rod after the collision implies minimising the translational kinetic energies of

both the projectile and rod. In the original frame, this occurs only if the final velocity of the

projectile equals to the final centre of mass velocity of the rod, i.e. the projectile and rod’s

centre of mass move together with equal velocity.

53
 This chapter provides a thorough and comprehensive mathematical analyses as well

as some generalisations of the abovementioned elementary problem in rotational mechanics,

involving a projectile and a rod on a smooth table. The generalisations include the unevenness

(non-uniformity) of the rod’s mass distribution, the elasticity of the collision, and the case

where the rod is pivoted. Here, we present nice analytical steps that address some expected

errors in students’ way of thinking due to the limitation of qualitative reasoning as well as

intuition.

3.1.1. Theory

Dynamics is a branch of classical mechanics concerned with the study of forces and

torques and their effect on motion, as opposed to kinematics, which studies the motion of

objects without reference to its causes. Isaac Newton defined the fundamental physical laws

which govern dynamics in physics, especially his second law of motion. Just as the study of

Newtonian dynamics begins by defining a force and its relation to the system’s momentum,

the study of rotational mechanics starts by defining our analogue to a force – a torque – and

its relation to the system’s angular momentum.

Firstly, it is necessary that basic concepts in rotational mechanics are revisited to

provide more thorough understanding of conservation laws.

(1) Conservation of Linear Momentum

In a closed system (one that does not exchange any matter with its surroundings and

is not acted on by external forces 𝐅⃗ext ) the total linear momentum is constant. This fact,

54
known as the law of conservation of linear momentum, is implied by Newton’s laws of

motion (Feynman et al., 2013).

∫ 𝐅⃗ext 𝑑𝑡 = 0 ⇒ 𝐩
⃗⃗𝑖 = 𝐩
⃗⃗𝑓

(3.1)

where ∫ 𝐅⃗ext 𝑑𝑡 denotes the total external linear impulse, which is necessary to be zero (or

negligible) for the system’s linear momentum to be preserved; 𝐩

⃗⃗𝑖 denotes the initial linear

⃗⃗𝑓 denotes the final linear momentum of the system. For a

momentum of the system, and 𝐩

rigid body, like the rod presented in Figure 3.1, its linear momentum can be obtained by

multiplying its mass with the velocity of its centre of mass.

(2) Conservation of Angular Momentum

Similar to linear motion, the law of conservation of angular momentum states that

when no (or negligible) external angular impulse by external torques 𝛕

⃗⃗ext acts on an object

or a closed system of objects, the system’s angular momentum remains constant (Podolsky,

1966). Commonly the point of collision is taken to be the reference point so as to make the

angular impulse acting on each body to be zero. But when a pivot is introduced, the pivot

needs to be taken as the reference point so as to make the angular impulse acting on the whole

system to be zero.

∫𝛕 ⃗𝑖 = 𝐋
⃗⃗ext 𝑑𝑡 = 0 ⇒ 𝐋 ⃗𝑓

(3.2)

55
where ∫ 𝛕
⃗⃗ext 𝑑𝑡 denotes the total external angular impulse, which is necessary to be zero (or

negligible) for the system’s angular momentum to be preserved, or for the conservation of

⃗ 𝑖 denotes the initial angular momentum of the system,

angular momentum to be applied; 𝐋

⃗ 𝑓 denotes the final angular momentum of the system.

and 𝐋

(3) Conservation of Energy

The law of conservation of energy states that when no (or negligible) work is done by

the non-conservative forces 𝐅⃗nc , the system does not lose its energy.

∫ 𝐅⃗nc ∙ 𝑑𝐫⃗ = 0 ⇒ 𝐸𝑖 = 𝐸𝑓

(3.3)

where ∫ 𝐅⃗nc ∙ 𝑑𝐫⃗ denotes the total work done by the non-conservative forces, which is

necessary to be zero (or negligible) for the system’s total mechanical energy to be preserved;

𝐸𝑖 denotes the initial mechanical energy of the system, and 𝐸𝑓 denotes the final mechanical

energy of the system. Note that some non-conservative forces may exist in a system, but as

long as they do not perform any work on the system, conservation of energy still holds. Some

examples would be static friction force and force acted by the pivot.

(4) Moment of Inertia

Moment of inertia can be defined as the “laziness of an object to rotate or to change

its angular velocity”. It is a property of a rigid body, which can be obtained by integration or

by making use of the symmetrical property of the shape using alternative methods, i.e. scaling

56
(Ricardo, 2015) and squashing (Wang & Ricardo, 2019) methods. The moment of inertia of
1
a uniform rod of mass 𝑚 and length 2ℓ about its centre of mass is given by 𝐼0 = 12 𝑚(2ℓ)2 =

1
𝑚ℓ2 . And if a rotational axis is located at a distance 𝑥 away from the centre of mass of the
3

rod, its moment of inertia about the axis can be obtained by using the parallel axis theorem,

1
𝐼 = 𝐼0 + 𝑚𝑥 2 = 𝑚ℓ2 + 𝑚𝑥 2
3

(3.4)

(5) Coefficient of Restitution

Suppose a projectile of mass 𝑚 and linear velocity 𝑣0 collides elastically and

perpendicularly with a rigid rod of mass 𝑀 and length 2ℓ having an initial linear velocity 𝑣1

in the same direction as 𝑣0 , on a smooth horizontal table. The mass of the rod is assumed to

be uniformly distributed.

Figure 3.2

Collision between Projectile and Moving Rigid Rod

57
 Let 𝑉 and 𝑣 indicate the final linear velocities of the rod’s centre of mass and the

projectile, respectively, and let 𝜔 indicate the final angular velocity of the rod (Figure 3.2).

For all linear velocities involved, positive sign signifies rightward motion and negative sign

leftward motion. For the angular velocity, positive sign signifies anticlockwise rotation and

negative sign signifies clockwise rotation.

Assuming an elastic collision, applying conservation laws in (3.1), (3.2) and (3.3),

and using the moment of inertia expressed in (3.4) yields

𝑚𝑣0 + 𝑀𝑣1 = 𝑚𝑣 + 𝑀𝑉

(3.5)

1
𝑀𝑣1 𝑎 = 𝑀𝑉𝑎 − ( 𝑀ℓ2 ) 𝜔
3

(3.6)

1 1 1 1 1 1
𝑚𝑣0 2 + 𝑀𝑣1 2 = 𝑚𝑣 2 + 𝑀𝑉 2 + ( 𝑀ℓ2 ) 𝜔2
2 2 2 2 2 3

(3.7)

From equations (3.5) and (3.7),

𝑎2
𝑚(𝑣0 − 𝑣)(𝑣0 + 𝑣) = 𝑀 (𝑉 − 𝑣1 )(𝑉 + 𝑣1 ) + 3𝑀 (𝑉 − 𝑣1 )2
ℓ2

𝑎2
(𝑣0 + 𝑣) = (𝑉 + 𝑣1 ) + 3(𝑉 − 𝑣1 )
ℓ2

𝑎2
𝑣0 − 𝑣1 = 𝑉 − 𝑣 + 3(𝑉 − 𝑣1 )
ℓ2

(3.8)

Coefficient of restitution characterises the type of one-dimensional collision

𝑉−𝑣
(Thornton, 1997). It is commonly defined as 𝜀 = 𝑣 . This definition does not account for
0−𝑣1

58
the rotation of the rod, and hence its value is not normalised – it is not equal to one when the

collision is elastic, neither is it zero when the collision is perfectly-inelastic (i.e. the projectile

sticks to the rod instantaneously after the collision). To normalise it, the coefficient of

restitution in this specific case can be defined as

𝑎2
(𝑉 − 𝑣) + 3(𝑉 − 𝑣1 )
𝜀= ℓ2
(𝑣0 − 𝑣1 )

(3.9)

This is valid to characterise the elasticity of the collision since the value of coefficient

of restitution is invariant across inertial frames of reference. For instance, in a frame moving

to the left with constant speed 𝑢, all the linear velocities in equation (3.9) are added by 𝑢,

and hence the value of 𝜀 remains unchanged.

From equation (3.8), it is obvious that using the definition in (3.9) for an elastic

collision, 𝜀 = 1. For a perfectly-inelastic collision, the projectile sticks to the rod, i.e. 𝑣 =

𝑎2
𝑉 + 𝜔𝑎 = 𝑉 + 3(𝑉 − 𝑣1 ) ℓ2 , and hence 𝜀 = 0. Therefore, for any dissipative collision, 𝜀 ∈

(0,1).

3.2. Generalisations

3.2.1. First Generalisation: Unevenness (Non-uniformity) of Rod

Consider the case where 𝑣1 = 0 and the collision is elastic. To generalise the problem,

let the centre of mass of the rod be located at a distance 𝑥 away (taken to be a non-negative

value) from the centre of the rod, 𝑥 ∈ [0, ℓ], and its moment of inertia about its centre of

mass be 𝐼CM (Figure 3.3). The point of collision is at 𝑎 ∈ [−ℓ, ℓ].

59
Figure 3.3

Generalisation to Projectile-Rod Collision: Unevenness (Non-Uniformity) of the Rod

As a result of the collision, the projectile and the centre of mass of the rod will move

horizontally – since there is only a horizontal force acting on them during the short time of

collision – and at the same time, the rod will rotate about its centre of mass.

Using the sign convention described above, applying equations (3.1), (3.2), and (3.3),

𝑚𝑣0 = 𝑚𝑣 + 𝑀𝑉

𝑀𝑉 (𝑎 − 𝑥 ) = 𝐼CM 𝜔

1 1 1 1
𝑚𝑣0 2 = 𝑚𝑣 2 + 𝑀𝑉 2 + 𝐼CM 𝜔2
2 2 2 2

yields

(𝑚 − 𝑀 )𝐼CM + 𝑀𝑚(𝑎 − 𝑥 )2
𝑣= 𝑣
(𝑚 + 𝑀 )𝐼CM + 𝑀𝑚(𝑎 − 𝑥 )2 0

(3.10)

2𝑚𝐼CM
𝑉= 𝑣
(𝑚 + 𝑀 )𝐼CM + 𝑀𝑚(𝑎 − 𝑥 )2 0

(3.11)

60
 2𝑀𝑚(𝑎 − 𝑥 )
𝜔= 𝑣
(𝑚 + 𝑀 )𝐼CM + 𝑀𝑚(𝑎 − 𝑥 )2 0

(3.12)

It is clear from equation (3.11) that there is no way for the rod to move leftward as the

expression is always positive. However, equation (3.10) indicates that the projective will
𝑀
move leftward if 𝑚 < 𝑀(𝑎−𝑥) 2
, and it will stop moving after the collision if the equality
1+
𝐼CM

holds. Moreover from equation (3.12), the rod will rotate clockwise if 𝑎 < 𝑥, and it will stop

rotating after the collision if the equality holds. This fact is rather obvious: the rod will rotate

clockwise if the projectile hits the rod at a point located higher than the rod’s centre of mass.

Where then should the projectile hit the rod for the rod to attain maximum angular

speed after the collision?” To answer this, the derivatives of 𝜔 with 𝑎 need to be explored.

𝑑𝜔 2𝑀𝑚𝑣0
= ((𝑚 + 𝑀)𝐼CM − 𝑀𝑚(𝑎 − 𝑥 )2 )
𝑑𝑎 [(𝑚 + 𝑀 )𝐼CM + 𝑀𝑚(𝑎 − 𝑥 )2 ]2

𝑑2𝜔 (2𝑀𝑚)2 𝑣0 (𝑎 − 𝑥 )
= − [3(𝑚 + 𝑀)𝐼CM − 𝑀𝑚(𝑎 − 𝑥 )2 ]
𝑑𝑎2 [(𝑚 + 𝑀 )𝐼CM + 𝑀𝑚(𝑎 − 𝑥 )2 ]3

The first derivative goes to zero if

(𝑚 + 𝑀 )𝐼CM
𝑎 =𝑥±√
𝑀𝑚

(3.13)

𝑑2 𝜔 (𝑚+𝑀)𝐼CM 𝑑2 𝜔
It is also obvious that < 0 and 𝜔 > 0 for 𝑎 = 𝑥 + √ ; > 0 and 𝜔 < 0 for
𝑑𝑎 2 𝑀𝑚 𝑑𝑎 2

(𝑚+𝑀)𝐼CM
𝑎 =𝑥−√ , which essentially correspond to the local maxima of the angular speed.
𝑀𝑚

The corresponding expressions for the maximum 𝜔’s are

61
 𝑀𝑚
𝜔max = ±√ 𝑣
(𝑚 + 𝑀 )𝐼CM 0

(3.14)

(𝑚+𝑀)𝐼CM
Note that since 𝑎 ∈ [−ℓ, ℓ] and 𝑥 is a non-negative value, for 𝑎 = 𝑥 + √ to
𝑀𝑚

(𝑚+𝑀)𝐼CM (𝑚+𝑀)𝐼CM
exist it is necessary to have 𝑥 ∈ [0, ℓ − √ ], and for 𝑎 = 𝑥 − √ to exist it
𝑀𝑚 𝑀𝑚

(𝑚+𝑀)𝐼CM (𝑚+𝑀)𝐼CM
is necessary to have 𝑥 ∈ [√ − ℓ, ℓ]. If √ ∈ [0, ℓ) then both local maxima
𝑀𝑚 𝑀𝑚

(𝑚+𝑀)𝐼CM (𝑚+𝑀)𝐼CM
exist for 𝑥 ∈ [0, ℓ − √ ]. If √ ∈ [ℓ, 2ℓ] then only one local maximum
𝑀𝑚 𝑀𝑚

(𝑚+𝑀)𝐼CM
exists for 𝑥 ∈ [√ − ℓ, ℓ] whose value never falls below the global maximum.
𝑀𝑚

1
For instance, consider a particular case where the rod is uniform, 𝑥 = 0, 𝐼 = 3 𝑚ℓ2

(from equation (3.4)), equations (3.13) and (3.14) yield

𝑚+𝑀
𝑎 = ±ℓ√
3𝑚

(3.15)

3𝑚 𝑣0
𝜔max = ±√
𝑚+𝑀 ℓ

(3.16)

𝑚+𝑀
with a condition that ℓ√ ≤ ℓ or 𝑀 ≤ 2𝑚. Here we see the limitations to qualitative
3𝑚

analysis as well as the use of intuition as the result is counterintuitive to many: the maximum

62
angular speed is attained not when the projectile hits the rod’s end. In fact, if 𝑎 = ±ℓ,

6𝑚 𝑣
equation (3.12) gives 𝜔 = ± 4𝑚+𝑀 ℓ0, which is at most equal to the one obtained in equation

(3.16). To prove that,

2
3𝑚 6𝑚 3𝑚(2𝑚 − 𝑀)2
−( ) = ≥0
𝑚+𝑀 4𝑚 + 𝑀 (4𝑚 + 𝑀)2 (𝑚 + 𝑀 )

The equality is achieved when 𝑀 = 2𝑚, 𝑎 = ±ℓ, i.e. the greatest angular speed is attained

when the projectile hits the rod at one end. When 𝑀 > 2𝑚, no local maxima can be found

and the rod should be hit at one end, 𝑎 = ±ℓ, to attain the greatest angular speed, 𝜔max =
6𝑚 𝑣
± 4𝑚+𝑀 ℓ0 .

Another pitfall commonly happens when one argues based on qualitative reasoning

that the maximum angular speed of the rod is attained when the projectile stops moving

immediately after the collision, so as to give off its whole kinetic energy to the rod or its

whole momentum to the rod. This alternative conception is similar to the one in the case of

one-dimensional collision between two blocks (as discussed in the previous section).
𝑚
Substituting equation (3.13) to equations (3.10) and (3.11), 𝑣 = 𝑉 = 𝑚+𝑀 𝑣0 , indicates that

the maximum angular speed of the rod is attained when the projectile and rod’s centre of

mass move with identical linear velocities after the collision. This is similar to the case of

inelastic collision in the absence of rotation. However, the supposedly-dissipated energy has

now appeared in the form of the rotational kinetic energy. Note that this is true even for the

case where the rod is not uniform.

63
 On the other hand, if the rod is hit at its centre of mass, 𝑎 = 𝑥, there will be no rotation

involved, and the collision is similar to a one-dimensional collision between two blocks, 𝑣 =
𝑚−𝑀 2𝑚
𝑣0 , 𝑉 = 𝑚+𝑀 𝑣0 , and 𝜔 = 0.
𝑚+𝑀

3.2.2. Second Generalisation: Elasticity of Collision

Consider the case where 𝑣1 = 0 and the mass of the rod is uniformly distributed, i.e.
1
𝑥 = 0, 𝐼CM = 3 𝑀ℓ2 (Figure 3.4). The point of collision is at 𝑎 ∈ [−ℓ, ℓ]. It shall be assumed

that the coefficient of restitution is 𝜀, where 𝜀 ∈ [0,1], inclusive of the two extreme cases:

elastic and perfectly-inelastic collisions.

Figure 3.4

Generalisation to Projectile-Rod Collision: Elasticity of Collision

As a result of the collision, the projectile and the centre of mass of the rod move horizontally

– since there is only a horizontal force acting on them during the short time of collision – and

at the same time, the rod will rotate about its centre of mass.

64
 Using the sign convention described above, applying equations (3.5), (3.6), and (3.7),

with 𝑣1 = 0,

𝑚𝑣0 = 𝑚𝑣 + 𝑀𝑉

1
0 = 𝑀𝑉𝑎 − ( 𝑀ℓ2 ) 𝜔
3

𝑎2
(𝑉 − 𝑣) + 3𝑉
𝜀= ℓ2
𝑣0

yields

3𝑎2
𝑚 (1 + ) − 𝑀𝜀
ℓ2
𝑣= 𝑣0
3𝑎2
𝑚 (1 + 2 ) + 𝑀
ℓ

(3.17)

𝑚 (1 + 𝜀 )
𝑉= 𝑣0
3𝑎2
𝑚 (1 + 2 ) + 𝑀
ℓ

(3.18)

𝑎 3𝑚(1 + 𝜀 )𝑣0
𝜔= 2
3𝑎 ℓ2
𝑚 (1 + 2 ) + 𝑀
ℓ

(3.19)

Equation (3.18) dictates that there is no way for the rod to move leftward as the

expression always yields a positive number. However, equation (3.17) indicates that the

3𝑎 2
projectile will move leftward if 𝑚 (1 + ) < 𝑀𝜀, and it will stop moving after the collision
ℓ2

if the equality holds. Moreover, from equation (3.19), the rod will rotate clockwise if 𝑎 < 0,

and it will stop rotating if the equality holds, which agrees to common sense.

65
 To find the local maxima of 𝜔, the derivatives of 𝜔 need to be explored.

3𝑚𝑎2
𝑑𝜔 3𝑚(1 + 𝜀 )𝑣0 𝑀+𝑚−
= ℓ2
𝑑𝑎 ℓ2 2
3𝑎2
[𝑚 (1 + 2 ) + 𝑀]
ℓ

𝑚𝑎2
2
𝑑 𝜔 2
54𝑚 𝑎(1 + 𝜀 )𝑣0 𝑀+𝑚−
=− ℓ2
𝑑𝑎 2 ℓ4 3
3𝑎2
[𝑚 (1 + ) + 𝑀]
ℓ2

The first derivative goes to zero if

𝑀+𝑚
𝑎 = ±ℓ√
3𝑚

(3.20)

𝑑2 𝜔
and is clearly negative when 𝑎 takes positive root and positive when 𝑎 takes negative
𝑑𝑎 2

root, which essentially correspond to the local maxima of the angular speed. The result is

surprising, and again, counterintuitive as regardless of the elasticity of the collision, the

𝑀+𝑚
projectile has to hit the rod at a point 𝑎 = ±ℓ√ from the centre of the rod to maximise
3𝑚

its angular speed. In general, these points are not the rod’s ends except if 𝑀 = 2𝑚. Since this

result is not even affected by the coefficient of restitution, it still holds for the extreme two

cases: 𝜀 = 1 (elastic collision) and 𝜀 = 0 (inelastic collision).

The corresponding expressions for the maximum 𝜔’s are given by

3𝑚 (1 + 𝜀 )𝑣0
𝜔max = ±√
𝑀+𝑚 2ℓ

(3.21)

66
which is identical to equation (3.16) for an elastic case. In fact, if 𝑎 = ±ℓ, equation (3.19)

6𝑚 (1+𝜀)𝑣0
gives 𝜔 = ± 4𝑚+𝑀 , which is at most equal to the expression in equation (3.21). This
2ℓ

3𝑚 6𝑚 2
fact has been proven above, i.e. 𝑀+𝑚 − (4𝑚+𝑀 ) ≥ 0.

The necessary condition that needs to be fulfilled is |𝑎| ≤ ℓ or 𝑀 ≤ 2𝑚. When 𝑀 =

(1+𝜀)𝑣0
2𝑚, the local maxima occur at both ends of the rod and 𝜔max = ± . If 𝑀 > 2𝑚, no
2ℓ

local maxima could be found and the rod should be hit at one of its ends, regardless of the

elasticity of the collision.

It is clearly an alternative conception if one thinks that for the rod to attain maximum

angular speed, the projectile has to stop after the collision to give off its entire momentum or

energy, similar to the previous case. Substituting equation (3.20) to equations (3.17) and

𝑀(1−𝜀)+2𝑚 𝑣0 𝑚(1+𝜀) 𝑣0
(3.18), 𝑣 = and 𝑉 = . Interestingly, if the collision is elastic, 𝑣 = 𝑉 =
𝑀+𝑚 2 𝑀+𝑚 2

𝑚
𝑣0 , i.e. the projectile and rod’s centre of mass move with identical linear velocity after
𝑀+𝑚

the collision, similar to the case of a perfectly-inelastic collision in the absence of rotation. If
𝑣0
the collision is perfectly inelastic, 𝑉 is half of the elastic case, and 𝑣 = 𝑉 + = 𝑉 + 𝜔max 𝑎,
2

signifying that the projectile instantaneously sticks to the rod.

3.2.3. Third Generalisation: Existence of Pivot

Consider the case where 𝑣1 = 0, the collision is elastic, and the mass of the rod is

1
uniformly distributed, i.e. 𝑥 = 0, 𝐼CM = 3 𝑀ℓ2 . To generalise the problem, let the rod be

pivoted at a distance 𝑥 away (taken to be a non-negative value) from its centre (Figure 3.5).

The point of collision is at 𝑎 ∈ [−ℓ, ℓ].

67
Figure 3.5

Generalisation to Projectile-Rod Collision: Existence of Pivot

After the collision, the projectile will move horizontally – since there is only a horizontal

force acting on it during the short time of collision – and the rod will rotate about the pivot.

In this case, equation (3.1) is no longer valid as an external force from the pivot is

introduced. Equation (3.2) can only be applied by taking the pivot as the reference point, for

otherwise there will be a non-negligible angular impulse on the system acted by the pivot.

Equation (3.3) still holds as the external force from the pivot performs no (or negligible)

work. Using the sign convention above, applying equations (3.2) and (3.3)

1
𝑚𝑣0 (𝑎 − 𝑥 ) = 𝑚𝑣(𝑎 − 𝑥 ) + ( 𝑀ℓ2 + 𝑀𝑥 2 ) 𝜔
3

1 1 1 1
𝑚𝑣0 2 = 𝑚𝑣 2 + ( 𝑀ℓ2 + 𝑀𝑥 2 ) 𝜔2
2 2 2 3

yields

1
𝑚(𝑎 − 𝑥 )2 − (3 𝑀ℓ2 + 𝑀𝑥 2 )
𝑣= 𝑣0
1
𝑚(𝑎 − 𝑥 )2 + (3 𝑀ℓ2 + 𝑀𝑥 2 )

(3.22)

68
 2𝑚(𝑎 − 𝑥 )
𝜔= 𝑣0
1
𝑚(𝑎 − 𝑥 )2 + (3 𝑀ℓ2 + 𝑀𝑥 2 )

(3.23)

It is somewhat clear from equation (3.23) that the rod will not rotate if 𝑎 = 𝑥. If 𝑎 > 𝑥 then

the rod will rotate anticlockwise, and if 𝑎 < 𝑥 then the rod will rotate clockwise.

To maximise the angular speed of the rod, the first and second derivatives of 𝜔 with

respect to 𝑎 need to be explored.

1
𝑑𝜔 (3 𝑀ℓ2 + 𝑀𝑥 2 ) − 𝑚(𝑎 − 𝑥 )2
= 2𝑚𝑣0 2
𝑑𝑎 1
[𝑚(𝑎 − 𝑥 )2 + ( 𝑀ℓ2 + 𝑀𝑥 2 )]
3

𝑑2𝜔 2 (
𝑚 (𝑎 − 𝑥 )2 − (𝑀ℓ2 + 3𝑀𝑥 2 )
= 4𝑚 𝑣0 𝑎 − 𝑥 ) 3
𝑑𝑎2 1
[𝑚(𝑎 − 𝑥 )2 + ( 𝑀ℓ2 + 𝑀𝑥 2 )]
3

The first derivative goes to zero if

𝑀 2
𝑎 =𝑥±√ (ℓ + 3𝑥 2 )
3𝑚

(3.24)

𝑑2 𝜔 𝑀 𝑑2 𝜔
It is also obvious that < 0 and 𝜔 > 0 for 𝑎 = 𝑥 + √3𝑚 (ℓ2 + 3𝑥 2 ), > 0 and 𝜔 < 0
𝑑𝑎 2 𝑑𝑎 2

𝑀
for 𝑎 = 𝑥 − √3𝑚 (ℓ2 + 3𝑥 2 ), which essentially correspond to the local maxima of the

angular speed. Substituting equation (3.24) to equation (3.25), the result also shows that the

maximum angular speed will be attained if the projectile stops moving after the collision,

which is intuitive since all its kinetic energy is converted into the rod’s rotational kinetic

69
energy in this case. Similar to the previous two cases, interestingly these points are not the

rod’s ends.

The corresponding expressions for the maximum 𝜔’s are given by

𝑚
𝜔max = ±√ 𝑣0
1 2 + 𝑀𝑥 2
𝑀ℓ
3

(3.25)

Note that since −ℓ ≤ 𝑎 ≤ ℓ, for two local maxima to be found, it is necessary to have −ℓ ≤

𝑀 𝑀
𝑥 − √3𝑚 (ℓ2 + 3𝑥 2 ) and 𝑥 + √3𝑚 (ℓ2 + 3𝑥 2 ) ≤ ℓ. If only one of them is fulfilled then only

one local maximum exists, and if none of them are fulfilled then no local maxima can be

found.

The first inequality is equivalent to 𝑓(𝑥 ) = 3(𝑀 − 𝑚)𝑥 2 − 6𝑚ℓ𝑥 + (𝑀 − 3𝑚)ℓ2 ≤

0. Its discriminant is given by 12𝑀ℓ2 (4𝑚 − 𝑀), and the function passes through the origin

𝑀
at 𝑚 = 3 . Hence, noting that 0 ≤ 𝑥 ≤ ℓ, the analysis can be split up into five cases: 𝑚 ∈

𝑀 𝑀 𝑀 𝑀
[0, ), 𝑚 ∈ [ , ), 𝑚 ∈ [ , 𝑀), 𝑚 = 𝑀 and 𝑚 ∈ (𝑀, ∞). In the first three cases, the
4 4 3 3

quadratic function is convex; the fourth case yields to a linear function; and in the fifth case

the quadratic function is concave. In the first case, it is impossible to find any real 𝑥. The

3𝑚−√3𝑀(4𝑚−𝑀) 3𝑚+√3𝑀(4𝑚−𝑀)
second case corresponds to 𝑥 ∈ [ ℓ, ℓ]. The third case
3(𝑀−𝑚) 3(𝑀−𝑚)

corresponds to 𝑥 ∈ [0, ℓ]. The fourth case also corresponds to 𝑥 ∈ [0, ℓ]. In the fifth case, the

quadratic function will never have any zeros, hence it also corresponds to 𝑥 ∈ [0, ℓ].

The second inequality is equivalent to 𝑓(𝑥 ) = 3(𝑀 − 𝑚)𝑥 2 + 6𝑚ℓ𝑥 + (𝑀 −

3𝑚)ℓ2 ≤ 0. Its discriminant is given by 12𝑀ℓ2 (4𝑚 − 𝑀), and the function passes through

70
 𝑀
the origin at 𝑚 = 3 . Hence, noting that 0 ≤ 𝑥 ≤ ℓ, the analysis can be split up into five cases:

𝑀 𝑀 𝑀 𝑀
𝑚 ∈ [0, 4 ), 𝑚 ∈ [ 4 , 3 ), 𝑚 ∈ [ 3 , 𝑀), 𝑚 = 𝑀 and 𝑚 ∈ (𝑀, ∞). In the first three cases, the

quadratic function is convex; the fourth case yields to a linear function; and in the fifth case

the quadratic function is concave. In the first case, it is impossible to find any real 𝑥. In the

second case, it is impossible to find any non-negative 𝑥. The third case corresponds to 𝑥 ∈

−3𝑚+√3𝑀(4𝑚−𝑀) ℓ
[0, ℓ]. The fourth case corresponds to 𝑥 ∈ [0, 3]. In the fifth case, the
3(𝑀−𝑚)

3𝑚−√3𝑀(4𝑚−𝑀)
quadratic function is concave, hence it corresponds to 𝑥 ∈ [0, ℓ].
3(𝑚−𝑀)

Therefore, one local maximum can be found if either one of the following conditions

𝑀 𝑀 3𝑚−√3𝑀(4𝑚−𝑀) 3𝑚+√3𝑀(4𝑚−𝑀) 𝑀
is satisfied: 𝑚 ∈ [ 4 , 3 ) and 𝑥 ∈ [ ℓ, ℓ], or 𝑚 ∈ [ 3 , 𝑀) and
3(𝑀−𝑚) 3(𝑀−𝑚)

−3𝑚+√3𝑀(4𝑚−𝑀) ℓ
𝑥∈( ℓ, ℓ], or 𝑚=𝑀 and 𝑥 ∈ (3 , ℓ], or 𝑚 ∈ (𝑀, ∞) and 𝑥∈
3(𝑀−𝑚)

3𝑚−√3𝑀(4𝑚−𝑀)
( ℓ, ℓ]; and two local maxima can be found if either one of the following
3(𝑚−𝑀)

𝑀 −3𝑚+√3𝑀(4𝑚−𝑀) ℓ
conditions is satisfied: 𝑚 ∈ [ 3 , 𝑀) and 𝑥 ∈ [0, ℓ], or 𝑚 = 𝑀 and 𝑥 ∈ [0, 3],
3(𝑀−𝑚)

3𝑚−√3𝑀(4𝑚−𝑀)
or 𝑚 ∈ (𝑀, ∞) and 𝑥 ∈ [0, ℓ].
3(𝑚−𝑀)

In the previous two cases, one may have thought that the reason why the point of

interest is not one of the rod’s ends is due to the linear velocity the rod gains after the

collision. But the result obtained in (3.24) tells us that even when the centre of mass of the

rod is fixed (𝑥 = 0), the local maximum is attained not when the projectile hits one of the
𝑀
rod’s ends, except for the case where 𝑚 = 3 .

71
 𝑀 𝑀 𝑀 3𝑚−√3𝑀(4𝑚−𝑀)
What if 𝑚 ∈ [0, 4 ), or 𝑚 ∈ [ 4 , 3 ) and 𝑥 ∈ [0, ℓ)? Then it would be
3(𝑀−𝑚)

impossible to find any local maximum for the angular speed, and the global maxima should

be taken, i.e. 𝑎 = ±ℓ. The corresponding expressions for the maximum 𝜔’s are 𝜔max =

2𝑚(±ℓ−𝑥)
1 𝑣0 . Intuitively, letting the projectile hit the rod at its end that is farther
𝑚(±ℓ−𝑥)2+( 𝑀ℓ2 +𝑀𝑥 2)
3

from the pivot, 𝑎 = −ℓ, should contribute to a greater angular speed. To verify this,

ℓ−𝑥 ℓ+𝑥
−
1 1
𝑚(ℓ − 𝑥 )2 + (3 𝑀ℓ2 + 𝑀𝑥 2 ) 𝑚(ℓ + 𝑥 )2 + (3 𝑀ℓ2 + 𝑀𝑥 2 )

1
2𝑥 [(𝑀 + 𝑚)𝑥 2 + (3 𝑀 − 𝑚) ℓ2 ]
=−
1 1
[𝑚(ℓ − 𝑥 )2 + ( 𝑀ℓ2 + 𝑀𝑥 2 )] [𝑚(ℓ + 𝑥 )2 + ( 𝑀ℓ2 + 𝑀𝑥 2 )]
3 3

which is obviously negative unless 𝑥 = 0. If 𝑥 = 0, the points of interest are symmetrical

with respect to the rod, and hence both of them should correspond to equal angular speed.

From equation (3.22), it is obvious that the projectile will stop moving after the

1 𝑀
collision if 𝑚(𝑎 − 𝑥 )2 = (3 𝑀ℓ2 + 𝑀𝑥 2 ), or equivalently 𝑎 = 𝑥 ± √3𝑚 (ℓ2 + 3𝑥 2 ).

Comparing with equation (3.24), it tells us that the local maxima occur if the projectile stops

moving after the collision.

3.3. Chapter Summary

The generalisations presented in this chapter include the unevenness (non-uniformity)

of the rigid body, the elasticity of the collision, as well as the case where the rod is pivoted.

For the rod to attain the greatest angular speed after the collision, the projectile should hit it

72
at a certain point, which is in general not the rod’s end, contrary to common intuition. This

is true even if the rod is uniform, its centre of mass is pivoted, and the collision is elastic.

In the first generalisation, the rod (mass 𝑀, length 2ℓ, centre of mass at 𝑥 from its

centre, moment of inertia about its centre of mass 𝐼CM ) is assumed to be non-uniformly

distributed, and the centre of mass of the rod is not at its geometrical centre. Only elastic

collision is considered. To maximise the angular speed of the rod, the projectile (mass 𝑚)

should hit the rod perpendicularly at positions 𝑎 from the centre of the rod, where

(𝑚 + 𝑀)𝐼CM (𝑚 + 𝑀)𝐼CM (𝑚 + 𝑀)𝐼CM

𝑥±√ , if 𝑥 ∈ [0, ℓ − √ ],√ ∈ [0, ℓ)
𝑀𝑚 𝑀𝑚 𝑀𝑚
𝑎=
(𝑚 + 𝑀)𝐼CM (𝑚 + 𝑀)𝐼CM (𝑚 + 𝑀)𝐼CM
𝑥−√ , if 𝑥 ∈ [√ − ℓ, ℓ] , √ ∈ [ℓ, 2ℓ]
𝑀𝑚 𝑀𝑚 𝑀𝑚
{

The corresponding angular velocities are respectively given by

𝑀𝑚 (𝑚 + 𝑀)𝐼CM (𝑚 + 𝑀 )𝐼CM
±√ 𝑣 , if 𝑥 ∈ [0, ℓ − √ ],√ ∈ [0, ℓ)
(𝑚 + 𝑀)𝐼CM 0 𝑀𝑚 𝑀𝑚
𝜔max =
𝑀𝑚 (𝑚 + 𝑀 )𝐼CM (𝑚 + 𝑀 )𝐼CM
−√ 𝑣 , if 𝑥 ∈ [√ − ℓ, ℓ] , √ ∈ [ℓ, 2ℓ]
(𝑚 + 𝑀)𝐼CM 0 𝑀𝑚 𝑀𝑚
{

ωℓ 𝑎 (𝑚+𝑀)𝐼CM
Figure 3.6 illustrates the behaviour of 2𝑣 as a function of ℓ in this scenario, for √ =
0 𝑀𝑚

(𝑚+𝑀)𝐼CM
.8, 𝑥 = .1ℓ; and for √ = 1.3, 𝑥 = .7ℓ.
𝑀𝑚

73
Figure 3.6

𝝎𝓵 𝒂 (𝒎+𝑴)𝑰𝑪𝑴 (𝒎+𝑴)𝑰𝑪𝑴
Graph of 𝟐𝒗 vs. 𝓵 for: (a) √ = 𝟎. 𝟖𝓵, 𝒙 = 𝟎. 𝟏𝓵 ; (b) √ = 𝟏. 𝟑𝓵, 𝒙 =
𝟎 𝑴𝒎 𝑴𝒎

𝟎. 𝟕𝓵

(a)

74
 (b)

This first generalisation was used to construct and create our main study test instrument with

an exception that the mass of the rod is uniformly distributed. In this case, the centre of mass

of the rod is located at its geometrical centre, and the problem has been greatly simplified.

In the second generalisation, the rod is assumed to be uniformly distributed (mass 𝑀,

1
length 2ℓ, moment of inertia about its centre of mass 𝑀ℓ2 ). The collision is assumed to be
3

𝑎2
(𝑉−𝑣)+3𝑉 2
dissipative with 𝜀 = ℓ
∈ [0,1]. To maximise the angular speed, the projectile
𝑣0

(mass 𝑚) should hit the rod perpendicularly at positions 𝑎 from the centre of the rod, where

𝑀+𝑚
𝑎 = {±ℓ√ 3𝑚 , 𝑀 ≤ 2𝑚

±ℓ, 𝑀 > 2𝑚

The corresponding angular velocities are respectively given by

3𝑚 (1 + 𝜀 )𝑣0
±√ , 𝑀 ≤ 2𝑚
𝜔max = 𝑀+𝑚 2ℓ
3𝑚 (1 + 𝜀 )𝑣0
{± 4𝑚 + 𝑀 ℓ
, 𝑀 > 2𝑚

𝜔ℓ 𝑎 𝑀
Figure 3.7 illustrates the behaviour of 3(1+𝜀)𝑣 as a function of ℓ in this scenario, for 𝑚 = .47
0

𝑀
and 𝑚 = 5.75.

75
Figure 3.7

𝝎𝓵 𝒂 𝑴 𝑴
Graph of 𝟑(𝟏+𝜺)𝒗 vs. 𝓵 for: (a) 𝒎 =. 𝟒𝟕 ; (b) 𝒎 = 𝟓. 𝟕𝟓
𝟎

(a)

(b)

76
 In the third generalisation, the rod is assumed to be uniformly distributed (mass 𝑀,

1
length 2ℓ, moment of inertia about its centre of mass 3 𝑀ℓ2 ). The collision is assumed to be

elastic. A pivot is introduced in the system at a distance 𝑥 from the centre of the rod. To

maximise the angular speed of the rod, the projectile (mass 𝑚) should hit it perpendicularly

at positions 𝑎 from the centre of the rod, where

𝑀 𝑚
𝑎 = 𝑥 ± √3𝑚 (ℓ2 + 3𝑥 2 ) and 𝜔max = ±√1 𝑣0 if any of the followings is satisfied:
𝑀ℓ2 +𝑀𝑥 2
3

𝑀 −3𝑚 + √3𝑀(4𝑚 − 𝑀)
𝑚∈[ , 𝑀) and 𝑥 ∈ [0, ℓ]
3 3(𝑀 − 𝑚 )
ℓ
𝑚 = 𝑀 and 𝑥 ∈ [0, ]
3
3𝑚 − √3𝑀(4𝑚 − 𝑀)
𝑚 ∈ (𝑀, ∞) and 𝑥 ∈ [0, ℓ]
{ 3(𝑚 − 𝑀 )

𝑀 𝑚
𝑎 = 𝑥 − √3𝑚 (ℓ2 + 3𝑥 2 ) and 𝜔max = −√1 𝑣0 if any of the followings is satisfied:
𝑀ℓ2 +𝑀𝑥 2
3

𝑀 𝑀 3𝑚 − √3𝑀(4𝑚 − 𝑀) 3𝑚 + √3𝑀(4𝑚 − 𝑀)
𝑚∈[ , ) and 𝑥 ∈ [ ℓ, ℓ]
4 3 3(𝑀 − 𝑚 ) 3(𝑀 − 𝑚 )
𝑀 −3𝑚 + √3𝑀(4𝑚 − 𝑀)
𝑚∈[ , 𝑀) and 𝑥 ∈ ( ℓ, ℓ]
3 3 (𝑀 − 𝑚 )
ℓ
𝑚 = 𝑀 and 𝑥 ∈ ( , ℓ]
3
3𝑚 − √3𝑀(4𝑚 − 𝑀 )
𝑚 ∈ (𝑀, ∞) and 𝑥 ∈ ( ℓ, ℓ]
{ 3 (𝑚 − 𝑀 )

2𝑚(ℓ+𝑥)
𝑎 = −ℓ and 𝜔max = − 1 𝑣0 if any of the followings is satisfied:
𝑚(ℓ+𝑥)2 +( 𝑀ℓ2 +𝑀𝑥 2)
3

77
 𝑀
𝑚 ∈ [0,
)
4
𝑀 𝑀 3𝑚 − √3𝑀(4𝑚 − 𝑀)
𝑚 ∈ [ , ) and 𝑥 ∈ [0, ℓ)
{ 4 3 3(𝑀 − 𝑚 )

𝜔ℓ 𝑎
Figure 3.8 illustrates the behaviour of 2𝑣 as a function of ℓ in this scenario, for 𝑚 = 𝑀, 𝑥 =
0

.2ℓ; 𝑚 = .5𝑀, 𝑥 = .5ℓ; and 𝑚 = .1𝑀, 𝑥 = .5ℓ.

In general, for a projectile-rod collision, maximum angular speed of the rod is attained

not when the projectile hits the rod’s end. The underlying concepts behind the problem

presented are simple, but students may proceed with an inaccurate analysis as it comes to this

presented higher order thinking problem.

Figure 3.8

𝝎𝓵 𝒂
Graph of 𝟐𝒗 vs. 𝓵 for: (a) 𝒎 = 𝑴, 𝒙 =. 𝟐𝓵 ; (b) 𝒎 =. 𝟓𝑴, 𝒙 =. 𝟓𝓵 ; (c) 𝒎 =. 𝟏𝑴, 𝒙 =. 𝟓𝓵
𝟎

(a)

78
(b)

(c)

79
 PART II

EDUCATIONAL STUDIES ON

MATHEMATICAL THINKING IN PHYSICS

80
 CHAPTER 4

LITERATURE REVIEW

This first part of this chapter provides the literature review on conceptual

understanding and its characteristics, one of them being representational fluency. Also,

alternative conceptions and conceptual change will be elaborated.

The second part deals mostly on mathematical thinking – being the main idea

explored in this thesis study –, its use in physics, its application in exploring a few topics in

mechanics, as well as the relevant studies pertaining to topics and studies on high-performing

students.

The third part of the literature review elaborates on the formats of diagnostic

instruments used in educational research. As mentioned in Chapter 1, some diagnostic

instruments are readily available. In fact, these instruments might be good to determine the

degree of students’ conceptual understanding and to obtain the right pool of participants for

this study. However, these instruments might not be sufficient if they were to be used for the

main studies in this thesis as they would be unable to excerpt students’ mathematical thinking

and they are also short of higher order thinking questions, which are necessary for this study.

Lastly, the framework that will be used for our teaching intervention strategy will be

discussed in detail. This strategy is unique, as it does not focus on addressing alternative

conceptions – which is commonly done in science education studies –, but rather on

promoting the use of mathematical thinking in analysis of physical situations.

81
4.1. What is Conceptual Understanding?

Concepts and conceptual understanding are common terms that are extensively used

and discussed in the field of science education. They are always thought-provoking to

educators and educational researchers. Concepts can be described as packages of meaning

that capture regularities (similarities and differences), patterns, or relationships among

objects, events, and other concepts that oftentimes represent different ways of organising the

nature of the world (West & Pines, 1985). Konicek-Moran and Keeley (2015) defined

conceptual understanding as one’s ability to make a science concept their own such that they

can think with it, apply it in broader areas, state it in their own words, find a metaphor or an

analogy for it, or build a mental or physical model of it. Aligned with this definition, one

assessable characteristic of conceptual understanding is its “transferability” i.e. the ability to

broadly apply knowledge in various contexts – not necessarily in the context in which it was

attained (McDermott, 2001; Mestre, 2002). In that sense, students who have gained

conceptual understanding should be able to connect and apply pieces of knowledge that were

acquired through their learning process. For example, many students try to solve as many

past year examination questions, hoping that in the examination, a question in a related

context is set. But when a new context which they have not encountered before is set for the

question, they may be caught unawares. Thus, the transferability aspect is a key aspect to test

for students’ understanding – some even describe it as the ultimate goal of education (Lupart,

1995). Although there has been much debate among researchers regarding factors that affect

the occurrence of transfer (Rebello et al., 2004; Barnett & Ceci, 2002), the importance of

transfer cannot be overstated – real knowledge and real learning must enable one to apply

the concept learnt outside the original learning context, for otherwise, they are very limited

82
in usefulness.

Researchers adopt varied views regarding the degree of coherence or structure of an

individual’s conceptions. DiSessa (1993) adopted a knowledge-in-pieces view of an

individual’s conceptions. He argued that these conceptions are formed by disjointed snippets

of ideas, called “phenomenological primitives”, “p-prims” or “knowledge in pieces” (p. 111).

These p-prims are phenomenological because they are formed by individuals to interpret and

understand their experiences in the physical world, and primitive because they are self-

evident and require no explanations. P-prims are descriptive, independent of each other and

able to evolve from previous knowledge. Hammer (2000) also subscribed to the same theory.

He hypothesised the existence of “conceptual resources”, which are cognitive structures

similar to, yet bigger than, p-prims. Some examples of p-prims and/or conceptual resources

on the topic of mechanics are as follows:

“Heavier (or bigger) body is harder to rotate” – it requires more effort to rotate a

heavier (or bigger) body.

“Maximum transfer (or transformation) implies a-hundred-percent transfer (or

transformation)” – whenever there is a transfer (or transformation) of physical

quantity (linear momentum, angular momentum or energy), maximum transfer occurs

when the quantity is fully transferred (or transformed).

“Longer arm length implies faster rotation” – longer arm length means easier to

rotate, and hence, faster rotation.

83
 DiSessa subscribed to a continuum view of conceptual change in that an individual’s

alternative conceptions, which are generated by one or more p-prims, can evolve into

scientifically-accepted views espoused by experts through some fine-tuning. He wrote that

“many misconceptions (alternative conceptions) come simply from using an element outside

its range of legitimate applicability” (p. 116). In other words, these conceptions may have a

certain extent of scientific truth in them, but there are limitations in their application, which

need to be addressed in order for students to understand they are actually alternative

conceptions in the context under discussion. In contrast, Hammer noted that alternative

conceptions are due to fervent misapplications of conceptual resources in particular contexts

or situations. Similar to DiSessa’s view, this means that alternative conceptions may involve

correct conceptual resources that are associated with wrong terms (Grayson, 1996).

DiSessa’s view of individual’s conceptions is supported by Mazens & Lautrey

(2003). In their study, it was found that young children were concerned about “local

coherency” but not “global coherency”. This means that these young children tended to be

satisfied with concepts that are applicable to a specific situation, even though they contradict

their reasoning in other situations. They were not aware of the importance of having a

generalised scientific concept that is applicable to any situation. Supporting this view, Reif

& Allen (1992) highlighted that students’ conception lacked coherence and appeared to be

composed of weakly linked knowledge elements. Due to this lack of coherence in the

students’ conceptions, these students would likely encounter paradoxes which they may find

difficult to resolve. It was noted that their participants exhibited difficulties in making general

inferences and tended to conclude that an occurrence is impossible simply because they could

not come up with a situation in which the event would be possible.

84
 It is notable that in science and science education, very often the words “concept” or

“conceptual understanding” is associated with qualitative thinking that does not invoke

numbers, mathematical notations or formulas. When referring to “conceptual questions” in a

test, for instance, many science teachers and students would think of questions that are

expressed in words and do not contain mathematics in it. However, in the subject of Physics,

this is far from the truth, as mathematics is often associated with and used in the description

of the physical situation being analysed.

Teaching experiences have allowed us to see how science concept representations

vary across students. The National Research Council’s report, Taking Science to School:

Learning and Teaching Science in Grades K-8, in the United States, describes student

proficiency in science to include learners who “know, use, and interpret scientific

explanations of the natural world” and “generate and evaluate scientific evidence and

explanations” (National Research Council [NRC], 2007). Proficient students have deep

understanding (Sawyer, 2014) that allows them to solve problems and use their knowledge

to explain the world (National Research Council [NRC], 2012).

There has been immense effort in past science education researches to focus on how

to improve students’ conceptual understanding in many different topics through various

means or interventions (Venville & Dawson, 2010; Olympiou & Zacharia, 2012; Gobert &

Clement, 1999; Zacharia, 2007). Repeatedly, science teachers across the world find it

difficult to get students to be “fluent” in applying facts, vocabularies and concepts they seem

to have mastered in class – showing a deficit of representational fluency. Though many

students are skillful in memorising the facts and necessary procedures, they are lacking in

building deep scientific arguments to solve science problems – a real and critical issue that

85
needs to be addressed (National Research Council, 2012; Braaten & Windschitl, 2011;

Applebee, 1986; Carey, 1986; Anderson & Smith, 1984; Champagne & Klopfer, 1977,

1984). This is especially observed when test questions cannot be solved by simple

memorisation of facts, formulas or methods, but rather require multiple layers of thinking.

4.2. Representational Fluency

Representational fluency is often defined in the physics discipline as the ability to

work within and translate among representations in which information can be presented with

ease (Bieda & Nathan, 2009; Hill & Sharma, 2015). This refers to the many ways that

information can be presented. Some examples of representations are the spoken or written

word, symbols, equations and images. The use of appropriate representations is important

because they can facilitate memorising (Aldrich & Sheppard, 2000), overcome cognitive load

limitations (Ainsworth, 2006), and portray relationships in a more obvious manner (Bowen

et al., 1999; Goldman, 2003). Moreover, the construction of representations has also been

linked with successes in learning science (Prain & Tytler, 2012).

Physics, like other sciences, uses various means to present information and ideas,

including diagrams, graphs, words and mathematics. Oftentimes, the same information can

be depicted through various representations (Figure 4.1). Each representation has particular

ways that it is more helpful, or less helpful in different situations, depending on the content

and purpose of the representation. As such, it is not surprising that multiple representations

are utilised in a physics class, assuming they can decipher the language used. Physics students

will require familiarity with individual representations and their affordances in order to

understand material as it is presented to them. In addition, they must develop the skills or

86
regulative techniques to be selective in choosing the combinations of representations they

use in order to achieve the best outcome for their own physics understanding, problem

solving, and communication.

Figure 4.1

Illustration of Multiple Representations Used in Physics to Depict Real World Situation

(Redish, 2003)

Learning to use representations in physics can be seen as similar to learning a new

language (Hill, 2015). Being fluent in a language is beyond being able to recognise and

identify words and their meanings. Fluency in a certain language is the ability to converse

fluidly and easily in that language. If mathematics is indeed one of the languages of physics,

one must become representationally fluent before using it to represent a physics concept,

otherwise physics will always seem like a foreign language. Importantly, someone who is

representationally fluent will begin to think in the language of representations, i.e. learn

87
physics and solve physics problems using mathematical thinking through mathematical

representations.

4.3. Alternative Conceptions

The word alternative conceptions will be used in this thesis rather than

misconceptions. An alternative conception is defined as students’ conception that differs

from, or is not consistent with those commonly accepted by the scientific community (Sanger

& Greenbowe, 2000). Alternative conceptions are usually persistent and pose strong

influence in the learning of new ideas (Carmichael et al., 1991; Pfundt & Duit, 1991). It is

notable that these alternative conceptions are largely constructed from the learner’s

experience (Wandersee et al., 1994). Students come to class with a repertoire of knowledge

and conceptual frameworks of scientific concepts, that is, prior knowledge (Teichert & Stacy,

2002). In this manner, students’ pre-existing inaccurate, superficial and flawed prior

knowledge could precipitate the formation of new alternative conceptions during the learning

of new knowledge (Pintrich et al., 1993; Dole, 2000). When a new phenomenon that demands

a thorough scientific analysis arises, students tend to use this flawed understanding in their

analysis, which has become a scientific intuition to them.

Wandersee et al. (1994) elaborates on a number of sources that could contribute to

the formation of alternative conceptions. These sources include language use (which covers

terminology), prior knowledge, teachers’ instruction as well as instructional materials such

as textbooks. Wenning (2008) supports these findings and noted that their identification

could be difficult, which include misunderstanding, miscommunication, miseducation and

misapplication of correct principles. Nevertheless, it is crucial for students’ alternative

88
conceptions to be identified and understood by educators so that these can be addressed

specifically and effectively in class.

As mentioned, language, being the key element for communications (including

scientific ones), is one of the main sources of alternative conceptions. Several researchers

(Clerk & Rutherford, 2000; Johnstone & Selepeng, 2001) found that language does pose

problems in the learning of science. Students, when interpreting scientific terms with

everyday meanings, could get an incomplete or totally different idea with what is being

communicated in the scientific text. When we say A is heavier than B, for example, it could

mean that A has larger mass than B or that the gravitational pull on A is stronger than that on

B. The language that is used in the science class would greatly affect the way students

perceive the idea of the scientific concept being taught. In this regard, it is crucial to have a

common language to communicate scientific ideas that is robust towards misinterpretations

and contains an absolute truth in itself, especially in the subject of Physics.

4.4. Conceptual Change

Conceptual change is defined as a process where existing conceptions are reorganised

(Tyson et al., 1997) or replaced so as to accommodate new ideas (Smith et al., 1993), which

are scientifically acceptable in students’ conceptual framework (Tekkaya, 2003). Previous

studies have shown that conceptual change is effective in addressing students’ alternative

conceptions (Basili & Sanford, 1991; Ebenezer & Gaskell, 1995). The issue here is that

students may not feel a need to change any of their existing conceptions if they can solve the

assessment problems given to them (Posner et al., 1982). The students must perceive a

cognitive dissonance, i.e. disagreement and dissatisfaction with one’s existing conception,

89
before they would abandon the conception and replace it with the correct one. This process

of abandonment of preconception and replacement of conception is what constitutes

conceptual change (Hewson, 1981).

In the past, a number of researchers have proposed theoretical explanations for what

occurs in the conceptual change process (Vosniadou, 1994; Chi et al., 1994; DiSessa, 1993;

Ueno, 1993). They described that learners’ ontological (Chi et al., 1994) and epistemological

presuppositions (Vosniadou, 1994), their “self-explanatory” everyday experiences (DiSessa,

1993) and the context (Ueno, 1993) play significant roles in the development of alternative

conceptions. In other words, conceptual change takes place only when learners could

compare, contrast, recognise, integrate and evaluate their existing conception and new ideas,

everyday experiences and contextual factors. For these processes to take place, one should

monitor the consistency between his/her existing concepts – that he/she is aware of – and

external information as well as evaluate the new concepts by providing justifications. Pintrich

and Sinatra (2003) stated that the theoretical models proposed to explain the change in

students’ conceptions make an “assumption about the importance of metacognitive

awareness” (metacognition is a superset of awareness, monitoring and evaluation). Their

view is supported by many other researchers (Georghiades, 2004; Ferrari & Elik, 2003;

Hennessey, 1999; Vosniadou, 1994, 2003; Beeth, 1998; White & Gunstone, 1989).

Metacognition is a process by which an individual is aware of his or her own brain

processes that occur during learning. In other words, metacognition is thinking about

thinking. A person with metacognitive skills is aware of his/her cognition and he/she is able

to evaluate and monitor his/her own thinking process. Metacognitive processes in facilitating

the change in students’ existing conceptions have been viewed as an important factor by

90
researchers in recent years (Georghiades, 2004; Vosniadou, 2003). These researchers

believed that educators should aim to implement teaching interventions that activated

students’ metaconceptual processes. In science education research, Posner et al. (1982)

proposed one of the most popular theoretical frameworks for conceptual change, the

Conceptual Change Model. This model describes the required conditions for one to change

his/her ideas and for his/her environment (components of one’s conceptual ecology).

Research studies conducted in the area of cognitive psychology have also proposed

theoretical frameworks about the nature of the change in students’ conceptions (Vosniadou,

1999).

Many researchers in recent years used the term “metaconceptual” to refer to the meta-

level thinking processes that are acting on students’ conceptions (Yürük, 2007; Mason &

Boscolo, 2000; Vosniadou, 1994, 2002, 2003; Wiser & Amin, 2001). Metaconceptual

knowledge and processes can be classified into four major components:

(a) metaconceptual knowledge refers to one’s acquired stable and statable knowledge

about concept learning and the factors affecting one’s concept development,

(b) metaconceptual awareness is one’s awareness of and reflection on existing concepts

and elements of conceptual ecology including one’s interpretation of experiences,

ontological and epistemological presuppositions,

(c) metaconceptual monitoring involves control processes which generate information

about one’s cognitive state and thinking processes, and

(d) metaconceptual evaluation involves processes in which learners make judgmental

decisions about the relative ability of the competing conceptions to explain the real

phenomenon.

91
The last two processes occur during one’s attempts to learn a new conception. Examples of

these processes includes justifying the truth contained in the new conceptions, contrasting

the existing and new conceptions, as well as monitoring the relevance and validity of existing

and new conceptions.

In the Conceptual Change Model (Posner et al., 1982), there are four essential

conditions for conceptual change to occur in students: (1) dissatisfaction, (2) intelligibility,

(3) plausibility and (4) fruitfulness. As mentioned above, the feeling of dissatisfaction with

the student’s existing conception becomes the motivation for conceptual change. Simply put,

an appropriate assessment must be constructed so as to surface the student’s alternative

conception in order for conceptual change to take place. A non-discriminating assessment is

definitely ineffective for this purpose. Intelligibility of a scientific concept simply means the

concept must make sense to the student and it can be constructed from other fundamental

existing concepts that have been previously learnt. To be plausible, the scientific conception

must be a credible and acceptable alternative. Finally, fruitfulness refers to the ability of the

new conception to produce acceptable solutions for the given problems.

In order for conceptual change to occur, the classroom environment is believed to be

one of the crucial parts in lesson intervention. Vosniadou (2007) noted that for conceptual

change to occur with students, classroom environment that provides opportunity for the

students to exchange ideas is necessary. In this regard, group discussions are seen as a

preferred element in teaching intervention.

One issue with the use of conceptual change strategies in this study is that alternative

conceptions are not being specifically addressed. Studies on conceptual change in science

have focused on specifically addressing alternative conceptions so that they become primed

92
towards canonical tenets. The teaching intervention in this study focuses more on promoting

mathematical thinking. If in the process, it leads to remediation of alternative conceptions,

then it is a welcome change. However, it is not possible to know of this before the start of

the intervention. In view of the foregoing, there is very little guidance in the literature on how

to configure a teaching intervention which leads to changes in mathematical thinking among

students from pre-test to post-test.

4.5. Mathematical Thinking in Physics

To address the problem of language as one of the main sources of alternative

conceptions, mathematics is found to be a perfect candidate to be adopted as a common

language for science communication (Aghadiuno, 1992). When a scientific idea is

communicated in mathematical language, any ambiguities are prevented and its robustness

towards misinterpretation can be minimised. Consider explaining a simple concept of linear

momentum and force in a physics class. To explain why a system’s momentum is conserved,

teachers may say “When there is no external force acting on a system, its linear momentum

𝑑𝐩
⃗⃗
must not change.” or they may say “From Newton’s second law, ∑ 𝐅⃗ext = . Since ∑ 𝐅⃗ext =
𝑑𝑡

⃗⃗, we have 𝑑𝐩⃗⃗ = 𝟎

𝟎 ⃗⃗ and therefore 𝐩
⃗⃗ is a constant vector.”. Technically, the two explanations
𝑑𝑡

may sound identical and there is no superiority in either one of them; they indeed promote

the same level of conceptual understanding. However, using mathematics as a common

language in explaining the concept of conservation of momentum provides benefits to

students, especially if they are mathematically inclined, in line with the aspect of

transferability. Firstly, consider the interpretation of the concept. What does it mean by “there

93
is no external force”? Does that mean that the system doesn’t experience any forces at all?

From mathematics, we understand that even though there may be some external forces acting

⃗⃗), momentum will still be conserved.

on the system, as long as the net force is zero (∑ 𝐅⃗ext = 𝟎

In this manner, mathematical language is strongly reliable as it provides a self-scrutiny that

guarantees an absolute truth. Secondly, using mathematical language stimulates students’

mathematical thinking in a way that they are able to see mathematically why linear

momentum is indeed conserved.

Mathematics has been one of the most basic subjects, introduced in the early stage of

education and believed to be necessary to develop other concepts, especially scientific

concepts. Some people would consider the fields of mathematics and science to be

overlapping, in which mathematics plays a role as the science that draws necessary scientific

conclusions (Peirce, 1881). With regards to this understanding, mathematics cannot be

considered as the originator of scientific laws – as scientific laws are usually discovered by

induction – nor the framer of theories – as scientific theories are usually hypothesised.

Instead, mathematics can be considered as one of the thinking tools in which science can be

communicated. The role of mathematics as a thinking tool in science is deemed to be

important and even fundamental as neither can scientific laws rule the universe we live in,

nor can scientific theory explain natural phenomena without the sanction of this tool. This

concurs with the words of the late scientist, Galileo Galilei (1564-1642), who once said “The

universe cannot be read until we have learnt the language and become familiar with the

characters in which it is written. It is written in mathematical language, and the letters are

triangles, circles and other geometrical figures, without which means it is humanly

94
impossible to comprehend a single word. Without these, one is wandering about in a dark

labyrinth.”, cited in Sautoy (2012). The above sentence highlights the significant role of

mathematics in the learning of science, in explaining and analysing science concepts,

especially when deeper analysis is required. Borrowing the famous quote from Immanuel

Kant (1724-1804), “In any particular theory there is only as much real science as there is

mathematics.”

Indeed, previous studies have shown that many students in several countries

experience great difficulties when they are asked to use previously learnt mathematical

concepts in problem-solving that involve selecting and using proper mathematical resources,

representations, and strategies to identify or construct mathematical relations (Ginsburg et

al., 2005). This is due to their lack of problem-solving competencies as they have not

developed “a way of thinking” to formulate questions, make conjectures, use various

representations, identify and explore mathematical relations, look for arguments to support

them, and communicate their results (Santos-Trigo & Camacho-Machín, 2009). Put simply,

although students can do mathematics, they may still not be able to perform mathematical

thinking.

Holyoak and Morrison (2005) describes a distinction between “thinking” and

“reasoning”. “Thinking” is a cognitive process which describes the systematic transformation

of mental representations of knowledge to characterise actual or possible states of the world.

Often, “thinking” is done in the service of goals. In other words, “thinking” is performed in

order for one to take or make an action. Although “thinking” and “reasoning” are both mental

processes, “thinking” encapsulates a larger arena of thought production that can be either

conscious or unconscious. A person would consciously or unconsciously “think” before

95
taking or making an action. In comparison, “reasoning” involves a deeper mental process –

it is scrutinising thoughts (products of thinking, ideas and beliefs) that one has thought about,

with facts or data, logic, etc. It is limited to the conscious production of mental thought with

the use of logic. The habit of thinking deeply needs to be developed before one can perform

reasoning in any specific context. As such, in this thesis the phrase “mathematical thinking”

is preferred rather than “mathematical reasoning”.

As this thesis explores the use of mathematical thinking in physics, the definition of

mathematical thinking has to come across clearly. Firstly, the phrase “mathematical

thinking” has a broader definition than quantitative thinking or quantitative reasoning. The

American Association of Colleges and Universities (AACU, 2010) defines quantitative

reasoning as a competency and comfort in working with numerical data and applying such

skills to solving quantitative problems within a range of contexts and everyday life situations.

Unfortunately, the definitions of “mathematical thinking” vary across different fields and

educational levels (Lutfiyya, 1998). Moreover, educators developed the definition of

mathematical thinking based on their own country’s curriculum (Isoda, 2006), resulting in

the diversity of different definitions of the term “mathematical thinking”. For example, Isoda

(2006) defines mathematical thinking in relation to problem-solving. He stated that

mathematical thinking requires higher thinking skills with a blend of mathematics to

understand the ideas, to identify and prove the relationship between the differences in ideas

and connect these differences together to solve an existing problem. Karadag (2009) defines

mathematical thinking as a thinking style supported by thinking skills. Among all the

thinking skills, he categorised mathematical thinking into seven major areas, which are

modelling, reasoning, symbolisation, representation, proving, abstraction, and

96
mathematisation. However, these major cognitive activities are hard to be categorised due to

their similarity and interrelation. Meanwhile, more recent literature argues that mathematical

thinking is a cognitive activity using mathematical techniques, concepts, and methods,

directly or indirectly (Uyangör, 2019). Although his definition still touches mostly the aspect

of problem-solving, there are many other aspects of mathematical thinking that can be

explored, such as conceptual learning, exploration of new methods, etc. This thesis study

adopts the definition of “mathematical thinking” from Sam and Yong (2006) due to its

compact and more comprehensive definition that can be applied to any aspects or fields; it is

defined as a mental operation used in solving problems affirmed by the mathematical

knowledge and dispositions, a definition that also includes attitude as an element in the

cognitive process. Mathematical thinking allows one to explore a physical situation and

perform a deep exploration resulting in various situations that require multiple layers of

analyses. As mentioned above, the emphasis on this skill is missing in most pedagogies

adopted in physics class.

Mathematical thinking is not only limited to thinking that involves numbers or

numerical data, but it may also include reasoning that involves variables, mathematical terms

and vocabularies, as compared to qualitative thinking. In Mathematics, many students are

usually able to perform well in their examinations and yet, when they go to a higher level of

study, surprise the teachers by their lack of facility in thinking mathematically. Researchers

reason that complete understanding should include the capacity to engage in the process of

mathematical thinking, in essence doing what makers and users of mathematics do: framing

and solving problems, looking for patterns, making conjectures, examining constraints,

making inferences from data, abstracting, inventing, explaining, justifying, challenging, and

97
so on (Pei et al., 2018; Devlin, 2000; Schoenfeld, 1992; Kitcher, 1984). This is on a much

higher level than students’ capacity to do mathematics or solve mathematics problems. In

this sense, mathematics should not be viewed as something static, but instead a dynamic

process of gathering, discovering and even creating new knowledge (Bukova, 2006;

Romberg, 1992).

In the context of science, specifically physics, mathematical thinking is crucial as a

complement to scientific intuition. Studies by Kaiser et al. (1986), Lemos (2008), Ricardo

and Lee (2015) and Ricardo (2016) show that in some cases, scientific intuition can be an

easy prey to deception, even when it is applied to simple physics problems. For example,

Ricardo (2016) addresses the question “Where should the rod be hit by a translating object

in order to attain maximum angular speed?” in a simple rigid body collision (Figure 4.2).

The students’ everyday experience with doors and levers will lead them to “think” that the

point should be at the rod’s end. Unfortunately, this is a common alternative conception in

physics. On the contrary, mathematics is something that one can reliably and correctly use at

all times (Schoenfeld, 2009), as compared to intuition. To overcome the abovementioned

alternative conception, for instance, a deep reasoning, which involves scrutinising

mathematical thinking at each stage, needs to be invoked. Such a problem that requires

multiple layers of thinking is categorised as a higher order thinking question. Another

example is on the idea of cycloid, especially the brachistochrone curve. The concept of

cycloid being a path that yields an oscillation with amplitude-independent period can always

be explained qualitatively using an analogy to simple harmonic motion. However, this is a

superficial explanation of cycloid. Several stages of mathematical thinking are crucial to be

98
performed if the concept were to be understood thoroughly. As such, teaching it to the

students with inability to perform mathematical thinking would be impossible.

Figure 4.2

Rigid Body Collision

Traditionally, it is believed that there is a significant relationship between

mathematical skills and physics achievement (Delialioğlu & Aşkar, 1999). The observations

on the relationship between mathematical skills and physics achievement have been shown

by many researchers (Long & Jiar, 2014; Peltzer, 1988; Hudson & McIntire, 1977;

Champagne et al., 1980; Hudson & Rottmann, 1981; Liberman & Hudson, 1979). Some other

studies found that mathematical skills can serve as a predictor for the physics courses

(Hudson & Liberman, 1982; Griffith, 1985). Although many science teachers have reported

that they see their students struggle with the use of mathematical terms and vocabularies to

reason, analyse, represent and communicate science problems, despite having taken

99
mathematics courses, the abovementioned studies show that this is not a valid reason to

downplay mathematical thinking in the teaching of physics concepts. In fact, it would be

interesting to see how concept representations vary and whether this relates to the degree of

their mathematical thinking skill.

When mathematical thinking is promoted in physics lessons and/or assignments,

there are a number of benefits. First and foremost, it helps students to bridge connections

across two subjects, pure Mathematics and Physics. Many high-performing students are

scarcely exposed to applying mathematical thinking in subjects other than Mathematics itself.

In fact, most students cannot find the beauty of articulating mathematical thinking until they

start performing it in the sciences. Moreover, the ability to transfer mathematics skills into a

chosen science discipline – in this case, Physics – is of crucial importance in students’

development as scientists, and in their future careers (Roberts et al., 2007). Unfortunately on

the contrary, many physics teachers assume that avoiding mathematics will keep their

students interested in the topic. Secondly, the habit of performing mathematical thinking in

learning basic physics concepts will translate to one’s ability in performing mathematical

thinking, where each stage of thinking is scrutinised and, in turn, approaching higher order

thinking questions with in-depth analysis. This is, in fact, beyond conceptual understanding

itself, leading to mastery of conceptual applications, where students are able to solve a wide-

range of problems. Problem-solving at a higher order thinking level is promoted in this thesis

study as students cannot leverage on ‘past experiences’ to solve the new problem posed,

allowing us to see the extent of transferability aspect (Reece, 2005; Darmer, 1995).

100
4.6. Relevant Studies in Physics Education on Momentum/Collision and Rotational

Mechanics

As this study focuses on high school students, a few topics in high school physics

were explored. Since the study involves higher order thinking problems, it was to our interest

to see the theoretical explorations on the topics of momentum/collision and rotational

mechanics involving several layers of mathematical thinking, as explored in Chapters 2 and

3. Nevertheless, prior to constructing new instruments, we sought a few studies in the

literature on these topics. A few alternative conceptions harboured by students and the

approaches to overcome these were found in the literature.

Table 4.1 describes the relevant studies in physics education on the topic of

momentum/collision.

Table 4.1

Studies on Students’ Alternative Conceptions and Conceptual Change on the Topic of

Momentum/Collision

Conceptual Change
Alternative Conceptions Level
Strategy

Samsudin et al. (2015) did not list down the alternative

High School Computer simulation
conceptions found

Şekercıoğlu & Kocakülah (2008)

• Bigger mass has bigger momentum (impulse)
• Momentum is pusher force
• It is harder to stop an object which has a bigger velocity, Conceptual change
mass is not important High School
strategy is not specified
• It is harder to stop a car which has bigger moment
• It is harder to stop a car which has a bigger momentum
• Momentum related to mass and velocity, and equal with
impulse

101
• If an object has a velocity, it has a force because of its
velocity
• A stationary object has a force and this force is not lost after
collision
• If soft objects collide each other, momentum does not
conserve
• For the momentum conservation, objects must collide
elastically
• If objects shape is spoilt momentum conserves, if not
momentum does not conserve
• Ball and astronaut have not got momentum because of have
not got any masses (weigh)
• Momentum is equal to force × distance
• If velocity’s doesn’t change, momentum doesn’t change
either
• Impulse means applying a force to an object which cannot
react to the power exerted on itself
• Impulse is momentum
• Impulse is applying a force to an object and speeding up of
that object as a result of the force applied or changing its
position
• Impulse changes objects’ position with applying force
• Impulse is a force which transferred energy from one form
to another form
• If an object changes its position there is impulse, if not there
is not impulse

Dalaklioğlu et al. (2015)

• Bigger mass has bigger momentum
• In an inelastic collision, the energy is fully transferred from Conceptual change
High School
one object to another strategy is not specified
• In a collision, momentum and energy are conserved for
each object

Hein (1999)
• The time in the impulse-force relationship represents the Undergraduate Writing (folder activity)
time taken for object to stop

REACT strategy
Ültay (2012) (Relating,
• Only pushing is considered as an impulse, and not pulling Undergraduate Experiencing,
• Impulse equals to momentum Applying, Cooperating
and Transferring)

Helm (1980) Conceptual change

Undergraduate
• Momentum is only conserved in an elastic collision strategy is not specified

Dega & Govender (2016)

• Elastic collision is always associated with “bouncing off” Conceptual change
Undergraduate
• ‘Moving faster and having an acceleration’ result in a larger strategy is not specified
momentum

102
 Problem-Based
Sahin (2010) did not list down the alternative conceptions found
Learning (PBL)

George et al., (2013) did not list down the alternative Microcomputer-Based
Undergraduate
conceptions found Laboratories (MBL)

Hewson & Hewson (1984) did not list down the alternative
Not specified Conceptual Conflict
conceptions found

In addition, there were also test instruments available as well as theoretical explorations that

can be used to construct test instruments on the topic of momentum/collision, such as those

by Ingec (2008), Chapman (1960), Diyanahesa et al. (2017), etc. These test instruments

mostly focused on high school students, and the list is certainly not exhaustive.

Unfortunately, most of the alternative conceptions in Table 4.1 and the theoretical

explorations in the previous studies mostly addressed basic physics concepts on

momentum/collision. They are not suitable for this thesis study. A comprehensive instrument

on this topic that is able to excerpt students’ alternative conceptions and mathematical

thinking concurrently is required.

Similarly, on the topic of rotational mechanics, the list of alternative conceptions

found in the literature is listed in Table 4.2. The topic of rotational mechanics is deemed to

be of higher level and as such researches in relation to this topic are very limited. Most of

these alternative conceptions only concern basic concepts on rotational mechanics. Although

there are some available conceptual surveys on this topic (Mashood & Singh, 2015;

Rimoldini & Singh, 2005), most of them deal only with basic concepts of rotational

mechanics. No available instruments were found in the literature on the topic of rotational

mechanics that explore higher order thinking and the use of the mathematical approach at the

same time.

103
Table 4.2

Studies on Students’ Alternative Conceptions and Conceptual Change on the Topic of

Rotational Mechanics

Conceptual Change
Alternative Conceptions Level
Strategy

Duman et al. (2015)

• Constant torque forms constant angular velocity
• Constant torque responsible for rotational balance
• Constant torque forms constant angular momentum
• The larger the mass of a wheel is, the greater the rotational
energy is
• The lighter wheel has more rotational kinetic energy…
Conceptual change
because it’s moving faster Undergraduate
strategy is not specified
• Moment of inertia depends on rollers’ angular acceleration
• Moment of inertia does not depend on rollers’ mass
• The instantaneous velocity with respect to the ground is
always tangent to the rolling circle
• The speed of all points should be the same with respect to
ground because they are all on the same wheel which is
rolling
Carvalho & Sousa (2005)
• Frictional forces acting on a rigid body always oppose the
motion such that it can never cause the motion itself Conceptual change
High School
• Analysis of rigid body is the same as analysis of point mass strategy is not specified
• Bicycle moves because there is a kind of driving force
produced by the engine (or the legs)

4.7. Relevant Studies in Physics Education Involving High-Performing Students

Excerpts of students’ mathematical thinking would be more visible if higher order

thinking is involved, as the multiple layers of thinking would provoke deeper use of

mathematical thinking in approaching a physics question. However, the involvement of

higher order thinking questions requires the subject participants to be high-performing

104
students who can tackle complex physics concepts easily. Therefore, this thesis study is

mainly targeted for high-performing students in physics.

Unfortunately, very few studies focusing on high-performing students were found in

the literature (Durkee, 1974; Coleman & Shore, 1991; Ziegler et al., 1996), and most of these

studies only explored students’ cognitive processes, gender differences, correlations between

their physics ability and other abilities, and so on. None of these studies looked into the

students’ mathematical thinking aspect as they approach physics problems in general, which

is the main crux of our study. Furthermore, none of these studies explored conceptual change

strategy for the improvement of their performance in higher order thinking problems.

Generally in educational research, the emphasis is more on mid-performing and low-

performing students. This is understandable as these groups of students need more support

for levelling up their conceptual understanding and their performance in problem-solving.

High-performing students are already at the top and they are unlikely to need further support

from stakeholders or study by researchers.

Many test instruments that were able to excerpt students’ alternative conceptions on

these two topics have been explored (refer to Tables 4.1 and 4.2). However, to the best of our

knowledge, there were no available instruments in the literature on the topic of

momentum/collision targeted specifically on high-performing students that were

comprehensive enough to excerpt students’ conceptual understanding, alternative

conceptions and mathematical thinking concurrently. As a solution to the shortcoming in the

literature, the theoretical explorations in Chapters 2 and 3 seem suitable for our study. The

analysis on alternative conceptions in those chapters was seen to be appropriate for high-

performing students.

105
4.8. Formats of Diagnostic Instruments for Use in Assessing Students’ Understanding

and Identifying Alternative Conceptions

Assessment is a powerful educational tool that serves at least four important

functions. First, it helps to evaluate students’ learning. This is a desired information not only

as a feedback for teachers but also the students themselves. Second, it helps students structure

their academic efforts. Crooks (1988), McKeachie (1986), and Wergin (1988) report that

students study in ways that reflect how they think they will be tested. If they think they will

be assessed on facts, memorising details will be their focus when preparing; if they think they

will be assessed on problem-solving skills or knowledge integration, they will work toward

understanding and applying information. Third, it helps teachers understand how

successfully the teaching material has been. This information may be desirable to other

parties too, such as parents. Last but not least, it reinforces continuous learning as it provides

indicators of what topics or skills students have not yet mastered and should concentrate on.

Despite these benefits, the outcome of the assessment may affect the emotional well-

being of students as it may produce a certain level of anxiety. As such, educational

researchers seem to use MCQ as a format that is effective in motivating, measuring and

reinforcing continuous learning. Multiple choice questions is a form of objective assessment

in which respondents are asked to select only correct answer(s) from the choices offered as

a list. MCQs are appropriate for assessing students’ mastery of details and specific

knowledge, ranging from simple concepts to complex ones. Moreover, since the questions

can be answered quickly, students’ understanding in several topics can be assessed in a short

106
period of time. The process of grading is also significantly shorter as compared to other

formats.

Additionally, there have been several tools available out there to easily test the

reliability of the MCQs, thus adding another advantage of using MCQ format. Reliability

here is an indication of how the test questions are internally consistent, i.e. the construction

of the questions is in such a way that they reflect students’ conceptual understanding

consistently. Cronbach’s alpha will be a measure used in this thesis study to measure the

consistency of answering the test items in both constructed assessments. The value of

Cronbach’s alpha will generally increase as the intercorrelations among test items increase,

indicating internal consistency estimate of reliability of test scores. Because intercorrelations

among test items are maximised when all items measure the same construct, Cronbach’s

alpha is widely believed to indirectly indicate the degree to which a set of items measures a

single unidimensional latent construct. The level of internal consistency will be determined

by the value of alpha. We will follow the commonly accepted rule for describing this internal

consistency (George & Mallery, 2003; Kline, 2000; DeVellis, 2016), as shown in Table 4.3.

Table 4.3

Level of Internal Consistency based on Cronbach’s Alpha

𝛼 Internal consistency

.9 ≤ 𝛼 excellent

. 8 ≤ 𝛼 < .9 good

. 7 ≤ 𝛼 < .8 acceptable

107
 . 6 ≤ 𝛼 < .7 questionable

. 5 ≤ 𝛼 < .6 poor

𝛼 < .5 unacceptable

Despite the benefits of MCQ assessment, science education researchers have

explored the use of multi-tier questions as compared to the traditional MCQs. For instance,

the use of 2-tier multiple-choice (2TMC) tests (Kwen & Cheng, 2005; Adodo, 2013; Haja &

Clarke, 2011; Loh et al., 2014; Lin, 2004) allows educators to identify alternative conceptions

harboured by students in a more robust manner compared to traditional MCQs. In a 2TMC

assessment, the first tier assesses students’ ability to arrive at the correct answer while the

second tier assesses their ability to provide the correct reasoning. As such, the possibility of

students scoring due to guessing is minimised. However, the findings of the study by Griffard

& Wandersee (2001) raised concerns about the validity of using 2TMC tests for diagnosing

alternative conceptions as it was still not certain whether the students’ mistakes were due to

alternative conceptions or the phrasing of the questions. Tamir (1989) expressed another

concern in that the options in a 2-tier test instrument might provide clues to correct answers

and reasons, a luxury that does not exist in a free-response test or interview. Table 4.4

provides a partial list of 2TMC test instruments in physics.

108
Table 4.4

2TMC Test Instruments in Physics

Reference Test Instrument

The Test of Image Formation by Optical Reflection

Chen et al. (2002)
(TIFOR)

Fetherstonaugh & Treagust (1992) Student Understanding of Light and Its Properties

Chu et al. (2009) Light Propagation Diagnostic Instrument (LPDI)

Two-tier Physics Questionnaire (on mechanics, electricity

Chang et al. (2007)
and magnetism, heat, sound and wave, and optics)

An example of 2TMC question is given in Figure 4.3.

Figure 4.3

Example of 2TMC Question (Chang et al., 2007)

109
 Caleon & Subramaniam (2010a) and Hasan et al. (1999) described significant

limitations of 2TMC tests in that, those tests cannot differentiate mistakes due to lack of

knowledge from mistakes due to existence of alternative conceptions; and that they cannot

differentiate correct responses due to scientific knowledge from those due to guessing.

Moreover, even when students exhibit incorrect reasoning, it is arguable whether their

incorrect reasoning is indeed a genuine alternative conception or a spurious one; the latter

occurs as students simply guess the reason.

The limitations mentioned for the 2TMC tests can be compensated significantly by

incorporating a third tier to each item of the test, asking for the confidence in the answers

given in the first two tiers. When this additional tier assesses the composite confidence rating

for both answer and reason, it is referred to as 3-tier multiple-choice (3TMC) test (Hoe &

Subramaniam, 2016; Saat et al., 2016; Aydın, 2007; Caleon & Subramaniam, 2010a;

Eryılmaz, 2010; Kutluay, 2005; Peşman & Eryılmaz, 2010; Türker, 2005). When the

confidence ratings are evaluated separately for answer and reason, the case is referred to as

4-tier multiple-choice (4TMC) test (Caleon & Subramaniam, 2010b; Sreenivasulu &

Subramaniam, 2013, 2014; Yan & Subramaniam, 2016). The confidence rating is an example

of evaluation of students’ metacognitive skills. Students with good metacognition would be

able to give an accurate description of whether they really know or don’t know about the

concept tested. This has provided a greater sense of alternative conceptions harboured by

students as it is supported by the numerical analyses of confidence ratings for each answer

and reason for every question (in the case of 4-tier question), or average confidence ratings

for both answer and reason for all questions (in the case of 3-tier question). This confidence

110
rating would at least provide an indication of the strength of student’s mistake for incorrectly

answered questions. For example, an incorrect answer chosen with a high certainty of

response can be considered as an alternative conception (Hasan et al., 1999; Potgieter et al.,

2010), and hence it helps the researchers to ensure that the test results are free from false

positives and false negatives, since each requires different remediation and treatment. This

certainty of response, used in the psychology literature, is similar to confidence rating in this

study. Tables 4.5 and 4.6 provide a partial list of 3TMC and 4TMC instruments in physics.

Table 4.5

3TMC Test Instruments in Physics

Reference Test Instrument

Eryılmaz (2010) Three Tier Heat & Temperature Test

Peşman & Eryılmaz (2010) Simple Electric Circuit Diagnostic Test (SECDT)

Caleon & Subramaniam (2010a) The Wave Diagnostic Instrument (WADI)

Kızılcık & Güneş (2011) Three Tier Circular Motion Test

Kaltakci & Didis (2007) Gravity Concept Test

Aykutlu & Şen (2012) Electricity Concept Test

Taslidere (2016) Photoelectric Effect Concept Test

111
Table 4.6

4TMC Test Instruments in Physics

Reference Test Instrument

Caleon & Subramaniam (2010b) Four Tier Wave Diagnostic Instrument (4WADI)

Kaltakçı (2012) Four Tier Geometrical Optics Test (FTGOT)

Hermita et al. (2017a) Static Electricity Concept Test

Fariyani et al. (2017) Geometrical Optics Concept Test

Kaltakci-Gurel et al. (2017) Geometrical Optics Concept Test

Afif et al., (2017) Energy and Momentum Concept Test

Sholihat et al., (2017) Fluid Dynamics Concept Test

Hermita et al., (2017b) Magnetism Concept Test

Caleon and Subramaniam (2010a) further classified these alternative conceptions

based on the confidence ratings. Genuine alternative conceptions are held by at least 10% of

the sample with mean confidence exceeding 3.5 (on a confidence scale of 1-6), while

spurious alternative conceptions are held by at least 10% of the sample with mean confidence

less than 3.5. High level of confidence in each occurrence of alternative conception suggest

to us that the ideas are anchored in students’ mental models strongly (McClary & Bretz,

2012), and more effort should be put in the intervention for the students to have better

conceptual understanding.

112
Figure 4.4

Example of 3TMC Question (Caleon & Subramaniam, 2010a)

113
Figure 4.5

Example of 4TMC Question (Caleon & Subramaniam, 2010b)

All the 2TMC, 3TMC and 4TMC formats have proven to be effective to excerpt

students’ conceptual understanding and alternative conceptions. However, if one of these

formats were to be used to excerpt students’ mathematical thinking in physics, the options in

the answer and/or reason tiers had to be written mathematically. This is a difficult task to do

as they need to encapsulate the thinking approach and not just the mathematical formulas or

results. Students might also be driven towards selecting a mathematical expression they have

memorised or that is familiar to them. As such, the existing formats still have limitations if

114
they were to be used in our study. A new tier which encourages students to express their

conceptual understanding mathematically in a fifth tier is thus necessary. It is thus a

refinement of the 4-tier format.

4.9. Mathematical Thinking as a Teaching Intervention Strategy

To support students’ conceptual understanding, decades of research in education have

produced numerous reports on how effective teaching intervention strategies have been

successfully applied. Each approach has its own strengths and weaknesses and leads to

different impacts and results in the teaching and learning processes. In the subject of Physics,

previous studies have explored a number of strategies for conceptual changes in various

topics, including our topics of exploration, i.e. momentum/collision and rotational

mechanics. Some of them are listed in Table 4.7, and the list is certainly not exhaustive. On

the topic of momentum/collision, for instance, problem-based learning (PBL), REACT

strategy, microcomputer-based laboratories and other strategies have been explored (Sahin,

2010; Ültay, 2012; George et al., 2013). More details on the alternative conceptions

identified and their respective conceptual change strategy can be found in Tables 4.1 and 4.2.

115
Table 4.7

Conceptual Change Strategies in Physics

Conceptual change
Topic Level
strategy

Vicarious learning through

Quantum Mechanics (Muller et al., 2007) Undergraduate
observation

Electric Circuits (Kim et al., 2006) Undergraduate Physics by Inquiry (PBI)

Instructional analogy
Refraction of light (Chinyere & Madu, 2006) High School
teaching

Computer-supported physics
Mechanics (Tao & Gunstone, 1999) High School
instructions

Mechanics (Brown, 1992) High School Examples and analogies

Pre-school and Gradual process of belief

Sound (Mauzens & Lautrey, 2003)
primary school revision

Mechanics (Bryce & Macmillan, 2005) High School Bridging analogies

Unfortunately, all these conceptual change strategies are not suitable for our study,

as the element of mathematical thinking with respect to the strategy could be minimally found

in the literature. In the PBI approach (McDermott, 1995), for example, the element of active

learning could be a strong factor in addressing alternative conceptions. This is different from

the traditional teaching approach where questions are only asked as summative assessment,

even if they are asked during the lecture itself. But even with all its positive features,

unfortunately the aspect of mathematical thinking is missing. In this approach, the concepts

are usually derived from experimental data, and mathematics is introduced only at the end

but not as thinking tools during the inquiry process itself. Intriguingly, as a consequence,

students with good conceptual understanding may view the activities or the guiding questions

116
during the inquiry process as not challenging enough to require explicit thinking in every

step of the scientific inquiry process. For example, in an interview to evaluate the inquiry

approach conducted in a study by Ma et al., (2007), the interviewer noted:

“He [the middle school student] almost saw exactly what it needed to be, and all the

steps in between he just didn’t see as necessary. It is kind of hard to tell somebody to

slow down, take it step by step, whenever he knows it is just a matter of intuition...”

This suggest that there is a strong tendency for students to rely on their scientific intuition to

obtain the final answer of a problem the moment the concepts have been established in their

thoughts. In other words, there is a tendency to jump into conclusion rather early without

considering the necessity of carefully articulating their thoughts in solving a particular

problem.

Another example was given by Steinberg (2012), based on his experience in teaching

a group of students about the relative motion between the Earth and the Sun. In his study,

students would successfully perform qualitative reasoning during the inquiry process. But as

a consequence, they might think that they were being directed to use more qualitative

reasoning whenever it comes to problem-solving, thus downplaying the importance of

mathematical thinking. To address the abovementioned issue, a teaching intervention that

encourages, promotes, expounds and makes use of mathematical thinking in learning physics

is considered to be effective for students with good basic conceptual understanding, i.e. high-

performing students. This teaching intervention is also deemed to result in conceptual

changes.

117
4.10. Gaps in the Literature

Physics is a study of nature and it requires conceptual understanding as well as use of

mathematics in the formulation of its laws and principles. A physics concept often requires

a lot of idealisations and involves a number of abstractions, sometimes involving principles

which are counterintuitive. In view of these, it is not surprising that myriad researchers have

investigated the nature and prevalence of students’ alternative conceptions and even

developed instruments of various formats as tools for determining alternative conceptions as

well as measurement of conceptual changes due to an intervention. Among all these studies,

high-performing students received the least attention as these educational researchers

probably felt that they do not require any help in levelling up their conceptual understanding.

Mathematical thinking is known as an important element for logical thinking in the

study of science, especially physics. Studies also identified students’ mathematical ability,

shown in their mathematics scores, as a crucial factor for them to do well in physics

assessment. However, the investigation of students’ mathematical thinking in a physics

assessment received minimum attention and, to the best of our knowledge, no study on the

excerpting of students’ mathematical thinking in physics assessment has been done. This gap

in the literature motivated our current study. In this study, mathematics was not only explored

as a problem-solving tool but also as a thinking tool. Students’ mathematical thinking was

excerpted from their mathematical approaches to addressing physical situations.

As much excerption of students’ mathematical thinking was desirable, it was

necessary for the physics assessments used in our study to contain multiple layers of thinking,

and not just a simple problem-solving set. Although the concepts involved are simple high

school physics concepts, mathematical thinking was expected to be done by student

118
participants as they approach the problems in the assessments. This need signifies the

construction of higher order thinking problems. Unfortunately, although there were a lot of

mathematical explorations in the physics literature, these were usually too advanced for high

school students. Hence, it was mandatory for theoretical explorations using mathematical

thinking on high school physics topics to be done as the first part of this thesis study. These

explorations were done and used for the development of the physics instruments for the

current study.

From the science education literature, there had been a lot of developments in the

conceptual surveys that were claimed to be effective in determining alternative conceptions

harboured by physics students from various levels and in various topics. However, even

multi-tier format is still not sufficient to excerpt students’ mathematical thinking. The 4-tier

format, for instance, evaluated students’ answers and reasons to physics questions as well as

their confidence ratings for the answers and reasons. This format of test instrument was

viewed to be robust in terms of evaluating students’ cognition and metacognition, and it had

been used by many researchers. Nevertheless, coming up with 4-tier questions where each

tier was mathematically-oriented is indeed difficult. If this were to be forced, it would create

another problem in which student participants might be selecting responses based on their

memory or the formulas they are familiar with. And hence, these newly-constructed

instruments had to include an additional tier in which students’ mathematical thinking could

be excerpted from students’ responses in that tier. Free-response format was preferred for

this fifth tier for better excerption and analyses of students’ mathematical thinking in relation

to their conceptual understanding. The construction of these 5-tier instruments for our current

study was also part of this thesis study.

119
 Last but not least, as higher order thinking problems were involved, it was mandatory

for our samples to be high-performing students. If alternative conceptions were even

expected to be harboured by high-performing students, then a great limitation would be faced

if the study involved a general population of students. To evaluate the suitability of our

sample for our main studies, the results of their performance in several physics conceptual

surveys would need to be analysed and compared against other students’ performance, as

reported in the literature. This analysis was also part of this thesis study prior to the execution

of our main studies. As the performance of these students was expected to be left-skewed,

neither could we use these instruments confidently for our main studies due to ceiling effect,

i.e. they might not be able to measure any improvement in scores.

In summary, the literature in physics education shows the following shortcomings:

(1) very minimal or hardly any studies have focused on exploring mathematical thinking

in students;

(2) existing formats of multi-tier instruments focus more on elucidating conceptual

understanding and identifying alternative conceptions, and do not seem to be suitable

for use in probing students’ mathematical thinking;

(3) the topics of momentum/collision and rotational mechanics have been the subject of

very few studies, and can benefit from further explorations; and

(4) studies on teaching interventions involving mathematical thinking have not been

given due recognition in the literature.

This thesis thus aims to address these shortcomings in relation to the education research

aspect.

120
 CHAPTER 5

METHODOLOGY

In this chapter, the methodology for the evaluation of suitability of our study sample,

the development and administration of the instruments in our pilot and main studies as well

as the framework for the teaching intervention will be presented. In addition, the process of

data analyses will be elaborated in detail.

Here, we could see several stages necessary for this research. Firstly, as the research

was specifically catered for high-performing physics students, assessments needed to be

administered to obtain the right pool of research subjects. Secondly, suitable assessments

need to be constructed in view of the gaps in the literature. Focus of this study on different

physics topics would provide stronger basis for the findings. As such, two topics were

explored: momentum/collision and rotational mechanics. Thirdly, the data analysis on

students’ performance in each tier was done to examine the association between conceptual

understanding and mathematical thinking as well as for the analysis of alternative

conceptions. Lastly, effective teaching interventions were designed and implemented to see

whether students’ mathematical thinking in physics can be improved. The effectiveness of

this strategy would be seen in the analysis results of the post-tests.

The participants of this study, as a representative of high-performing students, are

students from the NUS High School of Mathematics and Science in Singapore. NUS High

School has a unique feature as it focuses a lot more on mathematics and science subjects, and

thus attracts students with strong passion and interest in specialising in these subjects. The

school conducts rigorous selection at the entrance point, with very high criteria for admission.

121
Academic achievements and students’ interest are carefully assessed during the selection test,

selection camp and interviews. For Physics subject, students would complete the syllabus

beyond GCE O-level in the first three years of their study (Grades 7-10). In the last three

years of their study (Grades 11-12), students would complete the syllabus that is beyond GCE

A-level2, almost equivalent to first year undergraduate curriculum. Therefore, on average,

the advanced curriculum in NUS High School gave a strong indication that their pool of

students could be potentially classified as high performers, and thus suitable as candidates

for our main study. However, this claim still had to be validated. As conceptual understanding

is closely linked to the outcome of the learning process, it is often measured using assessment

instruments or surveys given to students.

5.1. Evaluating Suitability of Sample for Current Study

NUS High School administers several readily-available physics instruments to their

students every year to ensure that the quality of the students is maintained, as well as for

batch-to-batch comparison. These instruments were administered only to students who were

in the physics track. With the school’s permission, data from several batches of students were

collected and analysed. These data were a good reflection of the quality of NUS High School

students, especially in the subject of Physics. The questions cover a wide range of concepts

in mechanics as well as electricity and magnetism, and students would need to consolidate

their learning when sitting for the test.

2
Refer to the General Certificate of Education (GCE) O-Level and A-Level H3 Syllabi examined in 2019. The details of
the syllabus for the GCE O-Level can be obtained from https://www.seab.gov.sg/home/examinations/gce-o-level/2019-gce-
o-level-syllabus-for-school-candidates and for the GCE A-Level H3 can be obtained from
https://www.seab.gov.sg/docs/default-source/national-examinations/syllabus/alevel/2019syllabus/9814_2019.pdf

122
5.1.1. Materials

The instruments used for this preliminary study were the Force Concept Inventory

(FCI) (Hestenes et al., 1992), Mechanics Baseline Test (MBT) (Hestenes & Wells, 1992),

and Conceptual Survey in Electricity and Magnetism (CSEM) (Maloney et al., 2001). The

FCI assessment instrument consists of 30 multiple choice questions (MCQs) designed to

assess student conceptual understanding of basic concepts in Newtonian mechanics. They

assess student conceptual understanding and do not necessarily involve computation. The

MBT consists of 26 MCQs, which also assess student conceptual understanding of

Newtonian mechanics. The MBT is considered to be generally more difficult than the FCI,

and, unlike the FCI, more questions in the instrument are computational in nature. The FCI

and MBT are considered complementary to each other and, together, they give a fairly

complete profile of a student’s understanding of basic Newtonian concepts. In contrast, the

CSEM assessment instrument consists of 32 MCQs on the topic of electricity and magnetism.

The questions cover almost all concepts and it gives a fairly-sufficient baseline measurement

for a student before they go deeper on this topic, for instance, calculus-based calculations.

Students were not given any time limit to complete each instrument. However, each

year, all students managed to complete each instrument within less than 45 minutes. For the

purpose of data triangulation, interviews were conducted for a sample in one of the batches

to understand the students’ thinking pattern as they approached the questions. Students were

first ranked based on their performance and then randomly selected for the interview to

ensure that the interview participants were taken from the top, middle and bottom performers.

Several questions, with the lowest number of students who answered these correctly, were

123
selected for use in the interview. These questions were used to understand how students had

approached the questions and to see the prevalence of any alternative conceptions. These

interviews were also used to check that the participants were truly engaged during the

assessment and were not merely guessing.

5.1.2. Participants

The participants for this preliminary study were 158 Year 5 students across 3 different

batches (Grade 11; 34 females; mean age ± 𝑆𝐷 = 16.89 ± .76 years) and 65 Year 6 students

(Grade 12; 16 females; mean age ± 𝑆𝐷 = 17.51 ± .77 years). Thirty students of Years 5 and

6 from one of the years were randomly selected for interviews. Parents of all interviewed

participants provided written consent, and the interviewed participants provided written

assent under a protocol approved by the Nanyang Technological University (NTU)’s

Institutional Review Board. Years 5 and 6 (equivalent to Grades 11 and 12) students were

selected by the school to ensure that the participants have learnt the topics, i.e. mechanics,

electricity and magnetism.

5.2. Study on Momentum/Collision

With the limitations in the literature, we found that the content published by Ricardo

& Lee (2015) has unique strengths in which the content covered requires simple physics

concepts with the use of high school mathematics, and some pedagogical aspects are

elaborated in which many students have a strong tendency in harboring alternative

conceptions in higher order thinking questions. The participants in this study were ensured

124
to have gone through the concepts of momentum/collision and to have sufficient

mathematical skills for problem-solving.

5.2.1. Development of Initial Version of Instrument

This section elaborates on the initial version of the instrument constructed from the

concept of chain-collision elaborated in Chapter 2. We will present the questions in our

constructed assessment, students’ expected answers and correct solutions, which was used

for the pilot study. The assessment consists of questions ranging from conceptual questions,

standard problem-solving questions, to ones requiring higher order thinking. For the purpose

of the pilot study, a free response format was chosen as we would like to excerpt students’

mathematical thinking as much as possible from their mathematical representation of

answers.

This initial version of the instrument consists of eight free-response questions.

Questions 1-3 simply recall the students’ conceptual understanding on momentum/collision,

Questions 4-6 are standard problem-solving questions that students should be familiar with

at the time when they were tested, and Questions 7-8 are higher order thinking questions on

this topic. For each question, students were also expected to indicate their confidence rating

for their response using a Likert scale from 1 to 5, with 1 being “not confident at all” and 5

being “very confident”. One mark was assigned for correct answer for each question or sub-

question. The following are the elaboration of the questions in the test instrument for this

pilot study.

125
Questions on Momentum/Collision with Elaboration of Possible Responses and

Expected Solutions:

(1) How do you define momentum? [1 mark]

Some students were expected to answer this question qualitatively, while others were

expected to answer it mathematically. Answers such as “the strength of an object while it is

moving”, “the quantity of motion”, or somewhere along that line, would be acceptable as the

correct answers, as long as students showed some level of understanding of this physical

quantity. However, answers such as “the amount of force acting on an object” or “the amount

of energy an object possesses” would not be accepted as they refer to other physical quantities

that are of different dimensions. Mathematically, momentum is defined as the product of an

object’s mass and its velocity, i.e.

⃗⃗ = 𝑚𝐯⃗⃗
𝐩

(2) What is the necessary condition for the total momentum in a system to be conserved?

[1 mark]

This concept is a basic building block for the subsequent questions. As long as a system is

isolated, the total momentum of the system is conserved. Mathematically, the net external

force acting on the system equals to zero.

⃗⃗
∑ 𝐅⃗ = 𝟎

Both mathematical and qualitative answers are acceptable.

126
(3) What is the necessary condition for an elastic collision? [1 mark]

In a collision, energy can be dissipated as heat, sound and other forms as deformation takes

place. A collision is categorised as an elastic collision if and only if there is no energy loss

during the collision. In other words,

1 1 1 2 1 2
𝑚1 𝑣1 2 + 𝑚2 𝑣2 2 = 𝑚1 𝑣1′ + 𝑚2 𝑣2′
2 2 2 2

where 𝑚 and 𝑣 denote the mass and velocity of the blocks, prime (′) denotes the condition

after collision, index 1 refers to the first body and index 2 refers to the second body.

Qualitative answer such as “energy is conserved” would also be acceptable as it showed the

same degree of conceptual understanding.

(4) Consider the following one-dimensional collision. Block A, of mass 6 kg, moves with

velocity 3 m/s to the right. It hits block B, of mass 2 kg, which is at rest. After the

collision, block A moves with velocity 2 m/s to the right. Ignore friction.

(a) Explain how you can find the velocity of block B after the collision, and find

it. [1 mark]

(b) Explain whether the situation described above is possible to happen. Why? [1

mark]

This question was a standard problem-solving question. The only physical quantities to

consider are momentum and energy, and both concepts, including various types of collision,

have been covered in the core physics module prior to this assessment. Most students were

expected to be able to answer the first part of this question.

𝑚𝐴 𝑣𝐴 + 𝑚𝐵 𝑣𝐵 = 𝑚𝐴 𝑣𝐴′ + 𝑚𝐵 𝑣𝐵′

127
 (6)(3) + (2)(0) = (6)(2) + (2)𝑣𝐵′ → 𝑣𝐵′ = 3 m/s

To explain whether the situation is possible, students were expected to compare the initial

kinetic energy and the final kinetic energy of the system,

1 1 1 1
𝐾𝑖 = 𝑚𝐴 𝑣𝐴 2 + 𝑚𝐵 𝑣𝐵 2 = (6)(3)2 + (2)(0)2 = 27 J
2 2 2 2

1 2 1 2 1 1
𝐾𝑓 = 𝑚𝐴 𝑣𝐴′ + 𝑚𝐵 𝑣𝐵′ = (6)(2)2 + (2)(3)2 = 21 J
2 2 2 2

which indicated a dissipative collision. Hence, the scenario was possible. The same argument

could also be obtained by comparing the absolute value of the initial relative velocity and the

absolute value of the final relative velocity, or equivalently, the coefficient of restitution,

𝑣𝐵′ − 𝑣𝐴′ 3−2 1

𝑒=− =− =
𝑣𝐵 − 𝑣𝐴 0−3 3

Since 0 ≤ 𝑒 ≤ 1, the scenario was possible.

(5) Consider the following one-dimensional collision. Block C, of mass 6 kg, moves with

velocity 3 m/s to the right. It hits block D, of mass 2 kg, which is at rest. After the

collision, block C moves with velocity 1 m/s to the right. Ignore friction.

(a) Explain how you can find the velocity of block D after the collision, and find

it. [1 mark]

(b) Explain whether the situation described above is possible to happen. Why? [1

mark]

Similar to the previous question, this question was a standard problem-solving question.

Students were expected to be able to answer the first part of this question. Applying the

principle of conservation of momentum,

128
 𝑚𝐶 𝑣𝐶 + 𝑚𝐷 𝑣𝐷 = 𝑚𝐶 𝑣𝐶′ + 𝑚𝐷 𝑣𝐷′

(6)(3) + (2)(0) = (6)(1) + (2)𝑣𝐷′

𝑣𝐵′ = 6 m/s

By comparing the initial kinetic energy and the final kinetic energy of the system,

1 1 1 1
𝐾𝑖 = 𝑚𝐶 𝑣𝐶 2 + 𝑚𝐷 𝑣𝐷 2 = (6)(3)2 + (2)(0)2 = 27 J
2 2 2 2

1 2 1 2 1 1
𝐾𝑓 = 𝑚𝐶 𝑣𝐶′ + 𝑚𝐷 𝑣𝐷′ = (6)(1)2 + (2)(6)2 = 39 J
2 2 2 2

this scenario was clearly impossible assuming no explosive or coiled spring, etc., to supply

more energy, as the kinetic energy had increased. Equivalently, the coefficient of restitution,

𝑣𝐷′ − 𝑣𝐶′ 6−1

𝑒=− =− =2
𝑣𝐷 − 𝑣𝐷 0−3

Since 𝑒 > 1, the scenario was impossible.

(6) Consider the following one-dimensional collision. Block R, of mass 4 kg, moves with

velocity 9 m/s to the right. It hits block S, which is at rest. After the collision, block R

moves with velocity 1 m/s to the right, and block S moves with velocity 4 m/s to the

right. Ignore friction.

(a) Find the mass of block S. [1 mark]

(b) What percentage of the initial kinetic energy of block R is transferred to

block S during the collision? [1 mark]

Firstly, to ensure that the answer for part (b) is less than or equal to 100%, students were

expected to check for the coefficient of restitution. However, this step was not necessary as

129
students would have realised whether the scenario was possible after checking the coefficient

of restitution.

𝑣𝑆′ − 𝑣𝑅′ 4−1 4

𝑒=− =− =
𝑣𝑆 − 𝑣𝑅 0−9 9

Since 0 ≤ 𝑒 ≤ 1, the scenario was possible.

The first part of this question was also a simple problem-solving question. Applying the

principle of conservation of momentum,

𝑚𝑅 𝑣𝑅 + 𝑚𝑆 𝑣𝑆 = 𝑚𝑅 𝑣𝑅′ + 𝑚𝑆 𝑣𝑆′

(4)(9) + 𝑚𝑆 (0) = (4)(1) + 𝑚𝑆 (4) → 𝑚𝑆 = 8 kg

For the second part, the kinetic energy transfer is

1
( )( )2
2 8 4 = 32 = 39.5%
1
( )( )2 81
2 4 9

(7) Consider the following one-dimensional collision. Block E, of mass 4 kg, moves with

velocity 9 m/s to the right. It hits block F, of mass 9 kg, which is at rest. We want to

transfer the initial kinetic energy of block E as much as possible to block F. Ignore

friction.

(a) Find the velocities of block E and block F after the collision. [1 mark]

(b) Explain why you approach the question that way.

An alternative conception was anticipated for this question, as many students were expected

to think that block E should stop immediately after the collision to transfer its energy as much

as possible to block F. It was therefore expected that some students would consider the

transfer of 100% momentum from block E to block F, however, the answer obtained was

130
actually not so efficient for kinetic energy transfer. Some other students were also expected

to consider the transfer of 100% energy from block E to block F, however, this would either

violate conservation of momentum or not so effective as compared to elastic collision. Hence,

applying the principle of conservation of momentum and the coefficient of restitution for

elastic collision,

𝑚𝐸 𝑣𝐸 + 𝑚𝐹 𝑣𝐹 = 𝑚𝐸 𝑣𝐸′ + 𝑚𝐹 𝑣𝐹′

(4)(9) + (9)(0) = (4)𝑣𝐸′ + (9)𝑣𝐹′ → 4𝑣𝐸′ + 9𝑣𝐹′ = 36

𝑣𝐹′ − 𝑣𝐸′ 𝑣𝐹′ − 𝑣𝐸′

𝑒=− =− = 1 → 𝑣𝐹′ − 𝑣𝐸′ = 9
𝑣𝐹 − 𝑣𝐸 0−9

From the above equations,

45
𝑣𝐸′ = − m/s = −3.46 m/s
13

72
𝑣𝐹′ = m/s = 5.54 m/s
13

The second part of this question was simply to get the information on students’ approaches

to the question in the first part.

(8) Consider the one-dimensional collision in the previous question. Someone suggests

that we can increase the amount of kinetic energy transfer (kinetic energy that is

transferred from block E to block F), by inserting another block, call it block G, which

is also at rest, between block E and block F. So in this scenario, block E will hit block

G, and block G will hit block F afterwards. Explain why you think the suggestion

works or does not work. Assume no energy loss. Ignore friction. [1 mark]

131
Both qualitative reasoning and intuition have great limitations when approaching this

question. It was expected that many students will think that involving more masses in the

collision would only lead to more energy loss and, hence, the suggestion would not work.

However, inserting block G of an appropriate mass in between blocks E and F would increase

the kinetic energy transfer from block E to block F as the collision is perfectly elastic (no

energy loss). This is shown in the following. Let the mass of block G be 𝑚 and let block G

be stationary initially. Applying the principle of conservation of momentum and the

coefficient of restitution for elastic collision for block E and G,

𝑚𝐸 𝑣𝐸 + 𝑚𝐺 𝑣𝐺 = 𝑚𝐸 𝑣𝐸′ + 𝑚𝐺 𝑣𝐺′

(4)(9) + 𝑚(0) = (4)𝑣𝐸′ + 𝑚𝑣𝐺′ → 4𝑣𝐸′ + 𝑚𝑣𝐺′ = 36

𝑣𝐺′ − 𝑣𝐸′ 𝑣𝐺′ − 𝑣𝐸′

𝑒=− =− = 1 → 𝑣𝐺′ − 𝑣𝐸′ = 9
𝑣𝐺 − 𝑣𝐸 0−9

From the above equations,

72
𝑣𝐺′ =
𝑚+4

Applying the principle of conservation of momentum and the coefficient of restitution for

elastic collision for block G and F,

𝑚𝐺 𝑣𝐺′ + 𝑚𝐹 𝑣𝐹 = 𝑚𝐺 𝑣𝐺′′ + 𝑚𝐹 𝑣𝐹′

72 72𝑚
𝑚 + (9)(0) = 𝑚𝑣𝐺′′ + 9𝑣𝐹′ → 𝑚𝑣𝐺′′ + 9𝑣𝐹′ =
𝑚+4 𝑚+4

𝑣𝐹′ − 𝑣𝐺′′ 𝑣𝐹′ − 𝑣𝐺′′ ′ ′′

72
𝑒=− = − = 1 → 𝑣𝐹 − 𝑣𝐺 =
𝑣𝐹 − 𝑣𝐺′ 0−𝑚+4
72 𝑚+4

Solving the above equations,

132
 144𝑚
𝑣𝐹′ =
(𝑚 + 4)(𝑚 + 9)

Plugging in any value for 𝑚 within the range of 4 < 𝑚 < 9 yields a value for 𝑣𝐹′ that is

bigger than 5.54 m/s. Hence, the suggestion works.

Overall, the questions in this phase of the study do not require a specific approach to

be used. We expected that qualitatively we could see the improvement in the students’ ability

in performing mathematical thinking after undergoing a lesson on momentum/collision

where mathematical thinking was emphasised. This initial version of the instrument was first

validated by another physics teacher at NUS High School and verbal feedback was gathered

before it was administered to the students for the pilot study.

5.2.2. Pilot Study

The initial version of the instrument was administered to 112 Year 3 (Grade 9)

students constituting two different batches (Batch 2015: 46 students, 2 females, mean age ±

𝑆𝐷 = 15.22 ± .55 years; Batch 2016: 66 students, 12 females, mean age ± 𝑆𝐷 = 15.20 ± .50

years) of NUS High School, as part of their post-exam activity. All participants provided

written informed consent and assent in compliance with a protocol approved by the Nanyang

Technological University (NTU)’s Institutional Review Board. It was noted that although no

time limit was given to the participants, they were able to complete the assessment in less

than an hour.

One may argue that the scoring for open-ended assessment items may be somewhat

subjective, even with a clearly-defined marking scheme and, hence, this initial version of the

133
test instrument might not be robust enough for the analysis of conceptual understanding,

mathematical thinking and determination of alternative conceptions. Nevertheless, the results

analysis do provide rich insights into students’ thinking, expressed in the way they

approached each question.

5.2.3. Instrument Validation: Chain-Collision Assessment

The results analysis from the pilot study gave us a better idea on the kind of questions

to be constructed for the main study assessment and what kind of instructions to be put clearly

on the assessment. The higher order thinking questions used were expanded into a number

of questions in the amended version of the instrument so as to thoroughly extract students’

ways of thinking and to better identify the alternative conceptions harboured by them. After

a few rounds of feedback, the multi-tier format was adopted, as compared to free-response

format used in the pilot study for consistency in marking, analysis of answer and reason tiers,

analysis of confidence ratings, as well as the determination of alternative conceptions. This

format would also provide ease in computing the internal consistency of the assessment.

Meanwhile, the excerpt of students’ mathematical thinking could be done via the analysis of

the fifth tier, emphasising the benefit of a 5-tier test instrument. It was also learnt that the test

items needs to explicitly instruct the students to elaborate on their answer and/or reason

mathematically, in order to better understand their mathematical thinking. Failing to do so

would result in fewer students engaging in mathematical thinking when approaching the

questions.

After the first amendment, the instrument consisted of sixteen 5-tier questions, all

having equal weightage. For some of the questions, no clear distinction was made between

134
the reason and mathematical expression tiers as students were expected to provide their own

mathematical responses for the reason tier. This version of the instrument was then sent for

validation to three physics teachers teaching various levels in NUS High School. Several

feedbacks were given by them and these were used to further refine the instrument, with

some comments incorporated in revising certain questions, including the distracters for the

answer and reason tiers.

In the validation form, nine items in the Yes/No (Y/N) format were given in a

checklist to ensure that the instrument was targeted correctly for the specific group of

participants, i.e. Year 4 (Grade 10), and the specific topic tested was already taught to the

participants (Table 5.1). Moreover, it was necessary for the wordings to be clear to prevent

any ambiguity, be grammatically correct, and that the information to be excerpted would be

useful for this thesis study.

Table 5.1

Checklist for Validation of Chain-Collision Assessment by Physics Teachers (with

Responses)

Validator Validator Validator

No. Item
1 2 3

The questions are related to the topic of momentum

1 Y Y Y
and collision in the Year 4 Physics syllabus.

The required concepts have been taught in the Year

2 Y Y Y
4 Physics core module.

135
 The questions, answers and reasons are free from
3 Y Y N
grammatical errors.

There are an adequate number of questions to test

4 students’ understanding of the concept of Y Y Y
momentum and collision.

The questions in the diagnostic instrument are clear

5 Y Y N
and comprehensible.

The multi-tier format used in the diagnostic

6 instrument is effective in probing students’ Y Y Y
understanding of the topic.

The correct responses indicated in the Answer and

7 Y Y Y
Reason tiers are acceptable.

Each tier has only one correct response, with the rest
8 being misconceptions (alternative conceptions) or Y Y Y
incorrect responses.

Overall, the diagnostic instrument can provide

9 Y Y Y
useful information for the research study.

Additionally, the validators provided some written comments to improve the presentation of

the questions. The comments are listed in Table 5.2 and these were used to further improve

the instrument.

Table 5.2

Comments by Validators and Actions Taken for Refinement of Chain-Collision Assessment

Comment
No. Regarding Comment Action Taken
by

Validator Good question that unearths two This conceptual question is purposely
1 Q1
1 common misconceptions: (1) 𝑝 is put as the first question to trigger

136
 only conserved for elastic students’ mathematical thinking for a
collisions, (2) 𝑝 is always very simple question.
conserved.

The key concept tested here is

Language used is specific to the ‘isolated system’ in which there is no
two particles. You may want to net force acting on the system. Should
Validator
2 Q1 include a response involving impulse be included, it might add an
1
impulse in reason, if it is what you unnecessary terminology into this
are testing. assessment as it is not needed for the
subsequent questions.

Is it intentional for students to The concept of potential energy is not

consider gravitational potential required for this assessment.
energy first then dismiss it because Including the term would add another
Validator collision happens at a spot, more or layer of difficulty to the assessment.
3 Q2, Q9 less?
1 Hence, not mentioning it at all is
Should we eliminate any potential preferable as it might add an
energy conversion, e.g. charged unnecessary terminology into this
blocks? assessment.

This is a good correction from the

validator. Although the original
statement is still correct
Shouldn’t this (option B) be “<”
Validator mathematically, but excluding the
4 Q3 instead of “≤”? Equality only
1 equality eliminates the perfectly
holds for completely inelastic case.
inelastic case. The question has been
amended accordingly to a True/False
format.

Although it may catch many non-

thinking students, the distracters This is a good suggestion from the
can be further reused to cater for validator. However, the key concept
the range of misconceptions, e.g. tested in this question is the ‘relative
Validator no conditions needed, 𝑣1 > 𝑣2 velocity’ of the two blocks in a certain
5 Q3
1 only. In reasoning, students may type of collision, and not the direction
argue that if 𝑚2 > 𝑚1 , then 𝑚1 of motion. To avoid any confusion,
will go backwards if 𝑣1 is the question has been amended to a
sufficiently greater than 𝑣2 as an True/False format.
example.

Options C and D can be made

clearer in wording and the
Validator grammar needs to be corrected. I The options have been amended
6 Q6, Q9
1 suggest something like “… Block accordingly.
B’s gain in velocity is greater than
the loss of velocity of Block A.”

These two questions are the higher

Validator It may not be easy for students to
7 Q14, Q16 order thinking questions set for this
1 conclude that maximum transfer
assessment. It is expected that
means elastic, especially if they
students should perform multiple

137
 need to first consider 𝑚1 > 𝑚2 or layers of thinking too approach the
𝑚1 < 𝑚2 . Is it intentional? questions.
The question is highly advanced.

There is no time limit for the students

to complete the assessment,
Validator Whole How long will the test be? It seems especially as the students are required
8
2 paper like it is going to take a long time. to elaborate on their mathematical
thinking for each question. Hence,
this is not an issue.

“… the system of two particles

Validator above is conserved.” should be The question has been amended
9 Q1, Q2
2 written as “… the system of the accordingly.
two particles is conserved.”

It was intentional that reason tiers

were not provided for some questions
as students’ reasoning was to be
It is confusing that some observed from their mathematical
confidence ratings are linked to the expressions. However, after the
Validator
10 Q3 reason tier and some confidence feedback from the validator, all
2
ratings are linked to the questions were then designed to have
explanation-tier. answer, reason and mathematical
expression tiers, with confidence
ratings for the answer and reason
tiers.

“Which of the following…”

should be written as “Which one of
the following…”
“… while it is also partially lost…”
should be written as “… and is also These are good grammatical
partially lost…” to ensure clarity corrections from the validator to
Validator of the subject.
11 Q6-Q15 ensure clarity of the questions. The
2
“… is impossible to happen.” questions and options have been
should be written as “… is amended accordingly.
impossible.”
“There is a momentum transfer…”
should be written as “There is
momentum transfer…”

Even though they (the students) The question has been amended for
might have learnt the concept of more clarity. For the rest of the
Validator Whole
12 frame of reference, it is better to questions, the concept of frame of
3 paper
rephrase the sentence to avoid reference is only one among many
confusion. approaches that students can use.

All the suggestions have been taken

Validator Whole Diagrams can be drawn bigger
13 into account and the diagrams have
3 paper with clear annotation.
been redrawn for more clarity.

138
 Put the label of each block within
the block itself to avoid confusion
and for consistency (see Q16).
The floor should be drawn to
prevent students from thinking
about the effect of gravitational
force.

“… provide the confidence

level…” should be written as “…
circle the confidence level…” for These are good suggestions from the
Validator consistency.
14 Instructions validator. The instructions have been
3
To add “Choose only one right modified accordingly.
answer and reason for each
question.”

The questions have stated clearly that

Validator It is not clear whether 𝑚2 is 𝑚2 moves to the right with velocity
15 Q1-Q5
3 moving before the collision. 𝑣2. The picture also emphasises this
statement.

The words have been removed as all

Move the words “where
Validator Whole participants are required to express
16 necessary” to the beginning of the
3 paper their mathematical thinking in the
sentence.
fifth tier.

Option B tends to be
Validator misinterpreted as no change in the The question has been amended for
17 Q4
3 direction of motion of each block. better clarity.
Rephrase?

The participants are mostly Year 4

Validator Has the concept of centre of mass (Grade 10) students and they have
18 Q4
3 been taught to the students? learnt the concept of centre of mass
prior to participating in this research.

Validator Change the word “mass” to The question has been amended
19 Q5
3 “particle”. accordingly.

The diagram has been redrawn for

Put the mass label 𝑚𝐻 in the better clarity and mass 𝑚𝐻 has been
Validator
20 Q12, Q13 question and not above the block in put in the question. However, the
3
the diagram. label is still put above the block in the
diagram for consistency.

As there are multiple collisions, the

In Option A, it is not clear that the
Validator statement applies to all the collisions.
21 Q15 energy is transferred from which
3 Hence, no amendments were made on
block to which block.
this.

139
 This is a very good suggestion from
the validator as some students might
use a long mathematical expression to
Validator Space (for the explanation) might describe their line of thoughts for
22 Q15
3 be too narrow. their answer and/or reason. For each
questions, space has been added to
ensure that students would have
enough space to write.

After incorporating feedbacks from the three validators, significant improvements

were made to the format of the questions, phrasing of the questions, as well as grammar. All

the amendments were expected to improve the quality and clarity of the questions. The final

version of the instrument comprises 15 questions in 5-tier format with uniform weightage.

Each question requires five responses from the participants: answer, confidence rating for

answer, reason, confidence rating for reason, and mathematical expression. The

mathematical expression tier is targeted to see how much of mathematical thinking students

were able to perform while applying their conceptual understanding in approaching each

question. Five questions are in True/False format for the answer tier, and the remaining 10

questions have four distracters for the answer tier. All questions have five distracters each

for the reason tier, with one being “other reason”. Participants have the option to choose

“other reason” if they disagree with the given reasons provided in the reason tier. For the

confidence ratings, as previously done by Sreenivasulu & Subramaniam (2014), Caleon &

Subramaniam (2010a), Sreenivasulu & Subramaniam (2013), a six-point scale was used from

“Just guessing” to “Absolutely confident” as it seems suitable for our study.

Although it is a standard practice in item testing to leave out questions that exceed

75% of correct responses (Rust & Golombok, 2014), we feel that in the population of high-

performing students it is better not to exclude the analysis of these questions as this provides

140
rich insights as to how the students approach other questions. This is because all questions in

the assessment are closely-related, as shown by the analysis of internal consistency. It is also

customary to perform the calculation of Facility Index (FI) and Discrimination Index (DI) as

well as the percentage of each option chosen in the answer and reason tiers. The FI for either

answer tier or reason tier in a question is simply the proportion of students who provided

correct responses for that tier. The DI for a question shows how well the question can

discriminate between the top 25% of the sample and the bottom 25% of the sample, based on

overall test scores. The details of the formula are provided in Table 5.8. A DI value greater

than .3 indicates a good discriminating question in an assessment (Rust & Golombok, 2014).

The final version of the instrument was named the Chain-Collision Assessment and it

was finalised in 2018. A sample question in 5-tier format in the Chain-Collision Assessment

is shown in Figure 5.1. The complete instrument can be found in Appendix 3.

141
Figure 5.1

Sample Question in the Chain-Collision Assessment3

Question 13
Consider the following one-dimensional collision. Block I, of mass 4.0 kg, moves with velocity 9.0 m/s to the
right. It hits Block J, of mass 9.0 kg, which is initially at rest. It is desired to transfer the initial kinetic energy
of Block I as much as possible to Block J. The friction between the blocks and the floor can be ignored.

Before collision:
4.0 kg 9.0 kg
9.0 m/s
I J

After collision:
4.0 kg 𝑣𝐼′ 9.0 kg 𝑣𝐽′
I J
Figure 6

Which one of the following statements about the system’s momentum is correct?

Answer:
(A) During the collision, the momentum of Block I is partially transferred to Block J and is also partially
lost to the surroundings.
(B) During the collision, the momentum of Block I is partially transferred to Block J while the total
momentum of the system remains conserved.
(C) During the collision, all of the momentum of Block I is transferred to Block J while the total
momentum of the system remains conserved.
(D) During the collision, there is a momentum transfer from Block I to Block J while the total momentum
of the system remains conserved. Block I and Block J move together after the collision.
(E) The collision described in Figure 6 is impossible.

Confidence 1 2 3 4 5 6
Rating for Very Absolutely
Just guessing Unconfident Confident Very confident
Answer unconfident confident

Reason:
(A) The collision in Figure 6 is elastic.
(B) The collision in Figure 6 is possible, and block I comes to rest after the collision.
(C) Both blocks are constrained to move in one dimension.
(D) Momentum is always conserved in any case.
(E) Momentum cannot be transferred from one block to another block.
(F) Other reason: ____________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Absolutely
Just guessing Unconfident Confident Very confident
Reason unconfident confident

Express your answer and/or reason mathematically, and use texts for clarifications/explanations:

3
The space presented for mathematical expression is much more than that shown above.

142
5.2.4. Main Study Participants

In order to prevent respondent fatigue, the student participants for the main study on

momentum/collision were selected from a different pool as compared to the student

participants for the FCI, MBT and CSEM tests. However, the general results of the FCI,

MBT and CSEM tests was still valid as it was performed over four different batches of

students in NUS High School, ensuring the consistency of the students’ conceptual

understanding over the years. Hence, from a research point of view, the participants for this

main study on momentum/collision could also be classified as high-performing students.

We selected a total of 123 participants (16 females; mean age ± 𝑆𝐷 = 16.51 ± 1.06),

where 112 students are from Years 4-6 (Grades 10-12) and 11 accelerated students from

Years 2-3 (Grades 8-9) who have learnt the required content in advance. These levels were

chosen as students usually learn the topic of momentum/collision, including centre of mass,

in Year 3 (Grade 9), with the exception of accelerated students. Consent to participate in the

study was sought a month before the test administration. The students were informed that

they would be sitting for a diagnostic test to understand their way of thinking in approaching

questions on physics concepts they had learnt before. They were assured that only

summarised findings would be reported, that personal particulars would not be identifiable,

and that only the researchers have access to all the collected data.

5.3. Study on Rotational Mechanics

Similar to the main study on momentum/collision, we could not find in the literature

an instrument on the topic of rotational mechanics that can assess higher order thinking and

students’ mathematical approach at the same time. As such, a new instrument had to be

143
constructed to serve this purpose. This newly formulated assessment is expected to add value

to the library of physics assessments as it is targeted specifically for high-performing

students.

5.3.1. Development of Initial Version of Instrument

The questions in the test instrument were constructed from the theoretical exploration

by Ricardo (2016) (refer to Chapter 3). The initial version of the instrument on rotational

mechanics was constructed differently from the initial version of the instrument on

momentum/collision as we would like to further see the effect of mathematical thinking in

problem-solving. This means, the instrument has to comprise two sets of assessment, one

with qualitative questioning and another with mathematical questioning, though, in essence,

they are equivalent to each other. The distinguishing factor lies in the mathematical

presentation style for the second set of test, which encourages students to perform deeper

mathematical analysis.

The tests were administered as in-class assignments for one of the physics modules

in NUS High School. As highlighted in previous studies, it is important to test for prior

knowledge of the students before providing learning guidance (Hwang, 2003; Hwang et al.,

2008; Tseng et al., 2008) and, thus, the questions range from recall questions, standard

problem-solving questions, to the ones requiring higher order thinking. We foresaw that the

use of mathematical thinking would be more beneficial as compared to qualitative reasoning.

The initial instrument was validated by another physics teacher at NUS High School,

and useful feedback was provided to improve the instrument further. As a result, five free-

response questions were constructed. Free-response format was chosen as we would like to

144
excerpt the students’ mathematical thinking as much as possible. Students’ confidence level

was also assessed for each question.

The following are the questions presented in the first test where qualitative

questioning was used. For each question, participants were expected to elaborate on their

answers as well as express their confidence level using a Likert scale from 1 to 5, with 1

being “not confident at all” and 5 being “very confident”. One mark was assigned for correct

answer for each question, except for the last question where 4 marks were awarded as it

involved more steps.

Questions in the Initial Version of Instrument on Rotational Mechanics (using

Qualitative Questioning) with Elaboration of Possible Responses and Expected

Solutions:

Consider a point mass of mass 𝑚 moving with speed 𝑣0 . The point mass hits a uniform rod

of mass 𝑀 and length ℓ perpendicularly at a point a distance 𝑥 from its centre, which initially

lies at rest on a horizontal table. Assume elastic collision. The moment of inertia of the rod
1
about its centre of mass is 12 𝑀ℓ2 .

(1) Is the total linear momentum of the system (point mass + rod) conserved? Explain

why or why not. [1 mark]

This question tested participants’ conceptual understanding of the concept of momentum.

Students were expected to understand that the total momentum of a system is conserved if

145
and only if the system is isolated or, equivalently, no net force acts on the system. In this

case, the point mass and the rod are isolated, and there is no net external force acting on them.

Hence the total momentum of the point mass and the rod is conserved.

Figure 5.2

Elastic Collision between Point Mass and Uniform Rod

(2) Is the total angular momentum of the system (point mass + rod) conserved about:

(a) the centre of the rod? Explain why or why not. [1 mark]

(b) the point of collision? Explain why or why not. [1 mark]

This question tested participants’ conceptual understanding of the concept of angular

momentum and torque. Students were expected to understand that the total angular

momentum of a system is conserved if and only if no net torque acts on the system. Two

choices of reference points were given to see if they understood the dependency of torque on

the choice of the reference point. Nevertheless, in this scenario, only internal torques exist,

and no matter which point is taken as the reference, the net external torque is zero and, hence,

the angular momentum of the system is conserved.

Yes, it

146
(3) Is the angular momentum of the rod (without the point mass) conserved about:

(a) the centre of the rod? Explain why or why not. [1 mark]

(b) the point of collision? Explain why or why not. [1 mark]

This question tested participants’ conceptual understanding that the conservation of angular

momentum depends on the choice of the system. A torque can be considered as either an

internal or external torque, depending on the choice of physical system. In this case, since

the system only consists of the rod, the force exerted by the point mass during the collision

will create a torque unless the point of collision is taken as a reference.

(4) Is the total energy of the system (point mass + rod) conserved? Explain why or why

not. [1 mark]

This question tested the participants’ conceptual understanding of an elastic collision. As the

question has specified that the collision is elastic, the total energy of the system is conserved.

(5) At which point should the point mass hit the rod to produce the largest angular speed

of the rod after the collision? Comment on your answer. [4 marks]

The question required the students to perform analysis based on the conservation laws

described in the previous questions. Qualitatively, as the collision is elastic, the initial energy

of the point mass should be given partially to the translational kinetic energy and rotational

kinetic energy of the rod, depending on the point of collision and the ratio of their masses. In

the centre of mass frame, we want to minimise translational kinetic energy as much as

possible in order to maximise rotational kinetic energy. Thus, the point mass and rod’s centre

should move with equal linear velocity and the point of collision should lie somewhere

147
between the rod’s centre and its end. However, this train of thoughts is foreseen to be limited

if one simply uses qualitative reasoning or intuition that is solely based on common sense.

Using mathematical questioning, the questions in the second set of test were presented

differently. One mark was assigned for every item except for the last two items. Question (4)

was awarded 3 marks as it involves several steps (partial marks could be obtained if student

did not arrive at the final correct answer). Question (5) was awarded 4 marks.

Questions in the Initial Version of Instrument on Rotational Mechanics (using

Mathematical Questioning) with Elaboration of Possible Responses and Expected

Solutions:

Consider a point mass of mass 𝑚 moving with speed 𝑣0 . The point mass hits a uniform rod

of mass 𝑀 and length ℓ perpendicularly at a point a distance 𝑥 from its centre, which initially

lies at rest on a horizontal table. Assume elastic collision. The moment of inertia of the rod
1
about its centre of mass is 12 𝑀ℓ2 .

(1) How would you relate the initial and final linear velocities of the point mass and the

rod (𝑣0 , 𝑣 and 𝑉), given their masses (𝑚 and 𝑀)? Explain your answer. [1 mark]

As the system consisting the point mass and rod is isolated, their total linear momentum is

conserved.

𝑚𝑣0 = 𝑚𝑣 + 𝑀𝑉

148
Figure 5.3

Elastic Collision between Point Mass and Uniform Rod

(2) Which point should we choose as the origin to produce zero torque on the rod? What

can you say about the angular momentum of the rod (without the point mass) about

this point? Thus, show that the final linear speed of the rod (𝑉) is related to its length

(ℓ), its final angular speed (𝜔) and the location of the point of collision (𝑥) by 𝑉 =

𝜔ℓ2
. [1 mark]
12𝑥

Taking the point of collision would produce zero torque on the rod as the force exerted by

the point mass acts exactly on the point of reference. Therefore, about this point, the angular

momentum of the rod is conserved. Applying the principle of conservation of angular

momentum,

1
0 = 𝑀𝑉𝑥 − 𝑀ℓ2 𝜔
12

𝜔ℓ2
𝑉=
12𝑥

149
(3) Since the collision is elastic, how would you relate the quantities described in the

question (𝑣0 , 𝑣, 𝑉, 𝜔, 𝑚, 𝑀, ℓ)? [1 mark]

The conservation of energy can be applied as the collision is elastic.

1 1 1 1 1
𝑚𝑣0 2 = 𝑚𝑣 2 + 𝑀𝑉 2 + ( 𝑀ℓ2 ) 𝜔2
2 2 2 2 12

(4) From (1)-(3), show that the final angular speed of the rod (𝜔) as a function of the

location of the point of collision (𝑥), the length of the rod (ℓ), the initial speed of the

024𝑚𝑣 𝑥
point mass (𝑣0 ) and their mass (𝑚 and 𝑀) is 𝜔 = (𝑀+𝑚)ℓ2 +12𝑚𝑥 2. [3 marks]

This question can be approached using algebra, based on the equations they have already had

in the previous parts. Combining the equations from conservation laws,

2 2
1 2
1 𝑀𝜔ℓ2 1 𝜔ℓ2 1 1
𝑚𝑣0 = 𝑚 (𝑣0 − ) + 𝑀( ) + ( 𝑀ℓ2 ) 𝜔2
2 2 12𝑚𝑥 2 12𝑥 2 12

Multiply both sides by 288𝑚𝑥 2 ,

144𝑚2 𝑣0 2 𝑥 2 = (12𝑚𝑣0 𝑥 − 𝑀𝜔ℓ2 )2 + 𝑀𝑚𝜔2 ℓ4 + 12𝑀𝑚ℓ2 𝑥 2 𝜔2

144𝑚2 𝑣0 2 𝑥 2 = 144𝑚2 𝑣0 2 𝑥 2 − 24𝑀𝑚𝑣0 𝜔ℓ2 𝑥 + 𝑀2 𝜔2 ℓ4 + 𝑀𝑚𝜔2 ℓ4 + 12𝑀𝑚ℓ2 𝑥 2 𝜔2

24𝑚𝑣0 𝑥
𝜔=
(𝑀 + 𝑚)ℓ2 + 12𝑚𝑥 2

(5) At which point should the point mass hit the rod to produce the largest angular speed

of the rod after the collision? Comment on your answer. [4 marks]

Using differentiation to maximise the angular speed of the rod,

𝑑𝜔
= 0 → 24𝑚𝑣0 [(𝑀 + 𝑚)ℓ2 + 12𝑥 2 ] − 24𝑚𝑥𝑣0 (24𝑚𝑥) = 0
𝑑𝑥

150
 24𝑣0 (𝑀 + 𝑚)ℓ2 = 288𝑚𝑥 2 𝑣0

(𝑀 + 𝑚)ℓ2 = 12𝑚𝑥 2

ℓ 𝑀
𝑥=± √1 +
2√3 𝑚

we find that the point is generally not at the end of the rod and it depends on the ratio of the

masses of the point mass and the rod. However, this expression is only valid for

ℓ 𝑀 ℓ
√1 + ≤
2√3 𝑚 2

𝑀
𝑚≥
2

When the condition is not satisfied, then the rod must be hit at its end as no local maxima can

be found. Although it is not required by the question, one can also calculate the maximum

angular speed of the rod:

3𝑚 𝑣0
𝜔max = 2√
𝑚+𝑀 ℓ

and when this happens,

𝜔max ℓ2 𝑚
𝑉= = 𝑣
12𝑥 𝑚+𝑀 0

and from the principle of conservation of momentum,

𝑚
𝑣=𝑉= 𝑣
𝑚+𝑀 0

As mentioned, even with qualitative analysis, the same conclusion could be reached.

151
 One may argue that the marking scheme for an open-ended assessment may be

somewhat subjective and, hence, this initial instrument may not be robust for the analysis to

determine alternative conceptions. Nevertheless, the results analysis can provide some

insights into students’ thinking, expressed in the way the students approached each question.

5.3.2. Pilot Study

Data from 28 Year 6 students (Grade 12; 4 females; mean age ± 𝑆𝐷 = 18.32 ± .61

years) of NUS High School of Mathematics and Science were obtained. The students took

both sets of test, with and without mathematical questioning, as formative assessments as

part of the Year 6 (Grade 12) school physics module at different times. We have obtained the

school’s permission to analyse and publish the result.

After students learnt the topic of rotational mechanics, they were given 30 minutes to

complete the first set of test items, which used qualitative questioning. This was followed by

a 15-minutes break, when the teacher taught them unrelated materials to prevent discussion

among students regarding the test questions. The same group of students were then given

another 30 minutes to complete the second set of test items, which used mathematical

questioning.

5.3.3. Instrument Validation: Rigid Body Collision Assessment

The results of the pilot study provided us with insights on what kind of questions to

be constructed and the way the questions should be phrased. Similar to the study on

momentum/collision, after several rounds of discussions and feedback, multi-tier format was

preferred as compared to the free-response format used in the pilot study for consistency of

152
marking, analysis of answer and reason tiers, analysis of correlations, analysis of alternative

conceptions, as well as analysis of confidence ratings. It also provides for determination of

test reliability. Meanwhile, the excerpt of students’ mathematical thinking could be done via

the analysis of the fifth tier, thus emphasising the benefit of a 5-tier test instrument. It was

also felt that the mathematical expression tier needs to explicitly ask the students to elaborate

on their answer and/or reason mathematically, in order to excerpt more insights on their

mathematical thinking. Failure to do so would result in fewer students using mathematical

thinking when approaching the questions. The higher order thinking questions were then

expanded into a number of questions in True/False format for the answer tier.

The amended version of the instrument was validated by 3 physics teachers teaching

various levels in NUS High School. Feedbacks given by them were used to further refine the

instrument, with some comments incorporated in revising certain questions, including the

distracters for the answer and reason tiers. In the validation form, nine items in the Yes/No

(Y/N) format were given in a checklist to ensure that the instrument was targeted correctly

for the specific group of participants, i.e. Years 5 and 6 (Grades 11 and 12), and the specific

topic tested was already taught to the participants (Table 5.3). Moreover, it was to our interest

that any ambiguity should be prevented by making the questions as clear as possible and free

of grammatical errors.

153
Table 5.3

Checklist for Validation of Rigid Body Collision Assessment by Physics Teachers (with

Responses)

Validator Validator Validator

No. Item
1 2 3

The questions are related to the topic of momentum

1 Y Y Y
and collision in the Year 5 and 6 Physics syllabus.

The required concepts have been taught in the Year

2 Y Y Y
5 and 6 Physics core modules.

The questions, answers and reasons are free from

3 Y Y N
grammatical errors.

There are an adequate number of questions to test

4 students’ understanding of the concept of rotational Y Y Y
dynamics.

The questions in the diagnostic instrument are clear

5 Y Y N
and comprehensible.

The multi-tier format used in the diagnostic

6 instrument is effective in probing students’ Y Y Y
understanding of the topic.

The correct responses indicated in the Answer and

7 Y Y Y
Reason tiers are acceptable.

Each tier has only one correct response, with the rest
8 being misconceptions (alternative conceptions) or Y Y N
incorrect responses.

Overall, the diagnostic instrument can provide

9 Y Y Y
useful information for the research study.

154
Some comments given by the validators are listen in Table 5.4. The comments were also used

to further refine the instrument.

Table 5.4

Comments by Validators and Actions Taken for Refinement of Rigid Body Collision

Assessment

Comment
No. Regarding Comment Action Taken
by

The statement suggested by the

validator would be a correct
statement if the student were to
choose False in the answer tier.
Rotation without linear
Validator However, the reason tier should
1 Q1 acceleration should be a key
1 provide the “reason” of why the
distracter.
statement in the question is true or
false instead of some contradicting
statements. No amendments were
made for this question.

The correct option gives the

answer away for students who may
not have considered 𝐼 (moment of
inertia) initially. Possible to
change the wording?
You may want to catch students
These are very good suggestions from
who use only qualitative
Validator the validator. The question and
2 Q3 reasoning, i.e. one physical
1 options have been amended
quantity goes up and another goes
accordingly.
down resulting no change in
angular acceleration.
Also, you may want to catch
students who would see the effect
on 𝐼 (moment of inertia) but not 𝜏
(torque).

This is not a correct reasoning as it

Can student simply say “since the
Validator does not give a specific condition for
3 Q10 rod is moving, the ball (particle)
1 the angular velocity of the rod to be
also has to move”?
maximum.

155
 Students’ mathematical thinking can
Expect answer based on specific
Validator lead them into several ways to obtain
4 Q12 counterexample cases with values
1 the correct answer and/or reason.
substituted.
Substitution is also acceptable.

This is a good suggestion from the

Validator
5 Q13-15 Consider drawing the pivot in. validator. The picture has been
1
redrawn for better clarity.

How should the students do that

(show the physics concepts they
use to approach every single The instructions have been amended
question)? Do they need to write to make it very clear to the students
formula? Mention which topic? on what is expected. Instead of
Validator
6 Instructions showing physics concepts, students
2 “… confidence level for your are expected to “provide
answer and your explanation” mathematical workings in relation to
should be written as “… their responses for each question”.
confidence level for each answer
and explanation”.

(The diagram is confusing.)

Validator You may want to show (a better) All diagrams have been redrawn to
7 Q1 diagram to ensure students
2 ensure better clarity.
understand how it (the rod) is
placed.

“Very Unconfident” and No amendments were made to the

“Unconfident” should be written confidence ratings as they were
as “Not Confident at All” and “Not adopted from previous studies (Hoe
Validator Whole
8 Very Confident”, respectively. & Subramaniam, 2016; Saat et al.,
2 paper
Unconfident theoretically exists 2016; Sreenivasulu & Subramaniam,
(in dictionary), but is almost never 2014, 2013; Caleon & Subramaniam,
used. 2010a).

Mathematical elaboration is expected

to be seen in the mathematical
expression tier. However, sometimes
Validator Are you expecting working (in the mathematical statement can come in
9 Q1
2 mathematical expression tier)? words too, e.g. A is higher than B, A
and B move with equal speed, etc.
These are also acceptable as correct
mathematical explanation.

Option D should be written as

Validator The wording has been changed
10 Q2 “The force produces a smaller
2 accordingly.
amount of torque.”

“The rod will now experience The sentence has been modified to
Validator
11 Q3 bigger angular acceleration” “The rod will now rotate with a larger
2
should be written as “The rod will angular acceleration”.

156
 now experience a greater angular The rest of the words have been
acceleration”. modified accordingly.
“… twice longer…” should be
written as “… twice as long…”
“… bigger…” should be written as
“… greater…”

Validator Label the centre of the rod in the All diagrams have been redrawn for
12 Q4-Q8
2 diagram. better clarity.

Validator Q8 has been modified to a True/False

13 Q8 No MCQ reasons?
2 question for consistency.

There are minor grammatical

errors.
Diagrams could be better
represented. All the suggested grammatical
corrections and amendment in the
Validator Whole Need to indicate in the instructions
14 instructions have been implemented.
3 paper to choose only one correct answer All diagrams have been redrawn for
and reason. better clarity.
Put in the instructions to circle the
confidence level for answer and
reason.

It looks like the rod is standing

Validator perpendicular to a surface rather All figures have been redrawn and
15 Q1 than lies on a horizontal table.
3 labelled for better clarity.
Label the figure.

Put (True/False) beside each

question number to indicate the This is a very good suggestion from
Validator Whole
16 type of question as the question the validator. It has been
3 paper
looks like a statement rather than a implemented for the whole paper.
question.

The purpose of the fifth tier is to

excerpt students’ mathematical
Would just writing the formula be
Validator thinking. This mathematical thinking
17 Q2 accepted (for the mathematical
3 can be observed in the form of
expression tier)?
formula or words. Hence, either form
is acceptable.

“… twice longer…” should be

Validator written as “… twice as long…” The options have been amended
18 Q3
3 “… independent on…” should be accordingly.
written as “… independent of…”

Validator Q4-8, Q13- Label the centre of the rod in the All diagrams have been redrawn and
19
3 15 diagram and label the figure. labelled for better clarity.

157
 All options are similar, except for No amendments were made as this
Validator
20 Q4-6 option C. You might need was done intentionally to test the
3
differentiated distracters? students on conservation laws.

In options D and E for the reason

Validator The two options have been rephrased
21 Q9 tier, it is not clear what “this”
3 for better clarity.
refers to.

Four out of five options in the The options have been rephrased such
Validator
22 Q11 reason tier are for True answer. that they do not suggest any specific
3
You may need to diversify? answer.

The final version of the instrument was named the Rigid Body Collision Assessment.

It comprises of 16 multiple-choice questions in 5-tier format, with all questions having equal

weightage. The mathematical expression tier (fifth tier) is targeted to see how much of

mathematical thinking students were able to perform while applying their conceptual

understanding in approaching each question. All questions are in True/False format for the

answer tier and have five distracters for the reason tier, with one being “other reason”.

Participants have the option to choose “other reason” if they disagree with the options

provided in the reason tier. Confidence ratings were then added to the answer tier and reason

tier. As previously done by Sreenivasulu & Subramaniam (2013, 2014) and Caleon &

Subramaniam (2010a), a six-point scale was used from “Just guessing” to “Absolutely

confident” as it seems suitable for our study. The complete set of assessment items can be

found in Appendix 4.

A sample question in 5-tier format in our main study instrument in rotational

mechanics is shown in Figure 5.4.

158
Figure 5.4

Sample Question in the Rigid Body Collision Assessment 4

Question 12 (True/False)
Consider the following collision. A particle of mass 𝑚 moves to the right with velocity 𝑢. It then collides
elastically with a uniform rigid rod of mass 𝑀 and length ℓ that lies on a smooth horizontal table. The particle
hits the rod perpendicularly at a distance 𝑥 from its centre. Now, the rod is pivoted at its centre.
𝑢

Top view:
𝑢

pivot at the centre

of rod

After the collision, the particle moves with velocity 𝑣 horizontally and the rod rotates about its centre with
1
angular velocity 𝜔. The moment of inertia of the rod about its centre is 12 𝑀ℓ2 .

The total linear momentum of the system of particle and rod is conserved.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Absolutely
Just guessing Unconfident Confident Very confident
Answer unconfident confident

Reason:
(A) The collision is elastic.
(B) The system is isolated.
(C) The external force acting on the system does not produce a torque.
(D) The rod rotates after the collision, so the linear momentum is transferred/transformed to angular
momentum.
(E) Linear momentum is always conserved in any case.
(F) Other reason:

Confidence 1 2 3 4 5 6
Rating for Very Absolutely
Just guessing Unconfident Confident Very confident
Reason unconfident confident

Express your answer and/or reason mathematically, and use texts for clarifications/explanations:

4
The space presented for mathematical expression is much more than that shown above.

159
5.3.4. Main Study Participants

We selected a total of 126 participants (21 females; mean age ± 𝑆𝐷 = 17.16 ± 1.12)

from the NUS High School of Mathematics and Science in Singapore, where 99 students are

from Years 5-6 (Grades 11-12) and 27 are accelerated students from Years 2-4 (Grades 8-

10). These levels were chosen as students would normally learn the topic of rotational

mechanics, including rigid body collision, in Year 5 (Grade 11), where the accelerated

students would have learnt the content in advance. Consent to participate in the study was

sought a month before the test administration. The students were informed of the activity

they would be engaged in and that only summarised findings would be reported. They were

also aware that personal particulars would not be identifiable, and that only the researchers

would have access to all the collected data.

As some of these participants overlapped with the participants of the study on

momentum/collision, it was necessary for these participants to be taken from a different pool

as compared to the student participants for the FCI, MBT and CSEM tests, so as to avoid

respondent fatigue. However, as mentioned before, the results of the FCI, MBT and CSEM

tests was consistent as the instruments were tested on four different batches of students in

NUS High School, thus enabling us to generalise the results to the entire population of the

school. Thus, from the research point of view, the participants for this phase of the study

could be classified as high-performing students.

5.4. Interviews

Although the excerpting of students’ mathematical thinking was done via the analysis

of the fifth tier, the results had to be triangulated with interviews. This would provide a

160
stronger argument through qualitative analysis of the transcript. After the administration of

the assessments, a few students were randomly selected based on their performance in the

fifth tier for some questions: students who were awarded 0.0 marks, 0.5 marks, 1.0 mark as

well as 1.5 marks for the mathematical expression tier in those questions. Eight students were

selected from the study on rotational mechanics. Identical method was adopted as for the

studies on momentum/collision. The interviews were done about one month after the

administration of the test instruments. All interview participants provided written informed

consent and assent in compliance with a protocol approved by the Nanyang Technological

University (NTU)’s Institutional Review Board.

The purpose of the interviews was to understand more about the participants’

mathematical thinking in approaching the questions; this was explained clearly to them at the

start of the interview. Participants were then asked to elaborate on their line of mathematical

thinking that was expressed in the fifth tier. After the elaboration, participants with partial

marks, i.e. 0.5 marks or 1.0 mark, were probed by the interviewer to see if they could correct

or complete their thinking to provide a comprehensive explanation to the required approach.

All interviewees who eventually managed to come to the full analysis at the interview were

then probed even further by the interviewer using an extension to the question. This enabled

us to see if students could perform an extended analysis of the physics problem after

acquiring the correct mathematical thinking. However, participants with 0.0 marks in the fifth

tier were not asked further questions after the elaboration of their mathematical thinking.

This is because these interviewees were expected to provide an irrelevant approach in the

fifth tier.

161
 The transcripts of the interviews were then analysed and used as support for the

students’ ability/inability to perform mathematical thinking in physics. The full interview

protocols for both studies are provided in Appendices 5 and 6.

5.5. Teaching Interventions in Promoting Mathematical Thinking

This thesis study explores the issue of students’ difficulty in performing mathematical

thinking in the domain of physics. The topics selected for this study are momentum/collision

and rotational mechanics. These topics were chosen as the concepts involved were deemed

to be somewhat simple (they were covered in high school physics syllabus), yet various

higher order thinking questions could be constructed and explored, thus challenging students’

thoughts and intuition based on their daily experiences. As such, the participants for the first

study were expected to have learnt the topic of momentum/collision prior to this study, and

to have understood the concepts of momentum, impulse, elastic and inelastic collisions. In

contrast, the participants for the second study were expected to have learnt the topic of

rotational mechanics prior to participating in the study, and to have understood the concepts

of moment of inertia, torque, angular momentum, rotational kinetic energy and rigid body

collision. However, as mathematical thinking was still short in students’ approaches and

alternative conceptions were still expected to occur, as shown in the excerpts of the test

results, we would like to explore a teaching strategy to promote mathematical thinking that

would lead to students’ improvement in ability to perform mathematical thinking. As an

additional benefit, conceptual changes were also expected to take place.

Thus, for the purpose of this research study, a teaching intervention that strongly

promotes mathematical thinking in the process of conceptual learning, problem-solving as

162
well as inquiry, is needed. In this intervention, mathematics is not just seen as a problem-

solving tool, but also as a thinking tool. The framework of this newly-explored teaching

intervention strategy will be elaborated in this section.

5.5.1. Pedagogical Framework

Stacey (2006) elaborated on the framework for mathematical thinking and its

importance in the study of mathematics. However, no studies have reported the importance

of mathematical thinking in physics so far. Physical situation, modelled mathematically, is

indeed a unique problem that bridges the two subjects. Regardless of whether students have

learnt the concepts or the mathematics involved, mathematical thinking does not come

naturally, as will be elaborated in the results of this thesis study. However, students’ ability

to perform mathematical thinking was expected to correlate with their performance in physics

assessment comprising higher order thinking questions.

Figure 5.5

Framework of the Newly-Explored Teaching Intervention Strategy

Arising from this, we adopted some approaches from other pedagogies (Allmond &

Makar, 2010; Boaler, 1997; Schoenfeld, 1991) to construct suitable strategies for the teaching

intervention. This newly-explored strategy comprises four main stages, as described in

163
Figure 5.5, and was used for the teaching intervention in the two main studies. As the student

participants had learnt the concepts before, the teaching intervention was framed as a

revision, yet it was done in a unique approach. The rationale for this is elaborated below.

(1) Understanding Physical Context of the Problem

Context problems are defined as problem situations that are experientially real to the

student (Gravemeijer & Doorman, 1999). In other words, though the problems are abstract

and do not really happen (or have not happened) in life, the hypothetical situation in the

problem can be thought of as occurring realistically or physically with more than one aspect

being considered. Most physics problems are context problems as they usually contain

physical meaning in real life scenarios. Without a physical context, encouraging students to

think mathematically is no different from solving a mathematics problem.

(2) Mathematising the Problem

Mathematising is defined as a process of thinking about and making sense of the

world with a mathematical perspective. Using mathematical lens here involves seeing the

physical context mathematically, and exploring mathematical methods and analyses to

approach the problem. Although the physics problem may look simple, one might think that

using “trained” scientific intuition is enough to solve the problem. Nevertheless, once

students are used to performing mathematical thinking, they would start to realise that using

intuition alone may not encapsulate all the considerations, and that they have to explore all

possible mathematical tools and knowledge they know or even exhaust these to mathematise

the problem and create a more comprehensive solution.

164
 One of the challenges in encouraging students to perform mathematical thinking is

that common school problems have purposely removed multiple layers of thinking in the

problem, in order to make it more straightforward and doable for the students. Multiple layers

of thinking are seen by students to be a factor that increases the difficulty level of the

problems, and by teachers to be a complicating factor in the marking scheme. Thus, the

questions faced by students are generally straightforward, though they involve mathematics,

and students could easily use the list of formulas – given or memorised – to solve the

problems. However, this type of problems does not really encourage mathematical thinking

as little mathematisation is required.

In looking at a physics problem, mathematising can still be a major issue even when

students have, or claim to have understood the concept. This is because the students are not

used to reading and analysing a physical situation mathematically, as the habit of performing

mathematical thinking is not really developed. Having “trained” scientific intuition is

important, but when intuition fails to explain, mathematics should be used to analyse.

(3) Exploring Mathematical Strategies

Right after mathematising a physics problem, mathematical strategies involving any

mathematical tools are left to the students. On the contrary, common school problems tend

to tell students to perform a certain mathematical strategy. In “One-Dimensional Motion with

Constant Velocity”, for instance, when it comes to the mathematics, teachers tend to directly
𝑠 𝑠
tell students to memorise 𝑣 = , 𝑠 = 𝑣𝑡, 𝑡 = , and the students are expected to be able to
𝑡 𝑣

understand the concept and do all the related problems. From the constructivist point of view,

165
we should note that any physics problem has a certain correct answer. However, there is

obviously more than one approach or strategy to obtain that answer. The teaching

intervention we are exploring is different from the traditional teaching in that students are

expected to construct mathematical explanations. Students are then provided opportunities to

explain their mathematical thinking processes as part of their solutions, and it is this

presentation and discussion of ideas that provide the foundation for conceptual understanding

in mathematics (Chapko & Buchko, 2004).

(4) Understanding Physical Meaning

Right after performing thorough mathematical thinking, the results of the thinking

would be meaningless if they are not translated back to the conceptual reasoning. A physics

student is expected to be “fluent” in switching from physics concept to the inherent

mathematics, and vice versa, as these two form the backbone of physics conceptual

understanding. Some of the mathematical results, for instance, need to be rejected as they

don’t make sense physically. As an example, a negative value for mass needs to be rejected

as it does not conform to its physical meaning. In interpreting the mathematical results,

students need to articulate it in the language that conforms to the previously-learnt physics

concepts.

All the above aspects promote the framework to elaborate on the importance of this

newly-explored teaching intervention which is aimed at the development of deeper

conceptual understanding as opposed to procedural understanding in traditional teaching of

166
physics. Its prolonged effect is expected to be beneficial as students would be able to apply

mathematical thinking and perform deeper analysis in different situations.

5.5.2. Examples in Conceptual Teaching and Problem-Solving Questions

Teaching by inquiry has been quite popular in physics education. Previous studies

have shown that applying the inquiry process in teaching and learning would have positive

impact in terms of knowledge retention and logical reasoning (Kogan & Laursen, 2014;

Richardson & Liang, 2008; Kazemi & Franke, 2004; Kim, 2006; Witt & Ulmer, 2010).

Incorporating it to this newly-explored teaching intervention would mean that the guiding

inquiry questions need to be phrased mathematically so as to encourage students to perform

logical mathematical thinking. This can induce an additional strength in the physics class.

Some examples of how mathematical guiding inquiry questions can be incorporated

in teaching physics concepts in my physics class, in contrast with using qualitative inquiry,

are given below (Table 5.5).

Table 5.5

Examples of Qualitative vs. Mathematical Guiding Inquiry Questions in Physics

Conceptual Teaching

Example of Qualitative Guiding Example of Mathematical Guiding

Example of Physics
Inquiry Questions and Expected Inquiry Questions and Expected
Concepts
Responses Responses

In a rigid body collision, a Question: Why do you think you can Question: Since the relation between
suitable reference point or cannot apply the conservation of the external torque and system’s
must be taken such that the angular momentum in this case? angular momentum is ∑ ⃗𝛕⃗ = 𝑑𝐋 ⃗ /𝑑𝑡,
conservation of angular ⃗
why do you think 𝐋 is or is not
momentum can be applied to constant in this case?

167
 relate the physical Expected response: We can apply it Expected response: Yes 𝐋 ⃗ is
parameters before and after as there is no net external torque constant as the reference point is
the collision. acting on the system about the taken such that ∑ ⃗𝛕⃗ = ⃗𝟎⃗ in this case.
reference point.

Question: Why do you think you can Question: In this collision, how do
or cannot apply the conservation of you relate the masses of the two point
energy in this collision? masses and their velocities before
In an elastic collision
and after the collision?
between two point masses,
Expected response: As the collision Expected response: As the collision
there is no energy loss to the
is elastic, there is no energy loss to is elastic,
surrounding.
the surroundings. Hence, the 1 1 1 1
conservation of energy can be 𝑚 𝑢 2 + 𝑚2 𝑢22 = 𝑚1𝑣1 2 + 𝑚2 𝑣2 2
2 1 1 2 2 2
applied.

Question: What will happen to a Question: The moment of inertia of

rod’s moment of inertia if its length a rod of mass 𝑚 and length ℓ is
is doubled symmetrically? 1
The moment of inertia is the 𝑚ℓ2 . What will happen to the
12
“laziness” of an object to rod’s moment of inertia if its length
rotate about an axis of is doubled symmetrically?
rotation. It depends on the Expected response: It will increase Expected response: As the mass of
mass and the linear as it will require bigger torque to the rod will also double, its moment
dimension of the body. rotate the rod. of inertia will increase by a factor of
2 × 22 = 8. Hence, it will be 8 times
harder to rotate the rod.

The above examples show us how a simple physics concept can be taught by

mathematising the inquiry questions. The kind of responses by students is expected to mirror

the expression used to describe the guiding inquiry questions. If mathematics is used, they

would more likely use mathematics to interpret the problem, digest the context, analyse the

situation, solve the equations, and articulate the results. Mathematising the teacher-

questioning is expected to help the students develop the habit of performing mathematical

thinking.

What about incorporating problems that encourage mathematical thinking? Prior to

this thesis study, the newly-explored teaching intervention was tried out in the Physics

Olympiad class involving a very small number of Years 3 and 4 (Grades 8 and 9) students.

168
The following elaboration of class observations provides an idea on how mathematising the

teacher-questioning can be done in class. The students involved had just learnt the concept

of vectors and kinematics.

Example of Practice Question:

Two particles start moving from the positions given in Figure 5.6 with constant velocities of

equal magnitude. What will be the shortest distance between the two particles as compared

to the initial distance between them?

Figure 5.6

Two Particles Moving with Constant Velocities

It was interesting to see the responses given by two different classes as two different

pedagogies were used. To the first class, the concept of vectors and kinematics were taught

traditionally. To approach the question, I asked the students the following questions to guide

them:

• Do you think the two particles move closer or farther apart from each other as time

goes by?

169
• How is the motion of one particle as viewed from the other?

Most of the students in this class could figure out intuitively that the two particles would

move closer to each other at first, but farther apart from each other after a while. But

following this, only one student in the class was able to perform the right analysis leading to

the right answer.

To the second class, a different approach was used as it was in my interest to

encourage them to perform mathematical thinking. I asked the students the following

questions to guide them:

• Let the distance between the two particles be given by 𝑟. Does 𝑟 increase or decrease

with time?

• What would be the velocity vector of particle 1 as viewed from particle 2?

Figure 5.7

The Motion of Particle 1 as Viewed from Particle 2

It is interesting to see that even though the guiding questions are essentially identical – they

only differ in “language” – most of the students in this class managed to obtain the correct

answer by first mathematising the problem and then analysing it mathematically. Most

170
students would describe the relative velocity vector of particle 1 with respect to particle 2

using vector subtraction. Realising that this vector makes an angle of 45° with the horizontal,

1
the shortest distance between the two particles is found to be of the initial distance (Figure
√2

5.7). It was also surprisingly noted that some students in this class came out with alternative

methods using other mathematical strategies. When asked to present their solutions to their

peers, it was interesting to see their thinking pattern and how they performed the

mathematical thinking. Firstly, by denoting the speeds of the particles as 𝑣 and the initial

distance between them as 𝑠0 , the distance between the two particles at any time 𝑡 is given by

𝑠 = √(𝑠0 − 𝑣𝑡)2 + (𝑣𝑡)2

(5.1)

Some alternative methods used by the students are elaborated as follows:

Alternative method 1:

Re-expressing equation (5.1),

𝑠 2 = 2𝑣 2 𝑡 2 − 2𝑠0 𝑣𝑡 + 𝑠0 2

2𝑣 2 𝑡 2 − 2𝑠0 𝑣𝑡 + (𝑠0 2 − 𝑠 2 ) = 0

(5.2)

Students who used this method argued that equation (5.2) is indeed a quadratic equation in

𝑡, and it must have real solutions for 𝑡 for this physical situation to be valid. Hence, its

discriminant must be non-negative.

4𝑠0 2 𝑣 2 − 8𝑣 2 (𝑠0 2 − 𝑠 2 ) ≥ 0

171
 𝑠0 2
𝑠2 ≥
2

In other words, the quadratic equation in (5.2) is valid only when the above inequality holds.

Therefore the minimum distance between the two particles is given by

𝑠0
𝑠min =
√2

(5.3)

Alternative method 2:

The rate of change of the distance between the two particles is given by

𝑑𝑠 −𝑣(𝑠0 − 𝑣𝑡) + 𝑣 2 𝑡 𝑣
= = (2𝑣𝑡 − 𝑠0 )
𝑑𝑡 √(𝑠0 − 𝑣𝑡)2 + (𝑣𝑡)2 √(𝑠0 − 𝑣𝑡)2 + (𝑣𝑡)2

(5.4)
𝑑𝑠
Equation (5.4) tells us that there is a certain range of 𝑡 that causes to be negative, i.e. the
𝑑𝑡

distance between the two particles decreases with time, and there is another range of 𝑡 that

𝑑𝑠
causes to be positive, i.e. the distance between the two particles increases with time. The
𝑑𝑡

𝑑𝑠 𝑠
minimum distance is thus achieved when 𝑑𝑡 = 0, i.e. 𝑡 = 2𝑣0 . Plugging this value into equation

(5.1) yields

𝑠0 2 𝑠0 2 𝑠0
𝑠 = √(𝑠0 − ) +( ) =
2 2 √2

(5.5)

172
 It was indeed surprising to see how the students in the second class were able to

explore several methods by mathematising the given problem. This is a simple example of

the advantage of the newly-explored teaching intervention over the traditional teaching and

assessment problems. The difference in the students’ performance is expected to be seen even

more significantly in physics problems involving higher order thinking. In other words, the

students in the second class were able to retain their mathematical thinking ability and use it

in different contexts.

The following example of a physics problem involving higher order thinking on

momentum and energy was given as an assessment to the two different classes mentioned

above after teaching the necessary concepts. It was expected that after going through different

pedagogies, students in the second class would be more inclined to performing mathematical

thinking as compared to students in the first class.

Example of a physics problem in the assessment:

A particle with a certain kinetic energy explodes into three identical fragments. To maximise

the kinetic energy of one fragment, how should the other two fragments move?

Examples of guiding questions given (each of them invokes different layer of thinking):

• What conservation laws apply in this case and why do you think we can or cannot

apply them?

• How does maximising the kinetic energy of one fragment affect its momentum?

173
• How will the final kinetic energy of the system be compared to its initial kinetic

energy?

• How would you maximise the kinetic energy of one fragment using all the

information you have gathered?

It was observed that all students in both classes had no difficulties in mentioning and applying

the principle of conservation of momentum at the beginning. They also understood that

maximising the energy of one fragment is equivalent to maximising its momentum. Prior to

this assignment, the students had established the concept that in the centre of mass frame,

after a fragmentation of a particle into three particles, the three velocity vectors must be co-

planar, and that the final kinetic energy of the system should be larger than its initial kinetic

energy. Almost all students analysed the conservation of momentum a bit further and

concluded that the other two fragments should move in the opposite direction to the first

fragment in order for the first fragment to attain the highest momentum and, hence, the

highest kinetic energy. Up to this point, students in both classes still displayed the same level

of thinking. However, it was noted that most students in the first class were more inclined to

use qualitative analysis and only one of them arrived at the correct answer. Below is his

argument:

Student 1: “I think I should move to the centre of mass frame of the other two

fragments that move in the same direction. In this frame, the two fragments move in

the opposite direction with equal speed. To maximise the energy of the other

174
 fragment, the speed of these two fragments must be zero in this frame. Hence, in the

original frame, these two fragments should move with equal velocity.”

In contrast, it was astounding to see many of the students in the second class

creatively articulate their thoughts using mathematical language and perform inquiry

thinking in the assessment, an indication of their representational fluency using mathematical

language. All students in this class had no difficulties expressing the following equations

displaying their conceptual understanding:

𝑚 𝑚 𝑚
𝑚𝑣 = 𝑣1 − 𝑣2 − 𝑣3 → 3𝑣 = 𝑣1 − 𝑣2 − 𝑣3
3 3 3

(5.6)

where 𝑚 is the mass of the initial particle; 𝑣 is its velocity before the explosion; 𝑣1 , 𝑣2 , and

𝑣3 are the velocities of each fragment, respectively, after the explosion; and

1 1 1 1 2𝑈
𝑚𝑣 2 + 𝑈 = 𝑚𝑣1 2 + 𝑚𝑣2 2 + 𝑚𝑣3 2 → 𝑣 2 = 𝑣1 2 + 𝑣2 2 + 𝑣3 2 −
2 2 2 2 𝑚

(5.7)

where 𝑈 is the stored potential energy that is released during the explosion. The following

analyses were gathered from the students. Unfortunately, one of them failed to arrive at the

correct answer.

Student 2: “There are too many unknowns and we only have two equations.

Therefore, it is impossible to perform further analysis. The only thing we can

conclude is the other two fragments should move in the opposite direction to the first

fragment.” (Student 2 failed to arrive at the correct answer.)

175
Student 3: “It looks to me that the problem can be simplified to 𝑣1 − 𝑣2 − 𝑣3 =

𝐴 and 𝑣1 2 + 𝑣2 2 + 𝑣3 2 = 𝐵 (from equations (5.6) and (5.7)), where 𝐴 and 𝐵 are

some constants. I can apply Cauchy-Schwarz inequality (Steele, 2004) and obtain

(𝑣2 + 𝑣3 )2 ≤ 2(𝑣2 2 + 𝑣3 2 )

(𝑣1 − 𝐴)2 + 2𝑣1 2 ≤ 2𝐵

(5.8)

Maximising the kinetic energy of the first fragment is the same as taking equal sign

in the above inequality (5.8). From Cauchy-Schwarz inequality, that can only happen

when 𝑣2 = 𝑣3 . Hence, the other two fragments must move with identical velocity, and

opposite to the direction of the first segment.”

Student 4: “I can substitute 𝑣3 from the first equation (equation (5.6)) to the

second equation (equation (5.7)) and obtain a quadratic equation in terms of 𝑣2 :

2𝑈
𝑣 2 = 𝑣1 2 + 𝑣2 2 + (𝑣1 2 + 𝑣2 2 + 9𝑣 2 − 2𝑣1 𝑣2 − 6𝑣𝑣1 − 6𝑣𝑣2 ) −
𝑚
𝑈
𝑣2 2 − (𝑣1 + 3𝑣)𝑣2 + (𝑣1 2 + 4𝑣 2 − 3𝑣𝑣1 − )=0
𝑚

(5.9)

But it is fascinating to see that I can get identical quadratic equation in terms of 𝑣3

if I were to substitute 𝑣2 instead:

𝑈
𝑣3 2 − (𝑣1 + 3𝑣)𝑣3 + (𝑣1 2 + 4𝑣 2 − 3𝑣𝑣1 − )=0
𝑚

(5.10)

176
So assuming I know the maximum value of 𝑣1 , I can just substitute its value to the

above two equations and solve them to get the values of 𝑣2 and 𝑣3 , respectively. Thus,

it is obvious that 𝑣2 = 𝑣3 for a specific value of 𝑣1 , which is its maximum value.

Hence, the other two fragments must move with identical velocity.”

Student 5: “To maximise 𝑣1 , I can simply differentiate both equations (equations

(5.6) and (5.7)) with respect to 𝑣2 , and take the first derivative of 𝑣1 with respect to

𝑑𝑣
𝑣2 to be zero, that is 𝑑𝑣1 =0.
2

𝑑𝑣1 𝑑𝑣3 𝑑𝑣3

0 = 2𝑣1 + 2𝑣2 + 2𝑣3 → 𝑣2 + 𝑣3 =0
𝑑𝑣2 𝑑𝑣2 𝑑𝑣2

(5.11)

𝑑𝑣1 𝑑𝑣3 𝑑𝑣3

0= −1− → = −1
𝑑𝑣2 𝑑𝑣2 𝑑𝑣2

(5.12)

From these two equations, 𝑣2 = 𝑣3 .”

Student 6: “I can substitute the 𝑣 in the first equation (equation (5.6)) to the

second equation (equation (5.7)) and simplify it to

4(𝑣1 2 + 𝑣2 2 + 𝑣3 2 ) + 𝑣1 𝑣2 + 𝑣1 𝑣3 − 𝑣2 𝑣3 = 𝐶

(5.13)

where 𝐶 is a constant. Then I can just assign some value to 𝑣2 and check what value

of 𝑣3 will maximise 𝑣1 . But the same argument works, I can assign some value to 𝑣3

177
 and check what value of 𝑣2 will maximise 𝑣1 . Since both equations are identical, I

can conclude that 𝑣2 = 𝑣3 .”

Looking at all students’ responses above, it is evident that performing mathematical

thinking is indeed beneficial in getting them to perform a deeper analysis of physics

problems. It is also interesting to observe that students tend to be consistent in their approach,

i.e. students who have developed the habit of performing mathematical thinking in learning

physics concepts would tend to continue doing it when analysing a physical context, and

students in a traditional physics class would usually rely more on qualitative thinking and

use mathematics only at the end. The latter group usually gets stuck when multiple layers of

thinking are required.

A more challenging question involves the particle fragmenting into three unequal

fragments, in which the analysis will be more sophisticated though the concepts used remain

the same. It should be noted that in these examples we have ignored the shape and the size

of the fragments, i.e. only mass matters. Other examples of higher order thinking physics

questions in the topic of momentum/collision are discussed in Ricardo & Lee (2015), which

is part of the theory explorations in this thesis study (Chapter 2). Using both intuition and

qualitative reasoning show great limitations in approaching this problem.

5.5.3. Main Studies

For the main study on momentum/collision, after the administration of the assessment

the participants were split into two groups: 61 students in the control group (12 females;

178
mean age ± 𝑆𝐷 = 17.25 ± 1.23) and 65 students in the experimental group (9 females; mean

age ± 𝑆𝐷 = 17.08 ± 1.00). Baseline test was done to ensure the mean pre-test scores of both

groups were not significantly different. A non-equivalent control group design was chosen,

in that the experimental group underwent a teaching intervention a few weeks after the pre-

test followed by a post-test one week after, while the control group only underwent the

intervention after the post-test for equity purposes. For the intervention, the students in the

experimental group were split into five classes and a lesson intervention was administered to

each class at different times. All the lesson interventions were mostly done during the school

term-break. On average, there were 11-12 students attending each lesson, and for discussion

purpose, each class was split into groups of 3-4 students. The number of students in each

intervention class was kept deliberately small to ensure high quality of the lesson. The lesson

intervention lasted about two hours, inclusive of their time to discuss the problems and

present the solutions.

The complete PowerPoint slides titled ‘Momentum and Collision’, used for the

teaching intervention, can be found in Appendix 7. It can clearly be observed that the teacher

strongly encourages students to perform mathematical thinking in almost every step. At the

beginning of the lesson, the main objective of the lesson was described clearly, which is to

promote mathematical thinking in physics. It was also made clear to students that this lesson

served as a revision. A sample of the slides used is shown in Figure 5.8.

179
Figure 5.8

Sample Slides Used in Teaching Intervention on Momentum/Collision

180
 The main crux of this lesson is in its adopted strategy where mathematical thinking

was greatly emphasised throughout the lesson. The concept asked may be simple, but it

should encourage students to perform mathematical thinking and, at the same time, translate

it back to qualitative reasoning. Firstly, the physical context of the problem needs to be

presented clearly to the students. Secondly, the teacher needs to encourage the students to

mathematise the presented problem. Thirdly, mathematical strategies need to be invoked;

these strategies may range from simple algebra to advanced mathematics such as calculus.

Lastly, students need to be able to interpret the results back to its physical context.

For the main study on rotational mechanics, one of the physics lessons during the

semester was used for the teaching intervention. The participants in this study were split into

two groups: 64 students in the control group (8 females; mean age ± 𝑆𝐷 = 16.47 ± .71) and

59 students in the experimental group (8 females; mean age ± 𝑆𝐷 = 16.56 ± 1.34). Mean pre-

test scores of both groups were compared. If they were not significantly different, then mean

post-test scores of control group and experimental group can be compared to see

effectiveness of intervention. The experimental group underwent a teaching intervention a

few weeks after the pre-test and took the post-test in the following week, while the teaching

intervention for the control group took place after the post-test.

The experimental group was split into three classes with slightly bigger average class

size of 21-22 students. Each class was then split into groups of 4-5 students for discussion

purposes. Intervention lesson for each class lasted about two hours, similar to the first study.

181
Students also had a chance to do group discussion to discuss the problems and present the

solutions afterwards.

Figure 5.9 presents a sample of the PowerPoint slides titled ‘Rotational Dynamics’

used for the teaching intervention for this study. Similar to the other study, every slide was

used by the teacher to invoke students’ mathematical thinking. Although the lesson was

framed as a revision, as it was clearly conveyed to the students, the participants were

encouraged to learn to perform mathematical thinking in the analysis of the problems.

Similarly, much thought was put in designing the lesson intervention for the main

study on rotational mechanics. The complete lesson plan can be found in Appendix 8.

Although the students might find the revision to be easy, the mathematical thinking skill

acquired would benefit them in the assessment. The problems given to the students in this

teaching intervention were also selected carefully in order to avoid “teaching to the test”.

This teaching intervention was expected to help enhance students’ mathematical thinking

skills in approaching physics problems in an assessment, thus they should perform better in

our constructed assessment. The results were also expected to be consistent for both studies.

182
Figure 5.9

Sample Slides Used in Teaching Intervention on Rotational Mechanics

183
5.5.4. Instrument Validation: Survey Form

For the data gathering, an instrument in the form of a survey was constructed. The

survey sought to obtain feedback from students on their views about the teaching

intervention. It was meant to measure the affective responses of the students in attending the

lesson, listening to the teacher and actively participating in the problem-solving and group

discussions. The survey form initially consisted of three sections: the first one seeks to

measure students’ cognitive scores, the second one is a free response section and the third

one is meant to capture students’ reflections on the lesson intervention.

In the first section of the initial version of the instrument, two subsections were

constructed: on the nature of the topic – which covers the importance of the use of

mathematical thinking in the topic discussed, and on the lesson intervention – which explores

how they felt about the use of mathematical thinking during the physics lesson. A list of ten

statements, using a five-point Likert scale, was drawn up based on ideas from the survey

literature and on the author’s experiences. Adopting the scales used in the previous studies,

the Likert-type response for each statement ranges from 1 = strongly disagree (SD) to 5 =

strongly agree (SA) (Foong & Lam, 1988; Caleon & Subramaniam, 2008). Considering the

young respondents of this study, we presumed that the positively-worded statements would

be more helpful in eliciting more valid or accurate responses from the participants

(Schriesheim & Hill, 1981). This is because previous studies indicated that young students

did not fully comprehend negatively-worded statements and did not convey true attitudes

when faced with such statements (Benson & Hocevar, 1985). As such, in the initial version

of the survey instrument, all ten items used positive phrasings.

184
 A free response section was included in the second section of the initial version of

the instrument. This section is not less important than the previous one as it can provide

further evidence of learning and affect that might not be captured by the cognitive test. Their

responses in this section could give us further insights into which part of the lesson or

teaching strategy they found helpful or enjoyable during the teaching intervention. In this

regard, two questions were constructed to prompt the students: (1) “What is it about

mechanics that makes it difficult to understand?” and (2) “Do you understand the term

‘mathematical thinking’? Please describe it shortly.” In the third section of the initial version

of the instrument, students were requested to provide written reflections about the lesson,

class dynamics, group discussions, as well as the key points they had learnt from the lesson

intervention. All the sections in the survey instrument would give information on the level of

interest stimulated by the teaching intervention.

The initial version of the survey form was sent for validation to another physics

teacher at NUS High School and a researcher at the National University of Singapore (NUS).

They provided written feedbacks and comments to further refine the survey instrument. Table

5.6 summarises the feedbacks and comments from the validators.

185
Table 5.6

Comments by Validators and Actions Taken for Refinement of Rigid Body Collision

Assessment

Comment
No. Regarding Comment Action Taken
by

The statement suggested by the

validator would be a correct
statement if the student were to
choose False in the answer tier.
The responses need to be arranged
Validator However, the reason tier should
1 Responses from the highest to the lowest to
1 provide the “reason” of why the
make it clear to the participants.
statement in the question is true or
false instead of some contradicting
statements. No amendments were
made for this question.

The subsection title has been

amended to emphasise on a specific
Inconsistent use of uppercase topic of mechanics that the students
(Subsection A) sit for.
What does it mean by “use of The statements in items #2, #3 and #4
mathematics”? (Item #2) have been rewritten to ask of the
Validator importance of mathematical analysis
2 Section 1 Confusing statement (Item #3)
1 in the study of mechanics, whether
Need to restructure the statement mathematical analysis helps the
(Item #4) students to understand the concept
Need to explain what the word better, as well as whether they want
“this” refers to (Item #6) to explore more problems in
mechanics involving the use of
mathematical analysis.

If the students have different The question in Section 2 has been

Section 1 concept of mathematical thinking, amended and the understanding of
Item #10 will it invalidate their responses in the term “mathematical thinking” is
Validator
3 and Section Section 1? Would it be better to to be emphasised during the lesson
1
2 Question standardise what mathematical intervention. This will ensure all
#2 thinking means before presenting students have the same understanding
the items in Section 1? of mathematical thinking.

The question has been rephrased to

Section 2
Validators address a specific topic in mechanics
4 Question The question is ambiguous.
1, 2 that the students sit for in the
#1
intervention.

186
 Section 1 Are the questions arranged in the Most of the items have been
Validator
5 Items #2- intended order? The order might rephrased and rearranged
2
#5 affect the responses. accordingly.

The statement has been rephrased so

Validator Section 1 The word “otherwise” is
6 as to emphasise on the usefulness of
2 Item #3 ambiguous and unclear.
mathematical thinking.

Suggest to change to “What do you

like/dislike about the use of This is a good suggestion from the
Validator
7 Section 3 mathematical thinking? How do validator. The question has been used
2
you think this problem-solving to rephrase the question in Section 2.
approach would benefit you?”

After taking into account all the comments from the validators, the survey instrument was

modified quite significantly. The final version of the survey form consisted of two sections,

one on the cognitive aspects and another based on free response. The first section contains

eleven items – five on the nature of the topic and six on the lesson intervention. Out of the

eleven items, only one of them was negatively worded on purpose. This negatively worded

statement asks the participants whether after the lesson intervention they would still prefer to

do the topic qualitatively without the use of mathematics – it was clearly phrased to prevent

students from not conveying their true feeling or view. The final version of the survey forms

can be found in Appendices 10 and 11.

For each study, the lesson intervention lasted about two hours as we did not want to

fatigue the young participants by adding another task to them. For this reason the number of

items in the first section was minimised while the third section on students’ reflections was

taken out from the survey form. One week after the lesson intervention, students in the

experimental group were asked to complete this survey, prior to taking the post-test.

Additionally, a handful of the students in the experimental group were randomly asked to

provide written reflections and commentaries on how they felt about the topic and lesson.

187
5.6. Data Analyses

5.6.1. Evaluating Suitability of Sample for Current Study

Data analyses were accomplished using R statistical software (R Core Team, 2015)

and MATLAB R2015A (Mathworks, Natick, MA). Relevant statistical tests were two-tailed

and the significance level was set at 𝑝 < .05.

Firstly, the descriptive statistics of the students’ performance as well as the

characteristics of the distribution of the scores were analysed. Secondly, assuming

homogeneity of variance, an initial MANOVA was conducted to examine the consistencies

for FCI, MBT, and CSEM test scores as independent variables, and batches as the dependent

variable. If the MANOVA results were statistically significant, separate ANOVAs were to

be conducted to test whether FCI, MBT and CSEM differ across batches. Thirdly, we

compared students’ test scores with other schools’ test scores – from various overseas high

schools and universities (scores taken from Hestenes et al., 1992; Maloney et al., 2001;

Savinainen & Scott, 2002; Reay et al., 2008; Ding et al., 2008).

Interviews were done at the end of this preliminary study to triangulate the findings.

Possible alternative conceptions might also be identified through the interviews as the

interview questions were based on the scores distribution, such that these were questions

which the participants scored the least. Two questions from FCI, seven from MBT and eleven

from CSEM were selected as interview questions. However, the analysis of the alternative

conceptions in the test instruments used (FCI, MBT and CSEM) was not done as the aim of

this test was simply to evaluate the suitability of the sample for our main studies.

188
5.6.2. Excerpting Students’ Mathematical Thinking in Main Studies

All the test papers were marked and the responses for each tier were recorded and

analysed in an Excel file, namely option selected for answer tier, confidence rating for answer

tier, option selected for reason tier, confidence rating for reason tier, and mathematical

expression of answer and/or reason. One mark was awarded for each correct response in the

answer tier, and another one mark for each correct response in the reason tier. As the format

of the mathematical expression tier is open-ended, we felt it was necessary to use a fine-

grained marking scheme, with 0.0 marks being the lowest and 1.5 marks being the highest.

Partial marks of 0.5 or 1.0 were respectively given to students who showed limited

understanding or some understanding in the articulation of their mathematical thinking.

Although the subjectivity in scoring this tier can be an issue, it can be minimised by having

this scoring scheme (Table 5.7) and consistency across all participants. The confidence

ratings were not scored but analysed separately.

Table 5.7

Marking scheme for Main Study Assessments

Answer and Reason tiers Score Mathematical expression tier

Wrong answer and wrong reason 0 Not possible to get score

Wrong answer and correct reason 0 Not possible to get score

Correct answer and wrong reason 1 Possible to get score

Correct answer and correct reason 2 Possible to get score

189
 Mathematical expression tier Score

No response 0.0

Irrelevant mathematical explanation 0.0

Mathematical explanation shows little understanding in the correct direction 0.5

Mathematical explanation shows some understanding in the correct direction 1.0

Mathematical explanation shows acceptable understanding 1.5

For wrong response in the answer tier, the marks for the reason tier and mathematical

expression tier for each question were not taken into consideration as we felt that though the

student might have had the correct reasoning and the presented mathematical explanation

was seemingly correct, these still did not lead them to the correct answer. The maximum

score for each question is therefore 3.5, a combination of scores in the answer, reason and

mathematical expression tiers.

For the Chain-Collision Assessment, the maximum total score for all 15 questions is

therefore 47.5, whereas for the Rigid Body Collision Assessment, the maximum total score

for all 16 questions is 56.0. For the ease of data presentation, percentages are used instead of

nominal scores. For the determination of FI and DI, the criterion used was the total scores

from all tiers arranged sequentially.

Table 5.8 presents the summary of all the formulas used for the data analyses, as

adopted from Sreenivasulu & Subramaniam (2014) and Caleon & Subramaniam (2010b).

190
Table 5.8

Summary of Formulas Used for Data Analyses

Quantity Abbreviation Formula

number of students answered correctly

Facility index FI
number of students attempted the question

% students (in the top 25%) answered correctly

Discrimination index DI
− % students (in the bottom 25%) answered correctly

total confidence for students who answered correctly

Confidence when correct CFC
number of students answered correctly

total confidence for students who answered incorrectly

Confidence when wrong CFW
number of students answered incorrectly

total confidence for all students attempted the question

Mean confidence CF
number of students attempted the question

Confidence CFC − CFW

CDQ
discrimination quotient standard deviation for CF
CF − 1
Confidence bias CB − FI
5

The percentage of students in the various scoring ranges was also determined. For assessing

the internal consistency of the assessment, the value of Cronbach Alpha was determined,

based on the combined cognitive scores of answer, reason and mathematical expression tiers.

Reliability of the confidence ratings for the assessment was also determined.

All the alternative conceptions and their strengths were identified based on incorrect

answerreason combinations selected by at least 10% of the sample (Tan et al., 2002; Caleon

& Subramaniam, 2010b). Lastly, several correlation tests were performed between students’

scores in the answer tier vs. reason tier, answer tier vs. mathematical expression tier, as well

as the students’ combined scores of answer and reason tiers vs. mathematical expression tier.

191
The results of these correlation tests would give us some idea on the association between

students’ performance in one tier and another tier. These analyses were done using MATLAB

R2015A (Mathworks, Natick, MA).

The main strength and substance of this thesis study are in the analyses of the fifth

tier, which is the novelty of this study as compared to the previous studies. The fifth tier,

which is the mathematical expression tier, was marked and analysed carefully as the data

extracted were foreseen to be very meaningful. Firstly, the average percentage of people who

attempted to express the mathematical formalism in the correct manner for each question was

calculated as that would give us an idea of how many students are actually able to perform

and express their relevant mathematical explanation. The average score for the mathematical

expression tier for each question was also analysed and contrasted against the alternative

conceptions identified for that question, if any. This allowed us to see how students’

mathematical thinking leads them to correct or wrong answers and reasons and to confirm

the existence of alternative conceptions.

The transcripts of the interviews was analysed to triangulate the above findings. The

interviewees were expected to be able to explain their mathematical thinking process to

approximately the same level as their conceptual understanding. In other words, students who

obtained full marks in the fifth tier should be able to articulate the full correct analysis of the

physics problem using mathematical thinking and even extend it for further questions asked

by the interviewer. Likewise, students who obtained partial marks in this tier should be able

to express their mathematical thinking in the correct direction but still short of some

understanding. The interviewer would then try to probe them using some mathematical

questioning to see if they could arrive at the full correct explanation and to respond correctly

192
to the extended question by the interviewer. In contrast, students with zero marks in the fifth

tier would not be probed further after elaborating on their explanation.

5.6.3. Evaluating Effectiveness of Teaching Intervention

The effectiveness of the teaching intervention was measured by analysing the post-

test performance for both the control and experimental groups. Similar data analyses were

performed on their performance as in the pre-test. The t-test comparison between the pre-test

and post-test scores was then performed for each tier and for each group. For the experimental

group, it was also important to evaluate students’ conceptual changes that occurred after the

teaching intervention. This was done through the comparison of percentages of students in

which the tendency of alternative conceptions were indicated.

Furthermore, qualitative and quantitative analyses were also performed following the

students’ reflections and feedback in the survey form. Data were drawn from the survey form

whereby students were asked to respond to two open-ended question: ‘What is it about the

topic of <momentum/collision or rotational mechanics> that makes it difficult to

understand?’ and ‘What do you like or dislike about the use of mathematical thinking on the

topic of <momentum/collision or rotational mechanics>?’. Data were coded so as to identify

the emergent categories using the grounded theory approach (Strauss & Corbin, 1998). In

this approach, the categories emerge naturally from the data through engagement and re-

engagement with the data. The students’ responses were grouped in categories and the

connections among them were then studied.

The specific steps adopted in this study for categorising students’ feedback are as

follows:

193
(1) The responses were transcribed verbatim from the survey forms.

(2) A preliminary read-through of the transcribed text was done in order to get some

insights on the possible categories that could emerge inductively from the range of

suggestions documented.

(3) The transcripts were carefully examined and checked for similarities and differences.

When a meaningful text fragment was identified, it was assigned to a particular

category that was given a conceptual label.

(4) Text fragments that are similar at the conceptual level or share semantic commonality

were grouped together to form a category.

(5) The text fragments that were preliminarily grouped together into a category were

interrogated repeatedly to ensure that their placement therein makes sense and, when

this is not the case, they were assigned to a different or new category.

(6) The differences and commonalities in each category were identified and verified

repeatedly against the data. Some categories that did not emerge in the previous stage

surfaced in this stage. As a result, new categories were formed. Each category has its

specific characteristics and dimensions.

(7) The process is considered saturated when no additional data could be mined from the

transcripts to fit the category.

The categorisation of students’ feedback would be a crucial element for identifying

ideas of what really works in the teaching intervention strategy.

194
 CHAPTER 6

RESULTS

This chapter reports on the results of the studies done in this research. The first part

focuses on evaluating suitability of our sample for the main studies, i.e. students of NUS

High School of Mathematics and Science in Singapore, by administering conceptual surveys

and performing a comparison test with other overseas high school and undergraduate

students. The second part would be a thorough analyses of our main study results. These

include the analysis of test statistics, test scores, correlations, confidence ratings, students’

mathematical thinking – as elaborated in the mathematical expression tier, interviews and

alternative conceptions harboured by them.

As explained in the study design, after the administration of the test, we would like

to see the effectiveness of engaging and promoting mathematical thinking as our teaching

intervention. The results would also be analysed from students’ reflections and feedbacks in

the survey form.

6.1. Evaluating Suitability of Sample for Current Study

6.1.1. Test Scores and Comparison between Batches for FCI, MBT and CSEM

We first sought to identify the characteristics of the score distribution for each

instrument through plots (Figure 6.1) and examining their descriptive statistics (Table 6.1).

Figure 6.1 shows that the performances of the students, especially for FCI, were somewhat

on the high side.

195
 An initial MANOVA was conducted to examine the consistencies for FCI, MBT, and

CSEM test scores as independent variables, and batches as the dependent variable.

Significant difference was found in test scores based on batches (𝐹 (3,186) = 4.32, 𝑝 =

.006, Wilks’ Lambda = .935). As the MANOVA result was significant, separate ANOVAs

were conducted to test whether FCI, MBT and CSEM scores differ across batches. FCI and

MBT scores were found to be consistent across batches (FCI: 𝐹 (1,216) = .612, 𝑝 = .435;

MBT: 𝐹(1,216) = 2.665, 𝑝 = .104) while differences were found across batches for CSEM

scores (CSEM: 𝐹(1,193) = 15.11, 𝑝 < .001).

Figure 6.1 also shows that the performance of the participants, especially in the FCI,

was highly left-skewed. As such, the non-parametric rank-based Kruskal-Wallis H test was

used to test whether there were differences in FCI test scores across batches. Consistent with

the ANOVA result, the Kruskal-Wallis H test showed no significant difference in FCI scores

across the three batches (𝜒 2 (3) = 4.874, 𝑝 = .181), with a mean rank for FCI score of

104.18 for Year 5 Batch 2015, 107.96 for Year 6 Batch 2015, 128.08 for Year 5 Batch 2016,

and 102.30 for Year 5 Batch 2017.

The numerical data including the average, standard error and skewness for four

different batches in NUS High School are shown in Table 6.1.

196
Figure 6.1

Distributions of Participants’ FCI, MBT and CSEM Test Scores

197
Table 6.1

Comparison of Test Scores across Batches of Students

Year 5 (Grade 11) Year 6 (Grade 12) Year 5 (Grade 11) Year 5 (Grade 11)
Test
Batch 2015 Batch 2015 Batch 2016 Batch 2017
Instrument
Value ± 𝑆𝐸 Value ± 𝑆𝐸 Value ± 𝑆𝐸 Value ± 𝑆𝐸
N: 65 N: 42 N: 48
N: 63
Average: Average: Average:
Average:
26.40 ± .52 27.88 ± .33 26.71 ± .42
26.67 ± .40
Skewness: Skewness: Skewness:
Skewness:
FCI −1.88 ± .30 −1.36 ± .37 −1.19 ± .30
−1.19 ± .30
Kurtosis: Kurtosis: Kurtosis:
Kurtosis:
3.82 ± .57 1.61 ± .72 −.09 ± .67
1.05 ± .60
Min: 10/30 Min: 21/30 Min: 16/30
Min: 16/30
Max: 30/30 Max: 30/30 Max: 30/30
Max: 30/30

N: 63 N: 65 N: 42 N: 48
Average: Average: Average: Average:
19.94 ± .47 19.11 ± .45 21.31 ± .46 20.44 ± .51
Skewness: Skewness: Skewness: Skewness:
MBT −.48 ± .30 −.26 ± .30 −.32 ± .37 −.48 ± .30
Kurtosis: Kurtosis: Kurtosis: Kurtosis:
−.35 ± .60 −.13 ± .57 −.50 ± .72 . 38 ± .67
Min: 10/26 Min: 8 /26 Min: 15/26 Min: 10/26
Max: 26/26 Max: 26/26 Max: 26/26 Max: 26/26

N: 58 N: 57 N: 34 N: 46
Average: Average: Average: Average:
21.98 ± .56 24.98 ± .52 25.91 ± .67 24.98 ± .60
Skewness: Skewness: Skewness: Skewness:
CSEM . 17 ± .31 −.19 ± .32 −.87 ± .40 . 17 ± .31
Kurtosis: Kurtosis: Kurtosis: Kurtosis:
−.04 ± .62 −.80 ± .62 . 19 ± .79 −1.36 ± .69
Min: 13/32 Min: 17/32 Min: 16/32 Min: 17/32
Max: 32/32 Max: 31/32 Max: 32/32 Max: 31/32

Abbreviations. FCI, Force Concept Inventory; MBT, Mechanics Baseline Test; CSEM,
Conceptual Survey in Electricity and Magnetism; SE, Standard Error

198
6.1.2. Comparing between Subject Pool from NUS High School of Mathematics and Science

and Students from Other Schools

We then compared these students’ test scores with other schools’ test scores – for

example, overseas high schools and universities (scores taken from Hestenes et al., 1992;

Maloney et al., 2001; Savinainen & Scott, 2002; Reay et al., 2008; Ding et al., 2008). We

found that the NUS High School students performed significantly better than other overseas

high school students and that none of the other student groups from various overseas

universities performed significantly better than the NUS High School students. In other

words, these students were able to score higher, if not comparable, to students in the other

schools (𝑡 < 3.14, 𝑝 < .003) (Figure 6.2).

From the distributions of the results, we could see that there was still room for

improvement in the students’ performance. However, as we could not rigorously deduce from

these results that there was a lack of conceptual understanding, we investigated the scores

qualitatively by interviewing the students.

199
Figure 6.2

FCI, MBT and CSEM Score Comparisons across Schools

Note. Students of the NUS High School of Mathematics and Science performed either better
than or comparable to other schools across the three different test instruments. The
independent t-test was always done between the batch with the lowest mean score and other
school. Abbreviations: SEM, Standard Error of the Mean; Hon, Honours; Reg, Regular; IB,
International Baccalaureate; AP, Advanced Placement).
* Independent t-test p-value < .01.

200
6.1.3. Interview Findings

Twenty questions with the lowest percentage of students who answered correctly

were selected as the topics for interviews with the students, i.e. 2 from FCI, 7 from MBT and

11 from CSEM (see Figure 6.3). We then randomly selected 32 students who had taken the

assessments and interviewed them regarding their answers on the 20 selected questions (Q3

and Q15 from the FCI; Q5, Q9, Q12, Q18, Q19, Q20 and Q22 from the MBT; and Q2, Q13,

Q14, Q18, Q20, Q21, Q24, Q27, Q29, Q31 and Q32 from the CSEM) to see their thinking

pattern and the way they approached the questions.

During the interviews, we confirmed the performances of the students by asking them

the concepts and approaches they used in answer of the selected questions. Almost all

participants interviewed were able to give the same correct answers as they did during the

assessment or even made correction pertaining to their mistakes with correct reasoning, too.

This provides evidence that the students were fully engaged during the assessment and that

their performances were not driven by chance. When queried on questions where many

students answered incorrectly, a large number of the interviewees were even able to point out

their own previous mistakes, indicating that their alternative conceptions might be spurious

ones. Interestingly, some of their mistakes were due to over-reliance on incorrect qualitative

reasoning during the assessment, but as they carefully performed the calculations during the

interviews, many of them managed to get the correct answers without showing any signs of

alternative conceptions.

201
Figure 6.3

Scores Distribution across Questions – FCI, MBT, CSEM

FORCE CONCEPT INVENTORY

1
Fraction of students who
answered correctly

0.8
0.6
0.4
0.2
0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29
Question

MECHANICS BASELINE TEST

1
Fraction of students who
answered correctly

0.8
0.6
0.4
0.2
0
1 3 5 7 9 11 13 15 17 19 21 23 25
Question

CONCEPTUAL SURVEY ON
ELECTRICITY AND MAGNETISM
1
Fraction of students who

0.8
answered correctly

0.6
0.4
0.2
0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31
Question

202
 For example, the following is a partial transcript of the conversation between the

interviewer and Student 7 (Year 5 or Grade 11). The student made a mistake in Q9 in the

MBT.

Student 7: “Do you mind if I use the calculator?”

Interviewer: “Sure, please.”

(Student 7 started drawing a diagram and pressing the calculator, and answered after

a few seconds.)

Student 7: “For this one, I choose option D.”

Interviewer: “You choose option D because…?”

Student 7: “Wait, wait! No, no, sorry! I should check this again.”

(Student 7 started explaining the concept of static friction and circular motion, and

then performed the calculation which led to the correct answer.)

Student 7: “So the answer is A.”

Another example is the following partial transcript of the conversation between the

interviewer and Student 8 (Year 6 or Grade 12). In the written assessment, student 8 answered

Q21 and Q22 in the MBT (which were reflected as Questions 51 and 52 in the assessment)

incorrectly.

203
 Student 8: “Too little information for Questions 51 and 52. Because we don’t

know how long they are being pushed and we don’t know the work done for each

path.”

Interviewer: “So what other information do you need?”

Student 8: “The duration.”

Interviewer: “Okay what information is given in the question?”

(Student 8 mentioned everything correctly without any sign of misunderstanding or

misinterpretation of the questions. But student 8 then realized that the missing

information could have been obtained by calculation.)

Student 8: “Oh! Let me calculate!”

(Student 8 performed some calculations.)

Student 8: “I used some kinematics equations to slowly derive and I found the

results: C (for Question 51) and B (for Question 52).”

Another cause of incorrectness of answer was the ambiguity of the questions that led

to misinterpretation. This could be attributed to the phrasing of the question or perceived

ambiguity in the given information. However, from the interview, they demonstrated

understanding of the concept and could explain it correctly.

The following partial transcript is an example of Student 9 (Year 6). Student 9

attempted to answer Q3 in the FCI that was ambiguous to him/her.

204
 Student 9: “I guess it will be A, because of gravity it will accelerate, but then as

it speeds up the air drag force increases and after a while, it will reach terminal

velocity”

Interviewer: “Ok, are you sure with your answer?”

Student 9: “Hmm... (long pause). Not really. “

Interviewer: “Not really sure, but you will still choose A?”

Student 9: “Yes.”

Interviewer: “Is it because all other options are wrong to you?”

Student 9: “Hmm.. Technically C might be correct. But, I guess as I said earlier,

I am not really sure whether there is air drag or to consider it or not. So I take it from

a realistic point of view.”

Interviewer: “So you are having a dilemma between A and C”

Student 9: “Yes.”

From this partial transcript, Student 9’s explanations on whether to choose options A or C

were correct, but she chose the incorrect answer due to misinterpretation of the question.

The scores analysis and interview results confirmed the suitability of our subject

sample for our research by showing that they could be categorised as high-performing

students. Specifically, the students of NUS High School were considered as high-performing

because they were found to perform significantly better than other overseas students of

comparable education level reported in the literature in the three commonly used instruments

to test physics conceptual understanding: FCI, MBT and CSEM. More significantly, the

performance of the students in our sample is comparable with undergraduate honours

205
students from some of the world’s best universities (Hestenes et al., 1992) and also with high

school students who were given the chance to improve their scores after certain interventions

(Hestenes & Wells, 1992). The high performance of our potential subject pool was also

shown to be mostly consistent across the four batches, thus enabling us to generalise the

result to the whole school population. This is important as our main studies would be using

different batches of students.

The abovementioned instruments – FCI, MBT and CSEM – have been used widely

in physics education researches, especially for the evaluation of students’ conceptual

understanding, determination of alternative conceptions and measuring the effectiveness of

teaching interventions in physics. The results in this section showed that these instruments

were indeed able to differentiate the students of the NUS High School from other students

on their conceptual understanding of physics concepts. However, the score distributions

suggested that the instruments explored might not be suitable to measure gains for the

population of high-performing students as they were highly-skewed to the left and do not test

higher order thinking. Thus, with other reasons elaborated in the previous chapters, there is

a need for assessment instruments of higher order thinking to be constructed.

6.2. Study on Momentum/Collision

In this section, we would like to analyse the students’ performance in the pilot study

as well as the main study on the topic of momentum/collision. In general, although the

students were able to perform well in physics modules, it was interesting to observe that these

high-performing students did not perform well in the constructed instrument. The findings in

206
this study showed that these high-performing students still had strong tendencies to harbour

alternative conceptions based on wrong intuition or incorrect qualitative analysis.

6.2.1. Pilot Study

The higher order thinking questions in the test instrument used in the pilot study were

the last two questions. Overall, the students were able to perform well for all the questions,

except for these two questions. For the first one, it was found that only 41.0% of the

participants managed to associate the most efficient kinetic energy transfer with elastic

collision, while the rest of the students assumed that all momentum should be transferred or

that all kinetic energy should be transferred, which was clearly an alternative conception. For

the second one, it was found that only 20.3% of the participants managed to reason out that

inserting a stationary block in between two blocks could increase the portion of kinetic

energy transferred. These low percentages clearly showed that an intervention could be done

for conceptual changes. It was also interesting that although some students expressed their

understanding mathematically, most preferred to apply their scientific intuition to approach

the problems. Due to the latter scenario, we found that it was indeed difficult to excerpt

students’ mathematical thinking from the test paper without telling them specifically to

express it during the administration of the test instrument.

The ability to perform and articulate mathematical thinking might not translate to or

lead to conceptual understanding and problem-solving ability. Below are some examples of

students who expressed mathematical thinking but failed to attain the correct answers for the

first higher order thinking problem in the pilot study.

207
 Student 10: “To transfer the most kinetic energy, all kinetic energy should be

transferred which means 𝑣𝑋,final = 0. [Then he displayed the calculation using the

conservation of momentum, indicating 100% momentum transfer, which is a clear

alternative conception.] This is how I would interpret ‘transferring most kinetic

energy’.”

Student 11: “In order to transfer maximum KE, 𝑣𝑋 = 0. [Then he displayed the

calculation using the conservation of energy, indicating 100% energy transfer,

though the calculation was wrong. This also indicates a clear alternative

conception.] Assume 𝑣𝑋 to be 0 m/s to simplify the equation.”

Presented below are some examples for the second higher order thinking questions in

the pilot study. Although these students did not get the correct answer, their mathematical

thinking could still be excerpted and analysed in relation to their conceptual understanding.

Student 12: “No. Momentum must still be conserved in the system across all

collisions. As a result, the maximum transfer is still the same, and will not increase.”

Student 13:

362 4×9×4×9
𝐾𝐸𝑖 = = = 9 × 9 × 2 = 162
4×2 4×2

Index s is for ‘suggestion’:

𝐾𝐸𝑓𝑠 = 𝐾𝐸𝑓𝑍 + 𝐾𝐸𝑓𝑋 + 𝐾𝐸𝑓𝑌

208
 ′ ′
𝐾𝐸𝑓 = 𝐾𝐸𝑓𝑋 + 𝐾𝐸𝑓𝑌

′
As 𝐾𝐸𝑖 > 𝐾𝐸𝑓 , 𝐾𝐸𝑖 > 𝐾𝐸𝑓𝑠 , then 𝐾𝐸𝑓𝑌 > 𝐾𝐸𝑓𝑌

Suggestion does not work.”

Student 14: “No. ∑ 𝐾𝐸 = constant. Block Z will gain KE, which causes block Y to

have less KE because KE of Y = ∑ 𝐾𝐸 − KE of X – KE of Z. If Z does not exist, KE

of Z = 0, and if Z does, KE of Z > 0, which lowers KE of Y.”

Although the test instrument underwent a number of rounds of refinements after the

pilot study, the results of the pilot study still provide insights via the excerpts of students’

mathematical thinking. The subsequent subsections elaborate on the results analyses of the

students’ performance in the main study using the final version of our test instrument, i.e. the

Chain-Collision Assessment. The instrument was expected to better excerpt students’

mathematical thinking from the newly-added mathematical expression tier in relation to their

responses in the answer and reason tiers. Furthermore, students were clearly instructed to

elaborate on their answer and/or reason mathematically in this tier.

6.2.2. Main Study

6.2.2.1.Analysis of Test Statistics

For the Chain-Collision Assessment, the mean FI for the test was .62 for the answer

tier, .36 for the reason tier, .20 for the mathematical expression tier, and .37 for the overall

scores (answer, reason and mathematical expression tiers). These values suggest that the test

209
was generally difficult. Mean DI for the test was .39 for the answer tier, .38 for the reason

tier, .46 for the mathematical expression tier and .42 for the overall scores (answer, reason

and mathematical expression tiers). As the DI values are more than .30, on average, the test

was able to reasonably discriminate the top 25% from the bottom 25% of the samples on the

basis of all tiers.

Cronbach Alpha in relation to the reliability of the test with respect to cognitive scores

was .65 for the answer tier, .58 for the reason tier, .70 for answer and reason tiers, and .81

when all tiers were considered. The modest values of Cronbach Alpha for the answer tier

only and reason tier only are due to the fact that the test was generally difficult (Mehrens &

Lehman, 1978). However, when both answer and reason tiers were considered, or when all

tiers were considered, the value of Cronbach Alpha increased to a more acceptable value of

.70 (Nunnaly, 1978). This shows that the mathematical expression tier of the assessment is

important for the internal consistency of the overall assessment. The reliabilities for the

confidence ratings were even higher, .94 for both the answer tier and reason tier.

Table 6.2

Reliability Statistics for Chain-Collision Assessment

All
Reliability Answer tier Reason tier Answer + Reason tiers
tiersa
statistic
(𝑁 = 123)
Score Confidence Score Confidence Score Confidence Score

Cronbach
.65 .94 .58 .94 .70 .95 .81
alpha
a
All tiers include answer, reason and mathematical expression tiers.

210
6.2.2.2.Analysis of Test Scores

Table 6.3

Cognitive Scores for Chain-Collision Assessment

% students % students
% students % students
(considering both (considering only
Scores (%) (considering only (considering all
answer and reason mathematical
answer tier) tiers)
tiers) expression tier)

70 and above 34.2 13.8 1.6 2.5

60-69.9 27.6 17.9 5.7 8.9

50-59.9 10.6 17.9 7.3 13.8

Below 50 27.6 50.4 85.4 74.8

Mathematical thinking does not come easily. Even in the population of high-

performing physics students, a high percentage of these students could not entirely express

their mathematical thinking (refer to Table 6.3). When only the answer tier is considered,

34.2% of the participants were able to score 70% and above whereas 27.6% of the

participants scored below 50%. When both answer and reason tiers are considered, the

percentage of participants who scored 70% and above drops to 13.8% and the percentage of

participants who scored below 50% increased to 50.4%. This further reiterates that some

students may score highly in a traditional multiple-choice tests even with the lack of

conceptual understanding, as they may approach the question solely by guessing, using

partial knowledge, elimination of unlikely options, or they may even have memorised the

correct answers from previous contexts without good understanding of the concept used.

Thus, it can be seen that the reason tier enhances the efficacy of the test. Moreover, when

211
asked to elaborate on their mathematical thinking behind each answer, only 1.6% of the

participants were able to score 70% or above, whereas a large majority of them (85.4%)

scored below 50%. The scores in the fifth tier affected the whole performance greatly as it

can be seen from the last column in Table 6.3.

6.2.2.3.Analysis of Correlations

As expected, a strong positive correlation were found between students’ performance

in the answer tier and their performance in the reason tier (𝑟(121) = .774, 𝑝 < .001),

indicating that students with stronger reasoning ability are more likely to be able to attain

high scores for the answer tier. But since the focus of this study is more on students’

mathematical thinking, it is not sufficient to only look at their performance in the answer and

reason tiers. This is because the participants’ scores in the fifth tier would be more

informative to show their ability to translate their mathematical thinking into correct answers

and reasons. A closer look at Table 6.3 reveals that the percentage of students who are able

to perform mathematical thinking, or translate their answer/reason to mathematical

expressions, is quite low – much lower than the percentage of students who are able to obtain

correct answer and/or reason. Interestingly, there are significant positive correlations

between the participants’ scores in the mathematical expression tier and their scores in the

answer tier (𝑟(121) = .566, 𝑝 < .001), as well as between the participants’ scores in the

mathematical expression tier and their scores in the answer and reason tiers (𝑟(121) = .540,

𝑝 < .001). The moderate strength of these correlations suggests that the students’ ability to

perform mathematical thinking may be a good indicator of their ability to achieve correct

212
answers and reasons in a physics test. However, variability in the test scores can still be

observed due to other factors, such as students’ strong preference to use qualitative analysis,

overreliance on scientific intuition, or inability to perform mathematical thinking, which

could have led them to guess the answers and/or reasons. These factors may be the cause of

lower performance in the mathematical expression tier as compared to the answer and reason

tiers.

6.2.2.4.Analysis of Confidence Ratings

Traditional multiple-choice questions have some drawbacks, in that obtaining correct

answer may be treated as content proficiency in students and wrong answer may be treated

as lack of knowledge. Multi-tier tests, in contrast, provide for better analysis. For example,

correct answer with low confidence rating indicates lack of knowledge while wrong answer

with high confidence rating indicates alternative conception (Sreenivasulu & Subramaniam,

2014). In this thesis study, only confidence ratings for the answer and reason tiers were

evaluated.

The mean confidence of the students for the whole test was 4.5 out of 6.0, which

indicates that, on average, students were about 75% confident in the veracity of their

responses, which is relatively quite high. This shows an interesting behaviour of high-

performing students: those who understand the concepts are quite confident that they

understand it well, but those who do not understand the concepts think confidently that they

actually know the concepts. It was found that the mean confidence for the answer tier (mean

± 𝑆𝐷 = 4.6 ± .4) was significantly higher than the mean confidence for the reason tier (mean

± 𝑆𝐷 = 4.5 ± .3), with 𝑡(122) = 4.64, 𝑝 < .001. This trend is also confirmed by their

213
performance in the physics examination in general, where most students are usually able to

perform well in getting correct answers, but find it difficult when asked to elaborate on the

concepts involved. In other words, explanatory knowledge is generally more difficult than

content knowledge (Sreenivasulu & Subramaniam, 2014).

For the answer tier, the mean ± 𝑆𝐷 confidence-when-correct (CFC) for the test was

4.8 ± .4 while the mean ± 𝑆𝐷 confidence-when-wrong (CFW) was 4.2 ± .2. These values

are relatively on the high side, above the mid-point of the scale; it indicates that the students

are quite confident in the concepts they understand but unfortunately they are overconfident

about concepts they do not understand. The difference in the mean CFC and CFW was

statistically significant, with 𝑡(14) = 5.31, 𝑝 < .001. For the reason tier, the respective

values were 4.6 ± .5 and 4.3 ± .3, which are also relatively high. They were also statistically

different, with 𝑡(14) = 2.37, 𝑝 = .033. When both the answer and reason tiers were

considered, the respective values were 4.7 ± .5 and 4.4 ± .3. They are also statistically

different, with 𝑡(14) = 2.41, 𝑝 = .030. In general, the participants were more confident in

the concepts they understood as compared to the concepts they did not understand.

The confidence bias (CB) yields a positive number for all questions, when both

answer and reason tiers were considered, which indicates students’ overconfidence in the

accuracy of their responses. This is a common trend, as indicated in Boekaerts & Rozendaal

(2010).

Interesting findings can be observed by looking at the values of confidence

discrimination quotient (CDQ). The CDQ yields negative values for Questions 14 and 15

when only answer tier is considered, Questions 11-14, when only reason tier is considered,

214
and Questions 11, 13-15, when both tiers are considered. Negative CDQ indicates that

students are more confident when they are wrong than when they are correct. Interestingly,

these questions are also in relation to the higher order thinking questions used in the pilot

study. These high confidence ratings for wrong answers enable us to excerpt alternative

conceptions. It can also be seen that for these questions, students generally scored very low

for mathematical explanation as compared to other questions, showing a trend that they rather

preferred to use their false scientific intuition or incorrect qualitative analysis to approach the

abovementioned questions. Qualitatively, this also suggests to us that students are

overconfident of their intuitive skills than in their mathematical skills. The mean ± 𝑆𝐷 CDQ

values of . 5 ± .4 for the answer tier, . 3 ± .4 for the reason tier and . 3 ± .4 for both tiers show

that students have rather modest discriminating power between what they think they know

and what they think they do not know. It is expected that students will indicate higher

confidence when they are correct than when they are incorrect (Lundeberg et al., 2000).

215
 Table 6.4

Analysis of Cognitive and Confidence Measures per Item (Chain-Collision Assessment)

216
6.2.2.5.Analysis of Mathematical Expression Tier

As mentioned earlier, mathematical thinking does not come easily as the majority of

the participants were not able to express the mathematical formalism used in approaching the

questions. This is shown by the low percentage (37.5%) in the number of questions in the

diagnostic test where students attempted to perform mathematical thinking, be it right or

wrong. Interestingly, Table 6.5 shows that the lowest percentages of students who showed

some understanding in their mathematical formalism are in Questions 13, 14 and 15, which

are also the higher order thinking questions used in the pilot study.

Table 6.5

Students’ Performance in Mathematical Expression Tier across Questions (Chain-Collision

Assessment)

% students who Mean total score

Mean score (%) Mean combined
obtained non-zero (%) for the answer,
Question obtained in the score (%) for
score in the reason and
number mathematical answer and reason
mathematical mathematical
expression tier tiers
expression tier expression tiers

1 25.2 25.2 76.4 54.5

2 14.6 14.1 83.3 53.7

3 26.0 23.3 65.9 47.6

4 10.6 9.8 39.4 26.7

5 39.0 33.1 54.5 45.3

6 36.6 34.4 67.9 53.5

7 20.3 19.2 25.6 22.9

8 28.5 27.6 38.2 33.7

217
 9 26.0 23.6 55.3 41.5

10 26.0 24.7 56.9 43.1

11 23.6 23.3 60.6 44.6

12 26.8 3.8 44.3 36.6

13 4.1 3.8 31.3 19.5

14 4.9 5.0 26.8 17.3

15 2.4 1.9 12.2 7.8

Mean 21.0 18.2 49.2 36.5

The complete set of the Chain-Collision Assessment questions are in the appendix.

Samples of responses expressed by student participants for each question that earned full 1.5

credits and partial credits of 1.0 or 0.5 are given in Table 6.6. Apart from these, some students

expressed irrelevant mathematical thinking in approaching a question, or even left the

explanation blank. For instance, many students would equate the initial kinetic energy of

Block I to the final kinetic energy of Block J in Question 14. This is clearly an alternative

conception and, as mentioned, would not earn any mark in the mathematical expression tier.

218
Table 6.6

Possible Students’ Responses, Score Assignments and Teacher’s Commentary for

Mathematical Expression Tier of Each Question in Chain-Collision Assessment

Question 1
Marks
This is a conceptual question on the conservation of momentum.

𝑝 = constant 0.5
Teacher’s commentary. Student expresses that the total momentum of the system is conserved, but
he/she does not show sufficient conceptual understanding on the reason. Based on teaching
experiences, many students would think that momentum would always be conserved in any situation,
which is clearly an alternative conception.

𝑚1 𝑣1 + 𝑚2 𝑣2 = 𝑚1 𝑣1′ + 𝑚2 𝑣2′ 1.0

Teacher’s commentary. Student expresses complete equation for conservation of momentum,
involving the masses and velocities. However, the mathematical formalism of the reason tier is still
missing.

𝑑𝐩⃗⃗
∑ 𝐅⃗ext =
𝑑𝑡 1.5
∑ 𝐅⃗ext = ⃗𝟎⃗ → 𝐩
⃗⃗initial = 𝐩
⃗⃗final → 𝑚1 𝑣1 + 𝑚2 𝑣2 = 𝑚1 𝑣1 ′ + 𝑚2 𝑣2 ′
Teacher’s commentary. Student expresses complete understanding that in an isolated system, the
total external force acting on the system is zero. From the Newton’s second law, this implies the
conservation of momentum of the system.

Question 2
Marks
This is a conceptual question on elastic collision.

∆𝐾 = 0 0.5
Teacher’s commentary. Student expresses that the kinetic energy should not change in this collision.
Although this equation can be applied for the given situation, but it lacks of understanding as to why
it can be applied.
1 1
2
𝑚1 𝑣12 + 2 𝑚2 𝑣2 2 does not change before and after collision in this collision 1.0
Teacher’s commentary. Student expresses a certain level of understanding of the physical quantity
that should be constant, particularly in this collision, but it still lacks of understanding no reason is
mentioned.

elastic → 𝐾initial = 𝐾final

1 1 1 1 1.5
𝑚 𝑣 2 + 𝑚2 𝑣2 2 = 𝑚1 𝑣1′2 + 𝑚2 𝑣2 ′2
2 1 1 2 2 2
Teacher’s commentary. Student expresses complete understanding that the kinetic energy of the
system is conserved only if the collision is elastic.

219
 Question 3
This is a True/False question pertaining to the concept of elastic collision, and expressed using Marks
mathematical formalism. This question would encourage students to use their conceptual
understanding on elastic collision and to express it mathematically.

𝑣1 > 𝑣2
0.5
as the first particle needs to move faster than the second one for a collision to happen
Teacher’s commentary. Student expresses a partial understanding that only leads to half of the
answer. The second inequality is missing.

𝑣1 − 𝑣2 = 𝑣2 ′ − 𝑣1 ′ 1.0
Teacher’s commentary. Student expresses a very important equation for the given collision, but
unfortunately this is not sufficient to lead him/her to the right answer as the conclusion (which is
expected to be in terms of an inequality) is missing.

For a collision to happen, 𝑣1 > 𝑣2 has to be satisfied

This implies, that in the centre of mass frame, the two particles approach each other before the
collision.
1.5
Since the collision is elastic, kinetic energy has to be conserved and the two particles have to move
away from each other.
Hence, in the original frame, 𝑣1 ′ < 𝑣2 ′
Teacher’s commentary. Student expresses understanding about elastic collision and the conditions
that need to be satisfied in such a collision.

Question 4
Marks
This question is an application of the concept of centre of mass and centre of mass frame.

𝑚1 𝑣1 + 𝑚2 𝑣2
𝑣CM =
𝑚1 + 𝑚2 0.5
𝑣2 < 𝑣CM < 𝑣1
Teacher’s commentary. Student expresses a very limited understanding on the concept of centre of
mass. He/She knows how to obtain the velocity of the frame, but fails to do further analysis to
compare the momenta of the two particles in the centre of mass frame.

𝑚1 𝑣1 + 𝑚2 𝑣2
𝑣CM =
𝑚1 + 𝑚2
1.0
𝑣2 < 𝑣CM < 𝑣1
𝑣1 − 𝑣CM > 0 and 𝑣2 − 𝑣CM < 0
Teacher’s commentary. Student expresses a certain level of understanding as he/she is able to
conclude that the two particles move in the opposite direction, i.e. approach each other, in the centre
of mass frame. Although the student gets to analyse the directions of the velocities in this frame,
he/she still fails to conclude which particle has larger momentum.

𝑚1 𝑣1 + 𝑚2 𝑣2
𝑣CM =
𝑚1 + 𝑚2
𝑚2
𝑣1 − 𝑣CM = (𝑣 − 𝑣2 )
𝑚1 + 𝑚2 1 1.5
𝑚1
𝑣2 − 𝑣CM = − (𝑣 − 𝑣2 )
𝑚1 + 𝑚2 1
𝑚1 |𝑣1 − 𝑣CM | = 𝑚2 |𝑣2 − 𝑣CM |

220
Teacher’s commentary. Student expresses full analysis of the velocities of the particles in the centre
of mass frame. He/She would conclude that the two particles would approach each other in this
frame, and they should have equal momentum.

Question 5
This question is a problem-solving question that tests students’ understanding on elastic and inelastic Marks
collision.

(6.0)(3.0) + (2.0)(0.0) = (6.0)(2.0) + (2.0)𝑣𝐵 ′ 0.5

Teacher’s commentary. Student expresses the first step required for the analysis of the system, which
is the conservation of momentum. It does not matter whether or not he/she solves for the velocity of
Block B after the collision, as the analysis is still far from the required full analysis of this question.
The student needs to determine whether the momentum of the first block is fully or partially
transferred to the second block. And moreover, he/she needs to also mathematically verify whether
the collision is possible, i.e. no increase in the kinetic energy of the system.

(6.0)(3.0) + (2.0)(0.0) = (6.0)(2.0) + (2.0)𝑣𝐵 ′ → 𝑣𝐵 ′ = 3.0 m/s

(2.0)(3.0) < (6.0)(3.0) 1.0
No energy loss as friction can be ignored.
Teacher’s commentary. Student expresses the conservation of momentum, and that the fact that the
momentum is partially transferred from one block to the other. However, this analysis is not
sufficient as it does not analyse whether the collision is possible.

(6.0)(3.0) + (2.0)(0.0) = (6.0)(2.0) + (2.0)𝑣𝐵 ′ → 𝑣𝐵 ′ = 3.0 m/s

(2.0)(3.0) < (6.0)(3.0) 1.5
3.0 − 0.0 = 3.0 m/s > 3.0 − 2.0 = 1.0 m/s
Teacher’s commentary. Student express the conservation of momentum and use it to solve for the
final velocity of Block B, and that the fact that the momentum is partially transferred from one block
to the other. Moreover, he/she also analyses the relative velocity of the two particles before and after
the collision, as this physical quantity cannot increase, except in the case of explosion. This is
equivalent to saying that the kinetic energy of the system does not increase.

Question 6
Marks
This question is a problem-solving question that tests understanding on elastic and inelastic collision.

𝑣𝐴 ′ ≠ 0 hence energy is only partially transferred from Block A to Block B 0.5

Teacher’s commentary. Student is able to show that Block A does not come to rest after the collision
with Block B, and hence he/she might conclude that the kinetic energy of Block A is only partially
transferred to Block B. However, this analysis is only true to certain extent. It is definitely possible
that the kinetic energy of Block A is partially transferred to Block B and partially lost to the
surroundings, and Block A comes to rest after the collision.

1 1
(2.0)(3.0)2 < (6.0)(3.0)2
2 2 1.0
The final kinetic energy of Block B is less than the final kinetic energy of Block A and Block A is
still moving. Hence, the energy is only partially transferred from Block A to Block B.
Teacher’s commentary. Student is able to show that the kinetic energy is partially transferred from
one block to the other. However, this analysis is not sufficient as he/she needs to also mathematically
show that both blocks are in motion after the collision and that the collision is possible, i.e. kinetic
energy does not increase after the collision.

221
 𝑣𝐴′ ≠ 0
1 1 1 1
(6.0)(2.0)2 + (2.0)(3.0)2 = 21 J < (6.0)(3.0)2 + (2.0)(0.0)2 = 27 J 1.5
2 2 2 2
1 2
1 2
(2.0)(3.0) < (6.0)(3.0)
2 2
Teacher’s commentary. Student is able to present the full analysis of the given situation. Both blocks
are in motion after the collision, energy is only partially transferred from Block A to Block B and
the collision is possible as it describes an inelastic collision with some amount of energy loss to the
surroundings.

Question 7
Similar to Question 5, this question is a problem-solving question that tests students’ understanding Marks
on elastic and inelastic collision. However, students are expected to conclude that the given situation
is physically impossible.

(6.0)(3.0) + (2.0)(0.0) = (6.0)(1.4) + (2.0)𝑣𝐷 ′ 0.5

Teacher’s commentary. Student expresses the first step required for the analysis of the system, which
is the conservation of momentum. It does not matter whether or not he/she solves for the velocity of
Block D after the collision, as the analysis is still far from the required full analysis of this question.
The student needs to determine whether the momentum of the first block is fully or partially
transferred to the second block. And moreover, he/she needs to also mathematically verify whether
the collision is possible, i.e. no increase in the kinetic energy of the system. The last analysis is the
most crucial analysis for the given situation.

(6.0)(3.0) + (2.0)(0.0) = (6.0)(1.4) + (2.0)𝑣𝐷 ′ → 𝑣𝐷 ′ = 4.8 m/s

(2.0)(4.8) < (6.0)(3.0) 1.0
No energy loss as friction can be ignored.
Teacher’s commentary. Student expresses the conservation of momentum, and that the fact that the
momentum is partially transferred from one block to the other. However, this analysis is not
sufficient as it does not analyse whether the collision is possible from the angle of energy, which is
the most crucial analysis for the given situation.

(6.0)(3.0) + (2.0)(0.0) = (6.0)(1.4) + (2.0)𝑣𝐷 ′ → 𝑣𝐷 ′ = 4.8 m/s

(2.0)(4.8) < (6.0)(3.0) 1.5
3.0 − 0.0 = 3.0 m/s < 4.8 − 1.4 = 3.4 m/s
Teacher’s commentary. Student express the conservation of momentum and use it to solve for the
final velocity of Block D, and that the fact that the momentum is partially transferred from one block
to the other. Moreover, he/she also analyses the relative velocity of the two particles before and after
the collision, as this physical quantity cannot increase, except in the case of explosion. The last
inequality shows that the magnitude of relative velocity increases after the collision, and hence the
collision is physically impossible. This is equivalent to saying that the kinetic energy of the system
increases.

Question 8
Similar to Question 6, this question is a problem-solving question that tests students’ understanding Marks
on elastic and inelastic collision.

𝑣𝐶 ′ ≠ 0 hence energy is only partially transferred from Block C to Block D 0.5

Teacher’s commentary. Student is able to show that Block C does not come to rest after the collision
with Block D, and hence he/she might conclude that the kinetic energy of Block C is only partially
transferred to Block D. However, this analysis is only true to certain extent. It is definitely possible

222
that the kinetic energy of Block C is partially transferred to Block D and partially lost to the
surroundings, and Block C comes to rest after the collision. Moreover, his/her analysis is still short
from the analysis of the type of collision. In this case, he/she fails to notice that the collision is
physically impossible.

1 1
(2.0)(4.8)2 < (6.0)(3.0)2
2 2 1.0
The final kinetic energy of Block D is less than the final kinetic energy of Block C and Block C is
still moving. Hence, the energy is only partially transferred from Block C to Block D.
Teacher’s commentary. Student is able to show that the kinetic energy is partially transferred from
one block to the other. However, this analysis is not sufficient as he/she needs to also mathematically
show that the collision is possible, i.e. kinetic energy does not increase after the collision. In this
given situation, student fails to conclude that the collision is actually physically impossible.

1 1 1 1
(6.0)(2.0)2 + (2.0)(4.8)2 = 29 J < (6.0)(3.0)2 + (2.0)(0.0)2 = 27 J
2 2 2 2 1.5
Hence, the collision is impossible.
Teacher’s commentary. Student is able to present the full analysis of the given situation. The most
important analysis here is related to the total kinetic energy of the system. As the total kinetic energy
of Blocks C and D after the collision is higher than the one before the collision, student is able to
conclude that the collision is physically impossible.

Question 9
Similar to Questions 5 and 7, this question is a problem-solving question that tests students’
understanding on elastic and inelastic collision. It is quite thought-provoking as Block E comes to a Marks
stop after the collision. Yet, this does not exclude the possibility that the collision may be impossible.
Student is expected to do further analysis.

18 = 0 + 9𝑣𝐹′
0.5
𝑣𝐹′ = 2
Teacher’s commentary. Student expresses understanding that in a closed system, the total
momentum of the system is conserved. This is the first required step to analyse the collision and
hence, he/she expresses the equation for conservation of momentum. However, this is still far from
the expected analysis where the kinetic energy of the system needs to be analysed before and after
the collision, to see whether the collision is elastic, inelastic, or even impossible.

𝑚𝐸 𝑣𝐸 + 𝑚𝐹 𝑣𝐹 = 𝑚𝐸 𝑣𝐸′ + 𝑚𝐹 𝑣𝐹′
(6.0)(3.0) = 9.0𝑣𝐹′
′
𝑣𝐹 = 2.0 m/s < 𝑣𝐸 = 3.0 m/s
1 1.0
initial kinetic energy = (6.0)(3.0)2 = 27 J
2
1
final kinetic energy = (9.0)(2.0)2 = 18 J
2
Teacher’s commentary. Student presents the conservation of momentum and the calculation of the
initial and final kinetic energies of the system. These constitute the major steps required to analyse
the question. However, he/she thinks that the momentum of Block E is partially transferred to Block
F because the speed acquired by Block F after the collision is smaller than the initial speed of Block
E. This is clearly an alternative conception as momentum is determined by both mass and velocity,
and not only velocity. Although student shows a certain level of understanding in the correct
direction, but it is still short from the expected analysis.

223
 Conservation of momentum:
(6.0)(3.0) = (9.0)𝑣𝐹′ → 𝑣𝐹′ = 2.0 m/s and 𝑣𝐸′ = 0.0 m/s
|∆𝑣𝐹 | = 2.0 m/s < 3.0 m/s = ∆𝑣𝐸
1
∑ KE𝑖 = (6.0)(3.0)2 = 27 J 1.5
2
1
∑ KE𝑓 = (9.0)(2.0)2 = 18 J
2
∑ KE𝑓 < ∑ KE𝑖
Teacher’s commentary. Student presents the conservation of momentum and the comparison
between the velocity gain of Block F and the velocity reduction of Block E. This analysis is necessary
to eliminate option C in the answer tier. As 𝑣𝐸′ = 0.0 m/s, he/she concludes that the momentum of
Block E is fully transferred to Block F. Student also presents the comparison between the total initial
kinetic energy and final kinetic energy of the system. From this analysis, he/she is also able to
conclude that the collision is inelastic.

Question 10
Similar to Questions 6 and 8, this question is a problem-solving question that tests students’
understanding on elastic and inelastic collision. Though Block E comes to a stop after the collision, Marks
yet it does not mean all its kinetic energy is transferred to Block F. This is a further analysis that is
expected for this question.

1
KE𝑖 = (6.0)(3.0)2
2 0.5
1
KE𝑓 = (9.0)(𝑣𝐹′ )2
2
Teacher’s commentary. Student expresses a limited understanding that for this question the initial
and final kinetic energies of the system need to be calculated. However, he/she stops at this step and
it does not show enough understanding of the concept, as reflected in his/her mathematical
formalism. The value of 𝑣𝐹′ can be obtained from Question 9, and the comparison between the kinetic
energies need to be done to conclude if the collision is elastic, inelastic, or impossible.

Conservation of momentum: 𝑚𝐸 𝑣𝐸 = 𝑚𝐹 𝑣𝐹′

𝑚𝐸
𝑣𝐹′ = 𝑣 < 𝑣𝐸 1.0
𝑚𝐹 𝐸
Since KE ∝ 𝑣 2, KE𝑓 < KE𝑖
Teacher’s commentary. Student presents a good understanding leading to his conclusion that the
collision between Blocks E and F is inelastic. This is shown in his mathematical thinking that the
total energy of the system decreases. However, he/she only considers the decrease in the velocity
involved in the calculation for kinetic energy. Though this still leads him/her to the correct answer
and reason, but it still falls short of the complete analysis as kinetic energy should involve mass and
velocity in the calculation.

1
initial KE = (6.0)(3.0)2 = 27 J
2 1.5
1
final KE = (9.0)(2.0)2 = 18 J < 27 J
2
Teacher’s commentary. Student presents the complete required analysis for this question. The initial
and final kinetic energies of the system of both blocks are evaluated and compared. And since there
is a decrease in the kinetic energy of the system, kinetic energy of Block E is partially transferred to
Block F and the collision is inelastic.

224
 Question 11
Similar to Questions 5, 7 and 9, this question is a problem-solving question that tests students’
understanding on elastic and inelastic collision. Although it looks like this question is no different Marks
than the previous ones, but student is expected to notice that the collision is perfectly elastic from
their mathematical analysis of coefficient of restitution, relative velocity, or even the total kinetic
energy of the system before and after the collision.

1 1
(4.0)(3.0)2 + (𝑚𝐻 )(0.0)2
2 2
1 1
= (4.0)(1.0) + (𝑚𝐻 )(4.0)2
2
2 2
3.02 − 1.02 0.5
𝑚𝐻 = (4.0) ( ) = 2.0 kg
4.02
𝑚𝐻 has a unique solution.
𝑝𝑓 < 𝑝𝑖
Teacher’s commentary. Student expresses the calculation with conservation of energy being
assumed. Using conservation of energy as a first step can only be correct if the obtained value for
𝑚𝐻 is then used to check whether the system’s total momentum is conserved. If the momentum is
conserved, then the collision is indeed perfectly elastic. If it is not conserved, then the collision is
either inelastic or impossible. However, student does not show this analysis and hence his/her
analysis is still short from the expected one. In fact, students are actually encouraged to perform
conservation of momentum first before checking on the system’s kinetic energy.

Conservation of momentum:
(4.0)(3.0) + (𝑚𝐻 )(0.0) = (4.0)(1.0) + (𝑚𝐻 )(4.0) → 𝑚𝐻 = 2.0 kg 1.0
8.0 < 12
Teacher’s commentary. Student presents the conservation of momentum as a first step, which is the
correct analysis for an isolated system. He/She then continues to analyse that the momentum
transferred from Block G to Block H is less than the initial momentum of Block G. Hence,
momentum is only partially transferred. Although up to this point, the analysis is correct, but it still
falls short of the complete analysis as there is no analysis of the system’s kinetic energy. To
determine whether the collision is elastic, inelastic, or impossible, the system initial and final kinetic
energies need to be determined and compared. In the given situation, student fails to realise that the
collision is perfectly elastic.

12 = 4.0 + 4.0𝑚𝐻
𝑚𝐻 = 2.0 kg
8.0 < 12
1 1.5
(4.0)(3.0)2 = 18 J
2
1 1
(4.0)(1.0)2 + (2.0)(4.0)2 = 18 J
2 2
Teacher’s commentary. Although student’s presentation of mathematical thinking is not really clear,
but it can be deduced that he used the conservation of momentum to obtain the value of 𝑚𝐻 . The
marks given should not be based on clarity, but student’s mathematical thinking. In this case, his/her
mathematical thinking on the system’s momentum is expressed in his/her mathematical formalism.
Student then compares the momentum acquired by Block H with the initial momentum of Block G
to see that the momentum of Block G is partially transferred to Block H. Student also calculates the
initial and final kinetic energies of the system to conclude that the collision is indeed elastic. This
represents a full correct analysis although he/she chooses the wrong option for the reason tier.

225
 Question 12
Similar to Questions 6, 8 and 10, this question is a problem-solving question that tests students’ Marks
understanding on elastic and inelastic collision. Student is also expected to notice that the collision
is perfectly elastic.

Coefficient of Restitution = 1
0.5
Hence, the collision is elastic.
Teacher’s commentary. Student expresses the correct inference from his expression of coefficient of
restitution, i.e. elastic collision. However, he/she does not describe how he obtains the value of this
coefficient, and he/she might just simply recall the statement. It also lacks of analysis of whether the
kinetic energy of Block G is transferred partially or fully to Block H.

Before the collision, the two blocks approach each other at

3.0 − 0.0 = 3.0 m/s
After the collision, the two blocks move apart from each other at 1.0
4.0 − 1.0 = 3.0 m/s
Hence, it is an elastic collision.
Teacher’s commentary. Student presents a better mathematical thinking than the above. The
magnitude of the relative velocity between the two blocks is invariant across inertial frames. Hence,
analysing from the centre of mass frame yields constancy of each block’s speed before and after the
collision, implying conservation of energy. It can then be inferred that the collision is elastic,
regardless of the value of 𝑚𝐻 . However, he/she does not show whether the kinetic energy of Block
G is transferred partially or fully to Block H.

Momentum of the system is conserved.

𝑣𝐺 𝑚𝐺 + 𝑣𝐻 𝑚𝐻 = 𝑣𝐺′ 𝑚𝐺 + 𝑣𝐻′ 𝑚𝐻
3.0(4.0) = 4.0(1.0) + 4.0𝑚𝐻
𝑚𝐻 = 2.0 kg

1 1 1.5
KE𝑖 = 𝑚𝐺 𝑣𝐺 2 = (4.0)(3.0)2 = 18 J
2 2
1 ′2
1 ′ 2
1 2
1
KE𝑓 = 𝑚𝐺 𝑣𝐺 + 𝑚𝐻 𝑣𝐻 = (4.0)(1.0) + (2.0)(4.0)2 = 18 J
2 2 2 2
Kinetic energy is conserved.
Teacher’s commentary. Student presents the full mathematical analysis for this question. The initial
and final kinetic energies of the system of both blocks are compared. Since kinetic energy is
conserved before and after the collision, the collision is elastic. And since the final kinetic energy of
the system comprises of the kinetic energy of Block G and kinetic energy of Block H, it can be
inferred that the initial kinetic energy of Block G is partially transferred to Block H.

Question 13
Similar to Questions 5, 7, 9 and 11, this question is a problem-solving question that tests students’
understanding on elastic and inelastic collision. However, the question does not specify any of the
final velocity of the blocks. Students are expected to interpret the phrase “as much as possible” for
the kinetic energy transfer. This phrase implies a perfectly elastic collision. The analysis can be done Marks
from the centre of mass frame whereby both block would bounce exactly with the same speeds as
their initial speeds to conserve the kinetic energy and to make sure the total momentum to be zero in
this frame. A more qualitative analysis would be to think in terms of what type of collision is the
most effective in terms of conserving the system’s kinetic energy.

(4.0)(9.0) + (9.0)(0.0) = (4.0)𝑣𝐼 ′ + (9.0)𝑣𝐽 ′ 0.5

226
Teacher’s commentary. Student expresses the first step required for the analysis of the system, which
is the conservation of momentum. Many students are expected to get it correct up to this point, but
not be able to perform further analysis in terms of interpreting the phrase “as much as possible”.

(4.0)(9.0) + (9.0)(0.0) = (4.0)𝑣𝐼 ′ + (9.0)𝑣𝐽 ′

1 1 1 1
(4.0)(9.0)2 + (9.0)(0.0)2 = (4.0)𝑣𝐼 ′2 + (9.0)𝑣𝐽 ′2 1.0
2 2 2 2
Maximise 𝑣𝐽 ′
Teacher’s commentary. Student expresses the conservation of momentum and conservation of
kinetic energy, but fails to realise that they are actually sufficient conditions to approach the given
situation, i.e. elastic collision. Hence, this student is still not able to perform the full analysis.

(4.0)(9.0) + (9.0)(0.0) = (4.0)𝑣𝐼 ′ + (9.0)𝑣𝐽 ′

or equivalently, 𝑣𝐽 ′ − 𝑣𝐼 ′ = 9.0
1.5
45 72
𝑣𝐼′ = − m/s = −3.5 m/s ; 𝑣𝐽′ = m/s = 5.5 m/s
13 13
Teacher’s commentary. Student expresses the conservation of momentum and conservation of
kinetic energy, and realises that this is the required condition for maximum energy transfer. He/She
is also able to conclude that Block I should not come to rest after the collision.

Question 14
Similar to Questions 6, 8, 10 and 12, this question is a problem-solving question that tests students’
understanding on elastic and inelastic collision. However, as described in Question 13, students are Marks
expected to interpret the phrase “as much as possible” for the kinetic energy transfer as a perfectly
elastic collision.

1 1 1 1
(4.0)(9.0)2 + (9.0)(0.0)2 = (4.0)𝑣𝐼 ′2 + (9.0)𝑣𝐽 ′2
2 2 2 2 0.5
If 𝑣𝐼′ = 0, 𝑣𝐽′ = 6.0 m/s, but it violates the conservation of momentum. Hence, 𝑣𝐼′ ≠ 0.
Teacher’s commentary. Student is able to show that Block I should not come to rest as most people
would think. This is because kinetic energy of Block I should not get transferred fully to Block J.
However, this does not form the full analysis yet as it does not tell you how much the kinetic energy
should get transferred.

1 1 1 1
(4.0)(9.0)2 + (9.0)(0.0)2 = (4.0)𝑣𝐼 ′2 + (9.0)𝑣𝐽 ′2
2 2 2 2 1.0
(4.0)(9.0) + (9.0)(0.0) = (4.0)𝑣𝐼 ′ + (9.0)𝑣𝐽 ′
Maximise 𝑣𝐽 ′
Teacher’s commentary. Student expresses the conservation of momentum and conservation of
kinetic energy, but fails to realise that they are actually sufficient conditions to approach the given
situation, i.e. elastic collision. Hence, this student is still not able to perform the full required analysis.

(4.0)(9.0) + (9.0)(0.0) = (4.0)𝑣𝐼 ′ + (9.0)𝑣𝐽 ′

1 1 1 1
(4.0)(9.0)2 + (9.0)(0.0)2 = (4.0)𝑣𝐼 ′2 + (9.0)𝑣𝐽 ′2
2 2 2 2
or equivalently, 𝑣𝐽 ′ − 𝑣𝐼 ′ = 9.0
45 72 1.5
𝑣𝐼′ = − m/s = −3.5 m/s ; 𝑣𝐽′ = m/s = 5.5 m/s
13 13
1
(9.0)(5.5)2
𝜂=2 = 84.03%
1
(4.0)(9.0)2
2

227
Teacher’s commentary. Student expresses the conservation of momentum and conservation of
kinetic energy, and realises that this is the required condition for maximum energy transfer. He/She
is also able to conclude that Block I should not come to rest after the collision. In addition, he/she
also calculates the percentage of kinetic energy of Block I that gets transferred to Block J. Though
the last analysis is not required, it gives a full picture of how much he/she understands about the
concepts involved.

Question 15
This final question is the ultimate question that tests the students’ understanding on the concept of Marks
maximising kinetic energy transfer. Students with good understanding on momentum/collision are
expected to analyse this using the concept of elastic collision.

(4.0)(9.0) + 𝑚𝐾 (0.0) = (4.0)𝑣𝐼 ′ + 𝑚𝐾 𝑣𝐾 ′

4.0𝑣𝐼 ′ + 𝑚𝐾 𝑣𝐾 ′ = 36
𝑚𝐾 𝑣𝐾 ′ + (9.0)(0.0) = 𝑚𝐾 𝑣𝐾 ′′ + 9.0𝑣𝐽 ′ 0.5
Maximise 𝑣𝐽 ′
Teacher’s commentary. Student presents the first step of analysis, which is the conservation of
momentum. However, his/her analysis is very limited as it does not really lead to the correct answer.

(4.0)(9.0) + 𝑚𝐾 (0.0) = (4.0)𝑣𝐼 ′ + 𝑚𝐾 𝑣𝐾 ′

4.0𝑣𝐼 ′ + 𝑚𝐾 𝑣𝐾 ′ = 36
𝑣𝐾 ′ − 𝑣𝐼 ′ = 9.0 since the collision must be elastic for maximum energy transfer 1.0
𝑚𝐾 𝑣𝐾 ′ + (9.0)(0.0) = 𝑚𝐾 𝑣𝐾 ′′ + 9.0𝑣𝐽 ′
𝑣𝐽′ − 𝑣𝐾′′ = 𝑣𝐾 ′ since the collision must be elastic for maximum energy transfer
Teacher’s commentary. Student presents a very good analysis on this problem. He/She elaborates on
the necessary mathematical formalisms needed for the full analysis, i.e. conservation of momentum
and conservation of energy, which is reduced to the conservation of the magnitude of relative
velocity. Although the student presents correct analysis to a certain level, it is still short from the full
expected analysis as it still may not lead to the correct answer and/or reason.

(4.0)(9.0) + 𝑚𝐾 (0.0) = (4.0)𝑣𝐼′ + 𝑚𝐾 𝑣𝐾′

4.0𝑣𝐼′ + 𝑚𝐾 𝑣𝐾′ = 36
𝑣𝐾′ − 𝑣𝐼′ = 9.0
Solving,
′
72
𝑣𝐾 =
𝑚𝐾 + 4.0
72
𝑚𝐾 + (9.0)(0.0) = 𝑚𝐾 𝑣𝐾′′ + 9.0𝑣𝐽′
𝑚𝐾 + 4.0
72𝑚𝐾 1.5
𝑚𝐾 𝑣𝐾′′ + 9.0𝑣𝐽′ =
𝑚𝐾 + 4.0
72
𝑣𝐽′ − 𝑣𝐾′′ =
𝑚𝐾 + 4.0
Solving,
144𝑚𝐾
𝑣𝐽′ =
(𝑚𝐾 + 4.0)(𝑚𝐾 + 9.0)
Plugging in any value for 𝑚𝐾 within the range of 4.0 < 𝑚𝐾 < 9.0 yields a value for 𝑣𝐽′ that is
bigger than 5.5 m/s. Hence, it is true.
Teacher’s commentary. Student presents the full analysis required to approach this question. Firstly,
he/she realises that both collisions involved must be perfectly elastic for maximum energy transfer.
Secondly, varying the mass of Block K may yield higher velocity for Block J as compared to the one
obtained in Questions 13 and 14.

228
 As observed, the fifth tier is used to see the elaboration of students’ mathematical

thinking while applying the relevant physics concepts in approaching a question. In other

words, students’ conceptual understanding is also expected to be seen in the form of

mathematical language in this tier. As described above, the interviews were done to excerpt

students’ mathematical thinking even more, and it serves as triangulation for the results. In

the interviews, the overreliance on scientific intuition among high-performing students is

also interestingly observed as students expressed their mathematical thinking. Unfortunately,

the use of intuition often leads to incorrect answers and incorrect reasons.

6.2.2.6.Analysis of Interviews

Below are partial transcripts of the interviews. Students 1 and 2 were awarded partial

marks for Question 9 – 0.5 and 1.0 out of 1.5 marks, respectively. Though they both showed

some understanding, Student 15 showed very limited level of understanding as he solely

considered the conservation of momentum in this case. During the interview, he admitted

that he made a guess for the reason tier. After being probed further by the interviewer, Student

15 was able to provide further mathematical analysis that would lead him in the correct

direction.

Interviewer: “Why do you think you can apply the equation for conservation of

momentum?”

229
 Student 15: “In this case there are no external forces acting, there is no friction

on the floor either, so I thought momentum should be conserved and all momentum

should be transferred.”

Interviewer: “Why is all momentum transferred? Why is it not transferred

partially?”

Student 15: “If only partial momentum that is transferred, then Block E should

still be in motion.”

Interviewer: “You chose option B for the reason. How do you know that the

collision is inelastic?”

Student 15: “At that time, it felt that it was most correct to me. I made a guess.”

Interviewer: “What further mathematical analysis should you perform other than

conservation of momentum, if you want to know whether the collision is elastic or

inelastic?”

Student 15: “Is it the formula for conservation of kinetic energy? (short pause) oh

you could actually look at the difference in the velocities before and after the

collision… The magnitude of 𝑣2 minus 𝑣1 must be equal to 𝑣1′ minus 𝑣2′ .”

Interviewer: “So what if they are not equal?”

Student 15: “If they are not equal then it implies that the collision isn’t elastic.”

Although Student 16’s mathematical explanation is worth more marks, his explanation did

not lead him to the correct answer. This is also supported by the interviews. After queried by

the interviewer, he realised his mistake and managed to come out with the correct analysis,

even for an additional question.

230
 Interviewer: “Why do you think that the momentum is partially transferred?”

Student 16: “Because the final velocity of Block F is lower than the initial velocity

of Block E, so the momentum isn’t completely transferred from Block E to Block F

and some of the momentum will be lost to the surrounding.”

Interviewer: “But just now you mentioned that the total momentum is conserved,

and you also wrote down the equation for conservation of momentum. Are you sure

that the momentum is only partially transferred?”

Student 16: (short pause) “Okay… yeah, I think the momentum is totally

transferred from Block E to Block F.”

Interviewer: “Even though the final velocity of Block F is lower than the initial

velocity of Block E?”

Student 16: “Yeah, because momentum is mass times velocity.”

Interviewer: “What if both masses are equal?”

Student 16: “The momentum will still be conserved but the velocity of Block F

after the collision will be the same as the velocity of Block E before the collision.”

In contrast, Student 17 was able to show her conceptual understanding as required while

explaining her mathematical formalism for Question 9. She even managed to answer a further

question by the interviewer without showing any signs of alternative conceptions.

Student 17: “Whether the collision is elastic or inelastic depends on whether the

kinetic energy of the system is conserved. So initially we can find the kinetic energy

231
 to be half times six times three squared. And now that the final velocity of F is 2 m/s,

so the final kinetic energy is half times nine times two squared, which we notice that

the final kinetic energy in the final state is less than the kinetic energy in the initial

state. Hence, this collision is inelastic.”

Interviewer: “I change the masses here, and Block E stops moving after the

collision, can I still say that all its momentum is transferred to Block F?”

Student 17: “In this case, it is not the case of disintegration, so the final kinetic

energy cannot be more than the initial kinetic energy. The values of the masses must

be such that this does not happen.”

Below are some other transcripts for students’ interviews in relation to Question 10.

Students 18 and 19’s mathematical explanations are worth 0.5 and 1.0 marks respectively out

of 1.5. Both students managed to arrive at the correct answer and reason, but Student 4

displayed very limited understanding even during the interview. When prompted, he

admitted that his correct answer was just a guess. But after a short pause, and with some

guidance, Student 4 managed to arrive at the correct mathematical thinking.

Student 18: “So basically the question asks us to look at the system’s kinetic

energy and whether the kinetic energy is partially transferred or completely

transferred. So the equation I wrote here is the initial kinetic energy which is

expressed by the kinetic energy of the block that was first moving, and the final kinetic

energy which is the kinetic energy of the block which was collided and is now moving.

Interviewer: “How does that lead to your answer?”

232
Student 18: “From this, I would basically need to compare the initial kinetic

energy and the final kinetic energy and to see whether they have the same value, but

the issue is that I do not have the value for the final velocity over here.”

Interviewer: “So did you make a guess to your answer that the kinetic energy is

partially transferred?”

Student 18: “Yes, it was a guess.”

(short pause)

Student 18: “But I think I could have used the conservation of linear momentum

in the x-direction to get 𝑣F′ .”

Interviewer: “Oh, okay. So you realised that there is another equation you can use.

Why can you apply the conservation of linear momentum?”

Student 18: “Because if you consider Blocks E and F to be one system, there are

no external forces acting in the x-direction and therefore linear momentum is

conserved in that direction.”

Interviewer: “Okay, so from the conservation of linear momentum, you could get

the value of the final velocity. So, will it change your answer?”

(short pause)

Student 18: “I think it will be the same.”

Interviewer: “So the kinetic energy will be partially transferred?”

Student 18: “Yes, because the collision is inelastic.”

233
Student 19 was awarded more marks as he showed a certain level of understanding, which

led him to the correct answer and reason, though it was still incomplete. The interviewer then

tried to guide him a little further towards the correct answer.

Student 19: “Since momentum is conserved, and the first block comes to rest, then

the velocity is proportional to the ratio of the masses. For example, if the second mass

is twice heavier than the first one, then its final velocity will be twice smaller than the

initial velocity of the first mass. And since kinetic energy is proportional to 𝑣 2 , so the

kinetic energy will decrease too.”

Interviewer: “So you conclude that the kinetic energy decreases. What further

mathematical analysis should you do here? I guess you are forgetting the mass.

Kinetic energy is proportional to 𝑚𝑣 2 to account for the mass.”

Student 19: “Yes. Oh so in my example, the mass would increase by two and

velocity would decrease by 4, so overall the kinetic energy would still decrease.”

It is also interesting to see that as the interviewer probed the student with further

mathematical interrogations, the student was then able to perform mathematical analysis right

on the spot.

Interviewer: “Do you think there is any possibility that the kinetic energy is

completely transferred?”

234
 Student 19: “I do not think that it will conserve the momentum if kinetic energy is

completely transferred.”

Interviewer: “Is there any exception, for certain values of the masses perhaps?”

Student 19: “If both are of the same mass, then it is possible.”

In contrast, Student 16 managed to explain very well the mathematical formalism for

Question 10 and was awarded full marks (1.5 marks). The interviewer also tried to probe the

student to perform further mathematical analysis and though the student struggled at first,

eventually, he was able to do it on the spot.

1
Student 16: “Basically kinetic energy equals to 𝑚𝑣 2 . I calculated the total
2

kinetic energy initially and total kinetic energy after the collision. If they are the same,

it just means that the collision is elastic because energy is conserved. But in this case,

there is actually a decrease in the energy, which means there is energy loss to the

surroundings. So that is an inelastic collision.”

Interviewer: “Is it possible for the first mass to transfer all its kinetic energy to the

second mass?”

Student 16: “Yes, it should be possible, and the first block will come to rest…”

Interviewer: “So the first block will come to rest and the other one moves?”

(short pause)

Student 16: “Oh wait. I don’t think it is possible for one block to transfer all its

kinetic energy to another block.”

235
 Interviewer: “In the previous question, you mentioned that if both masses are

equal, the first block will stop moving after the collision, and the second block will

move with speed equal to the initial speed of the first block? Does that mean the

energy is fully transferred?”

Student 16: “Umm… Yes. Oh so I think it is possible.”

Interviewer: “So what is the required condition then?”

Student 16: “Both masses must be equal.”

The analysis of the interviews shows reasonable marks for students’ explanation in

the Chain-Collision Assessment, and it provides further indication of the depth of students’

conceptual understanding, as expressed in their mathematical formalism.

6.2.2.7.Analysis of Alternative Conceptions in Momentum/Collision

Table 6.7 shows a list of 10 alternative conceptions on the concept of

momentum/collision documented in our study. The mean confidence for the alternative

conceptions ranges from 3.4 to 5.1. Based on the classification scheme of Caleon &

Subramaniam (2010b), only one of these alternative conceptions is spurious and the rest are

genuine. Although the concepts are explicitly taught in a physics class on

momentum/collision, many students still show a strong tendency of misinterpreting these

when applied to higher order thinking problems. A large percentage of students showed

alternative conceptions for Questions 13 (32.3%), 14 (35.8%) and 15 (68.3%).

236
Table 6.7

Alternative Conceptions on Momentum/Collision

% sample
Question with Mean
No. Alternative conception
number alternative confidence
conception

In a collision between two masses, conservation of

1 momentum applies only when the collision is Q1 AA 17.1 5.1
elastic.

Q1 AE 26.8 5.0
Q5 BD 32.5 4.8
Q7 BD 15.4 5.0
In a collision between two masses, conservation of
2 Q7 CD 11.4 3.9
momentum applies with no required conditions.
Q9 DD 17.1 4.5
Q11 BD 35.8 4.7
Q13 BD 22.8 4.7

In a collision between two masses, only the case of

3 one-dimensional collision imposes constraints to Q3 AB 15.4 5.0
the velocities of the masses.

In the centre of mass frame of two bodies, larger

4 Q4 AA 23.6 3.4
mass implies larger momentum, and vice versa.

In the absence of friction, a collision between two

5 bodies is always elastic, i.e. no loss of kinetic Q5 BA 11.4 4.0
energy during the collision.

In a collision between two masses, as long as Q7 BB 10.6 3.8

6 conservation of momentum applies, then the
collision is possible. Q8 AB 28.5 4.2

In a collision between two masses, if one comes to

7 rest after the collision, its kinetic energy must have Q10 DB 10.6 4.4
all been transferred to the other.

After a collision between two masses, if both

8 masses move in the same direction, then the Q12 AA 13.8 4.0
collision must have been inelastic.

In a collision between two masses, the best case Q13 CB 19.5 4.4
9 scenario for kinetic energy transfer occurs when Q14 AB 25.2 4.5
one block comes to rest after the collision. Q15 BA 50.4 4.7

237
 In a collision between two masses, the best case Q14 DB 10.6 3.8
10 scenario for kinetic energy transfer occurs when
the collision is inelastic. Q15 BB 17.9 3.9

As some of the alternative conceptions listed in Table 6.7 are overlapping in terms of their

underlying conceptual constructs, we have categorised these items into five classes for

discussion.

(1) Concept of conservation of momentum (Items No. 1-3)

The concept of conservation of momentum is a consequence of Newton’s second law or

Newton’s third law, depending on the choice of system. The total momentum of a system is

only conserved when the system is isolated, namely when the vector sum of the external

forces acting on the system equals to zero. This forms the only condition for conservation of

momentum. Unfortunately, after students thought that they had the concept mastery, a lot of

them took it for granted, especially after doing a lot of problem-solving where conservation

of momentum always works. Hence, this led to an alternative conception: In a collision

between two masses, conservation of momentum applies without any required condition –

expressed by about 35.8% of the samples for Question 11, with mean confidence 4.7. This

confirms that students tend to apply their false intuition instead of careful application of the

concept when it comes to a physics situation that they think they are familiar with (Ricardo,

2016).

In a collision involving two masses, as long as no external forces act on the system,

momentum is conserved, irrespective of whether the interaction between them dissipates or

conserves energy, that is, whether a collision between them is elastic or inelastic. The crux

238
of this assessment is to see if students were able to move to the next layer of thinking beyond

just applying the conservation of momentum but also understanding why they could apply it

in a physical system.

Additionally, the existence of collision between two masses itself imposes a

constraint to the initial velocities of the masses. Mathematically, the unit vector of their

relative velocity must be in the opposite direction to the unit vector of their relative position.

𝐯⃗⃗2 − 𝐯⃗⃗1 𝐫⃗2 − 𝐫⃗1

=−
|𝐯⃗⃗2 − 𝐯⃗⃗1 | |𝐫⃗2 − 𝐫⃗1 |

(6.1)

Conservation of momentum dictates the relation between masses and velocities, and it

imposes another constraint to the motion of two masses, regardless of their degrees of

freedom.

(2) Concept of centre of mass frame (Item No. 4)

By definition, the centre of mass of a system is located at the average of the weighted position

vector of all constituent masses. It moves with the average of the weighted velocity vector of

all constituent masses. Therefore, in the centre of mass frame, the vector sum of the

momentum (which is equivalent to weighted velocity) of all the system’s constituent masses

is equal to zero. In a system of only two blocks, the magnitudes of the momentum of both

blocks must then be equal, otherwise, the two momenta do not add up to zero. It is definitely

an alternative conception to think that: In the centre of mass frame of two bodies, larger mass

implies larger momentum, and vice versa. This alternative conception led 23.6% of the

participants (with mean confidence 3.4) to think that in the centre of mass frame of a two-

239
body system, the body with larger mass has larger momentum, as expressed in Question 4.

Although it is an alternative conception, the mean confidence for this choice of answer is

below 3.5, indicating it might be a spurious alternative conception. However, as this mean

confidence is only slightly below the mid-point scale, it does show that these students were

not merely guessing when answering this question.

(3) Concept of relative velocity in elastic and inelastic collisions (Items No. 5-6, 8)

In a one-dimensional collision, the relative velocity of two masses can be analysed from the

centre of mass frame. This is because relative velocity is invariant across inertial frames. In

the centre of mass frame, a collision can only occur if both masses approach each other,

indicating that 𝑣1 > 𝑣2 in the original frame. After the collision, the two masses move away

from each other or stick together, indicating that 𝑣1′ ≤ 𝑣2′ in the original frame. However,

elastic collision demands that the total kinetic energy be conserved both in the original frame

as well as in the centre of mass frame. As we see it in the centre of mass frame, conserving

kinetic energy implies each mass moves away from the other mass with exactly the same

speed as its initial speed before the collision. This implies a stricter inequality in the original

frame, 𝑣1′ < 𝑣2′ .

An elastic collision happens when no kinetic energy is dissipated in the process of

collision. In contrast, in an inelastic collision (a dissipative collision), energy is lost to the

surroundings in the form of heat, sound, or stored potential energy of the masses. This

happens commonly due to the deformation of the bodies (Ricardo & Lee, 2015). Whether a

collision is elastic or inelastic is determined by the amount of energy loss during the short

240
time of collision itself, regardless of the presence of friction between the bodies and the

ground.

The crux of the underlying concepts in these questions lies in the following statement:

the kinetic energy of the system cannot increase after a collision, which may seem very

intuitive. However, many students are usually reluctant to perform this very important check,

as the following alternative conception was harboured by 28.5% of the participants in

Question 8 (with mean confidence 4.2): In a collision between two masses, as long as

conservation of momentum is satisfied, then the collision is possible. Conceptually, a one-

dimensional collision will only be possible if three conditions are satisfied: (a) a collision

happens, i.e. 𝑣1 > 𝑣2 ; (b) the total momentum of the system is conserved, i.e. 𝑚1 𝑣1 +

𝑚2 𝑣2 = 𝑚1 𝑣1′ + 𝑚2 𝑣2 ′; and (c) the total kinetic energy of the system does not increase, or

equivalently, the magnitude of the relative velocity between the two blocks does not increase,

i.e. 𝑣2′ − 𝑣1′ ≤ 𝑣1 − 𝑣2 . Although the percentage is not very high, about 13.8% of the

participants (with mean confidence 4.0) thought that: In a collision between two blocks, as

long as both blocks still move in the same direction after the collision, the collision must have

been inelastic. And hence, in an elastic collision, the body must bounce back after the

collision, which is shown in Question 12. However, the three conditions above do not dictate

the directions of the final velocities; in an elastic collision, the magnitude of the relative

velocity between the two blocks is preserved.

(4) Transfer of all momentum vs. transfer of all kinetic energy to another body (Item No.

7)

241
An alternative conception was observed in Question 10 as 10.6% of the samples (with mean

confidence 4.4) thought that: In a collision between an initially-moving block and another

block at rest, if the initially-moving block comes to rest after the collision, it must have

transferred all its kinetic energy to the other block. Students were expected to perform a

deeper mathematical analysis to see why it is impossible. Transferring all kinetic energy to

an initially-stationary block implies no energy loss, i.e. elastic collision. Applying the

conservation of momentum and relative velocity, we obtain 𝑚1 𝑣1 = 𝑚2 𝑣2 ′ and 𝑣2′ = 𝑣1 ,

which is only possible if 𝑚1 = 𝑚2 . If the two blocks have different masses, then one of the

equations needs to be compromised. As conservation of momentum cannot be violated in an

isolated system, the only way is for the collision to be inelastic. In other words, if the initially-

moving block comes to rest after the collision with a stationary block, then it must have

transferred all of its momentum to the other block. However, this does not imply transferring

all of its kinetic energy to the other block. In fact, a careful check of the system’s initial and

final kinetic energies needs to be performed to ensure that the collision is possible. If kinetic

energy increases, then the collision is impossible.

(5) Elastic collision being the best-case scenario for momentum transfer or kinetic energy

transfer (Items No. 9-10)

About 50.4% of the samples (with mean confidence 4.7) chose to interpret the concept in

Question 15 this way: In a collision between an initially-moving block and another block at

rest, the best-case scenario for kinetic energy transfer occurs when the initially-moving block

comes to rest after the collision. The same alternative conception was shown by 25.2% of

242
the participants (with mean confidence 4.5) in Question 14, with some interpreting it as the

first block transferring all its momentum to the second block, and some others interpreting it

as the first block transferring all its kinetic energy to the second block. This violates physics

laws, as described in point (4) above.

Analysing the system in the centre of mass frame enables us to see that elastic

collision provides the best-case scenario for momentum transfer or kinetic energy transfer.

This is because the second mass would bounce off with the greatest possible speed after the

collision, both in the centre of mass frame and in the original frame. In fact, intuitively, the

best case scenario should not happen when the first mass transfers all of its momentum as

that would imply some energy loss to the surroundings unless the masses are equal, nor does

it happen when it transfers all of its kinetic energy as that would cause violation of

conservation of momentum unless the masses are equal. The best-case scenario should refer

to an elastic collision as there would be no loss of energy to the surroundings.

Mathematically, it can be shown that the closer the values of the masses, the higher

the percentage of kinetic energy transferred. Ricardo and Lee (2015) also describe the

possibility of increasing the kinetic energy transferred by putting another stationary block in

between the two blocks, as long as the value of the mass of this additional block is in between

the values of the masses of the two blocks – describing the concept of chain-collision. The

best-case scenario would happen as the ratio of two adjacent masses is equal throughout the

chain of masses.

243
6.3. Study on Rotational Mechanics

This section presents analyses of the pilot study and main study on rotational

mechanics. Similar to the study on momentum/collision, in general the findings show that

mathematical thinking does not come easily, even for high-performing students with strong

conceptual background.

6.3.1. Pilot Study

As expected, in the first set of the assessment, low performance was observed for

three questions, which are the higher order thinking questions incorporated in the assessment.

It was noted that none of the students expressed their mathematical thinking for the first

higher order thinking question, only 17.9% expressed their mathematical thinking for the

second one, and only 46.4% tried to perform some mathematical analysis (on top of

qualitative analysis) for the third one. Nevertheless, only 53.6% managed to answer the first

higher order thinking question correctly, 42.9% managed to answer the second one correctly,

and none of these students managed to answer the third one correctly.

For example, for the last question, Student 17 was using the formula for torque while

Student 18 was using the formula for angular momentum to analyse the problem but they

ended up performing incorrect analyses:

Student 17: “The point mass should hit the rod at one of it ends. Given that the

angular speed would depend on the magnitude of 𝛕 ⃗⃗ = 𝐫⃗ × 𝐅⃗, and

⃗⃗, torque, and 𝛕

244
 maximising 𝐫⃗, the distance from the centre of mass, would result in the greatest torque

possible.”

length of rod
Student 18: “𝑥 = . This will allow for the furthest distance from the
2

centre of mass, allowing for the greatest torque, allowing for the greatest angular

⃗ =𝐑
momentum 𝐋 ⃗⃗ × 𝐩
⃗⃗. Since 𝐩 ⃗⃗ must be maximised. This results in an
⃗⃗ is the same, 𝐑

overall greatest increase in angular velocity.”

In contrast, for this particular question in the second set of assessment where

mathematical questioning was used, 14 students (50%) were observed to elaborate on their

mathematical thinking, another 4 students (14.3%) performed pure qualitative analysis or

relied on intuition, and the remaining 10 students (35.7%) chose not to answer this question.

On average, these 14 students managed to obtain 78.5% of the full score, with 9 of them

(32.1%) obtaining full score for this particular higher order thinking question. The number

of students who at least approached the question in the right manner increased significantly

in this set of assessment though many still could not arrive at the desired final answer.

Mathematical questioning was observed to be effective in provoking students’ mathematical

thinking.

245
Figure 6.4

Distribution of Students’ Performance in the Pilot Study on Rotational Mechanics

Note. All questions were considered in the score calculation, with 10 marks being the
maximum score. Average improvement due to mathematical questioning can be seen from
the shift of the mean scores from the left to the right (shown using asterisks *).

Below are some extracts of students’ mathematical thinking for the higher order

thinking question. These students are top performers in the physics class but still performed

incorrect analyses in this particular question.

Student 19: “[attempted to differentiate but he did not continue] Since 𝜔 is

ℓ ℓ ℓ
increasing on the bound (− 2 , 2), 𝑥 = 2 is the point for 𝜔max .”

246
 Student 20: “Since all of the energy is transferred by the ball to the ball-rod

system, the maximum rotational kinetic energy there is, the higher the angular speed

of the rod. To increase the angular kinetic energy [he meant rotational kinetic

energy] would require the linear kinetic energy [he meant translational kinetic

energy] to decrease which can be done by minimising 𝑉 [the linear speed of the rod],

which can then be done by maximising 𝑥. Thus, the largest angular speed is provided

when the ball hits the end of the rod.”

Although students’ mathematical thinking could be somewhat excerpted, it was not

easy to ask the students to elaborate on their mathematical thinking. Therefore, the instrument

used in the pilot study was refined significantly into the final version of our test instrument,

i.e. the Rigid Body Collision Assessment, which was used in our main study. Nevertheless,

the results analysis of the pilot study still provided useful insight on students’ overreliance

on their scientific intuition to answer higher order thinking questions.

6.3.2. Main Study

6.3.2.1.Analysis of Test Statistics

For the Rigid Body Collision Assessment, the mean FI for the test was .64 for the

answer tier, .33 for the reason tier, .11 for the mathematical expression tier and .32 for the

overall scores (answer, reason and mathematical expression tiers). These values suggest to

us that the overall test was generally difficult. Moreover, among these high-performing

247
students, on average, only 11% of the participants were able to express their mathematical

thinking for each question. Mean DI for the test was .32 for the answer tier, .34 for the reason

tier, .25 for the mathematical expression tier and .30 for the overall scores (answer, reason

and mathematical expression tiers). As the values of DI were .30 and above, except for the

mathematical expression tier, this shows that the test was able to reasonably discriminate the

top 25% from the bottom 25% of the samples on the basis of answer and reason tiers.

The reliability of the test with respect to cognitive scores, based on Cronbach Alpha,

was .48 for the answer tier, .56 for the reason tier, .60 for answer and reason tiers and .88

when all these tiers were considered. The modest values of Cronbach Alpha for the answer

tier only, reason tier and even answer and reason tiers combined are due to the fact that the

test was generally difficult (Mehrens & Lehman, 1978). However, when all tiers are

considered, the value of Cronbach Alpha was considered good. In fact, this shows that the

fifth tier of the assessment is beneficial for the consistency of the overall assessment. The

reliabilities for the confidence ratings were even higher, .92 for the answer tier as well as for

the reason tier.

248
Table 6.8

Reliability Statistics for Rigid Body Collision Assessment

Reliability Answer + Reason

Answer tier Reason tier All tiersa
statistic tiers
(𝑁 = 126) Score Confidence Score Confidence Score Confidence Score

Cronbach
.48 .92 .56 .92 .60 .93 .88
alpha
a
All tiers include answer, reason and mathematical expression tiers.

6.3.2.2.Analysis of Test Scores

The results presented in this section also show that mathematical thinking does not

come easily even for students with strong conceptual understanding in physics. In Table 6.9,

30.2% of the participants performed very well (70% and above from the full score) if only

the answer tier is considered, whereas only 11.1% of them scored below 50%. This is indeed

a good achievement in a conceptual test, considering that Rotational Mechanics is one of the

relatively difficult topics in Physics. When the reason tier is included in the analysis, the

above percentages changed significantly to 10.3% and 53.2%, respectively. This reiterates

the significance of using multi-tier tests to excerpt students’ conceptual understanding as

compared to traditional MCQs. The reason tier clearly differentiates students who approach

the questions through guessing, or elimination of unlikely option, or even using incomplete

knowledge of the concept tested and, thus, increases the efficacy of the test. However, when

all tiers (including the fifth tier) are considered in calculating the test scores, these

percentages change even more significantly to 2.4% and 92.0%, respectively. This shows

249
that the majority of the students are not able to translate their conceptual understanding into

mathematical thinking.

Table 6.9

Cognitive scores for Rigid Body Collision Assessment

% students % students
% students % students
(considering both (considering only
Scores (%) (considering only (considering all
answer and reason mathematical
answer tier) tiers)
tiers) expression tier)

70 and above 30.2 10.3 1.6 2.4

60-69.9 34.9 7.9 .8 2.4

50-59.9 23.8 28.6 1.6 3.2

Below 50 11.1 53.2 96.0 92.0

6.3.2.3.Analysis of Confidence Ratings

Similar to the Chain-Collision Assessment, the Rigid Body Collision Assessment

only uses confidence ratings for the answer and reason tiers. The analyses of these confidence

ratings provide a more nuanced understanding on the profile of the participants in terms of

their conceptual understanding. For example, correct answer with low confidence rating

indicates lack of knowledge while wrong answer with high confidence rating indicates an

alternative conception (Sreenivasulu & Subramaniam, 2014).

The mean ± 𝑆𝐷 confidence of the students for the whole test was 4.1 ± .4 out of 6.0,

i.e. they are about 68% confident in the veracity of their responses. For the answer tier, the

mean ± 𝑆𝐷 confidence-when-correct (CFC) for the test was 4.2 ± .5 while the mean ± 𝑆𝐷

for confidence-when-wrong (CFW) was 3.9 ± .3. These values are relatively on the high

250
side, above the mid-point of the scale. For the reason tier, the respective values are 4.2 ± .9

and 3.8 ± .3, which are also relatively high. When both tiers are considered, the respective

values are 4.3 ± .8 and 3.9 ± .3. This shows an interesting pattern in the population of high-

performing students, in which the students with strong conceptual understanding are

confident in the concepts they know, but unfortunately students who are short on their

understanding overconfidently thought they understood the concepts well. The difference in

the mean CFC and CFW for the answer tier is statistically significant, with 𝑡(15) = 3.43,

𝑝 < .01. For the reason tier, they are moderately different, with 𝑡(15) = 2.11, 𝑝 = .052.

And when both tiers are considered, they are also statistically different, with 𝑡(15) = 2.41,

𝑝 = .029. This, in general, shows that the participants are more confident in the concepts

they understand compared to the concepts they do not understand.

It was also found that the students perceived explanatory knowledge to be more

difficult than content knowledge, as expressed in the mean confidence for the answer tier

(mean ± 𝑆𝐷 = 4.2 ± .4) that is significantly higher than the mean confidence for the reason

tier (mean ± 𝑆𝐷 = 4.0 ± .4), with 𝑡(125) = 5.54, 𝑝 < .001, thus confirming the findings in

the previous study (Sreenivasulu & Subramaniam, 2014).

The confidence bias (CB) yields a positive number for all questions except Question

2, when both answer and reason tiers are considered, which indicates students’

overconfidence in the accuracy of their responses. This is a common trend, as indicated in

Boekaerts & Rozendaal (2010).

The confidence discrimination quotient (CDQ) yields negative value for Questions

11 and 16 when only answer tier is considered. The same trend was observed for Questions

251
8, 11 and 16 when only reason tier is considered as well as when both tiers are considered.

Negative CDQ means students are more confident when they are wrong than when they are

correct, which is a strong indication of the existence of alternative conceptions. Interestingly,

these questions are based on the alternative conceptions harboured by students in the pilot

study, but revised in the final version of the instrument. It can also be seen that for Questions

8 and 16, students generally scored very low for mathematical explanation, showing a trend

that they rather preferred to use their qualitative reasoning or intuition for these questions

which may lead to false conclusions. In other words, students are more confident in using

their scientific intuition than their mathematical thinking. The mean ± 𝑆𝐷 CDQ values of

. 3 ± .3 for the answer tier, . 4 ± .5 for the reason tier and for both tiers show that students

have rather modest discriminating power between what they think they know and what they

think they do not know, confirming previous findings that students are generally more

confident when they are correct than when they are incorrect (Lundeberg et al., 2000).

252
 Table 6.10

Analysis of Cognitive and Confidence Measures per Item (Rigid Body Collision Assessment)

253
6.3.2.4.Analysis of Mathematical Expression Tier

As mentioned, mathematical thinking does not come easily even in the population of

high-performing students in the subject of Physics. This is revealed by the very low

percentage of questions in the diagnostic test in which students attempted to perform

mathematical thinking to explain their answer and/or reason, regardless of whether the

analysis was right or wrong. Table 6.11 shows that the lowest percentages of students who

showed some understanding in their mathematical formalism occur for Questions 8, 14 and

16. While Question 14 tests the students’ knowledge on elastic collision, Questions 8 and 16

correspond to the alternative conceptions harboured by most students in the pilot study, thus

confirming the findings of CDQ.

Table 6.11

Students’ Performance in Mathematical Expression Tier across Questions (Rigid Body

Collision Assessment)

% students who Mean total score

Mean score (%) Mean combined
obtained non-zero (%) for answer,
Question obtained in the score (%) for
score in the reason and
number mathematical answer and reason
mathematical mathematical
expression tier tiers
expression tier expression tiers

1 20.6 18.8 75.8 51.4

2 55.6 51.3 90.9 73.9

3 34.1 30.7 50.4 42.0

4 10.3 9.5 42.9 28.6

5 11.1 11.1 42.1 28.8

6 11.1 10.3 32.1 22.8

254
 7 4.0 4.0 76.2 45.2

8 2.4 1.6 37.7 22.2

9 4.8 3.7 55.2 33.1

10 7.1 7.1 52.4 33.0

11 4.8 4.5 17.9 12.1

12 5.6 4.0 51.2 31.6

13 4.0 4.0 40.1 24.6

14 1.6 1.6 65.5 38.1

15 6.4 5.0 36.9 23.2

16 2.4 2.4 13.9 9.0

Mean 11.3 10.6 48.8 32.5

Interestingly, a number of participants chose option F (other reason) for some

questions and wrote “by calculation” for the reason, regardless of whether their mathematical

expression was correct. Table 6.12 elaborates on this further. These are not classified as

alternative conceptions as they are not concept-related mistakes. Rather, it indicates that

although some students are able to perform mathematical thinking to some extent, they are

largely unable to articulate it conceptually. Although the percentages are not high, this

portrays a certain level of disconnection in students’ minds between the concept and its

mathematical formalism in general.

255
Table 6.12

Failure in Representing Mathematical Thinking in terms of Conceptual Reasoning –

Answer is Correct but “Other Reason” was Selected (Rigid Body Collision Assessment)

Question number % sample Mean confidence

Q6 13.5 3.8

Q8 15.9 4.4

Q9 14.3 4.6

Q11 15.1 4.2

Q15 15.9 4.0

Q16 12.7 4.1

Selected samples of students’ responses for some questions that earned full 1.5 marks

and partial marks of 1.0 or 0.5 are presented in Table 6.13. There were also quite a number

of participants who did not earn any mark for the mathematical expression tier as they

expressed wrong or showed irrelevant mathematical thinking, or left it blank. An example of

an irrelevant explanation would be “the total angular momentum of the system is not

conserved as the system is initially at rest and finally in motion after the collision”. In fact, it

is possible for the system to be in motion, yet still possess zero angular momentum. This case

happens when the translational angular momentum vector perfectly cancels the rotational

angular momentum vector.

256
Table 6.13

Possible Students’ Responses, Score Assignments and Teacher’s Commentary for

Mathematical Expression Tier of Each Question in Rigid Body Collision Assessment

Question 1
Marks
This question tests students’ understanding on the concept of translation and rotation.

ℓ
𝜏=𝐹× 0.5
2
Teacher’s commentary. Although student’s mathematical formulation shows the existence of
external force and torque, but this is still far from acceptable explanation. He/She does not explain
why the centre of the rod moves, nor why it rotates about its centre. The existence of an external
force itself does not guarantee that the body would undergo a linear motion, neither does the
existence of an external torque guarantee that the body would undergo a rotation. Newton’s law
equations for both translation and rotation are missing here.

ℓ
𝜏=𝐹 1.0
2
𝐹 = 𝑚𝑎
Teacher’s commentary. Student shows some understanding on the linear acceleration caused by the
external force 𝐹. He/She presents the equation to obtain the external torque from the force. However,
the Newton’s law equation for rotation is still short. The existence of a torque does not imply angular
acceleration, Students are expected to relate the net torque to the angular acceleration of the rod via
Newton’s law. Although he/she has shown a certain level of understanding in the correct direction,
but it is not sufficient to form a full analysis.

ℓ
𝜏=𝐹 = 𝐼𝛼
2 1.5
𝐹 = 𝑚𝑎
The body will experience acceleration since 𝐹 is the only force on it.
Teacher’s commentary. Student presents a complete understanding on the presented case in Question
1. Firstly, he/she notices that 𝐹 is the only external force acting on the system and hence it forms the
net force. This force, based on Newton’s law for translation, causes an acceleration of the rod’s
centre. At the same time, the force exerts a torque on the rod about its centre. This torque produces
an angular acceleration via Newton’s law for rotation.

Question 2
Students are expected to interpret what it means by “more difficult” in this question. Torque is not Marks
the only factor to be considered here. Students also need to consider the moment of inertia of the
rod, which is constant in this case and not constant in the next question.

torque = force × length 0.5

Teacher’s commentary. Student expresses a limited understanding that does not really lead to the
conclusion on why he/she chooses the options for his answer and reason, respectively. He/She
expresses the correct equation relating torque and force. However, he/she stops at this step and it
does not show enough understanding of the concept.

𝜏 = 𝐹 × 𝑅 → If 𝑅 decreases, 𝜏 decreases. Hence, it is harder to rotate. 1.0

257
Teacher’s commentary. Student presents a good understanding beyond the equation relating torque
and force. He/She concluded that if the force acts at some other point on the rod other than the rod’s
end, it would lead to the decrease in the amount of torque produced. However, this is still short from
the expected analysis as our concern is the “rotation of the rod” and not just what causes the rotation.

⃗⃗ × 𝐅⃗
⃗𝛕⃗ = 𝓵
𝜏 = ℓ𝐹 (since 𝜃 = 90°) 1.5
When ℓ decreases, 𝜏 decreases. Since 𝜏 = 𝐼𝛼 and 𝐼 is constant, 𝛼 also decreases.
Hence, it is more difficult to rotate the rod.
Teacher’s commentary. Student’s mathematical explanation depicts what is expected in the full
analysis of the given physical scenario. On top of relating the torque and force, he/she acknowledges
the decrease in the amount of torque as the force acts at some other point on the rod other than the
rod’s end. He/She also uses Newton’s law for rotation to explain that the angular acceleration of the
rod will have to decrease too as the moment of inertia of the rod remains unchanged. Above and
beyond this analysis, he/she also shows the vector cross product between position vector and force
vector to obtain the expression for torque. Although this is not really expected from the participants,
but this shows the student’s strong ability in performing mathematical thinking.

Question 3
Similar to Question 2, students are expected to analyse all variables that change when the length of Marks
the rod is doubled. In this question, both the torque and moment of inertia change.

1
𝐼= 𝑀ℓ2
12
𝜏
𝛼= 0.5
𝐼
ℓ
𝜏=𝐹
2
Teacher’s commentary. Student expresses a limited understanding on the relations between torque
and force as well as between torque and angular acceleration. This does not amount to ample analysis
on why the angular acceleration changes.

𝐹 = 𝑚𝑎
ℓ
𝜏 = 𝐹 = 𝐼𝜃̈ 1.0
2
Now the 𝜏 is multiplied by 2.
Teacher’s commentary. Student presents a good understanding on the Newton’s law. On top of that,
he/she also acknowledges that the torque is now larger by a factor of 2 as the length of the rod is
doubled. However, he/she still fails to see how the moment of inertia is affected in this case.

ℓ 1
𝜏 = 𝐹 ( ) = 𝐼𝛼 = 𝑚ℓ2 𝛼
2 12
6𝐹
𝛼=
𝑚ℓ 1.5
ℓ 1
𝜏′ = 𝐹 (2 ) = 𝐼𝛼 = 𝑚(2ℓ)2 𝛼
2 12
3𝐹
𝛼′ =
𝑚ℓ
Teacher’s commentary. Student shows a full understanding of the presented case and his/her
mathematical analysis shows a thorough understanding of the physics concepts involved. By
doubling the length of the rod, the torque is now doubled but the moment of inertia is quadrupled.
These two changes cause the angular acceleration of the rod to decrease.

258
 Question 4
The question tests on students’ understanding of the conservation of linear momentum. Although Marks
the concept is simple, but the existence of rotational motion adds another layer of difficulty.

∑ 𝐅⃗ext = ⃗𝟎⃗ 0.5

Teacher’s commentary. Student expresses that the net external force is zero, but he/she does not
show sufficient conceptual reasoning.

∑ 𝐅⃗ext = ⃗𝟎⃗ → 𝐩
⃗⃗ = constant 1.0
Teacher’s commentary. Student expresses complete equation for the required condition for
conservation of momentum. As there is no net force acting on the system, the momentum of the
system is conserved. However, the mathematical formalism of the conservation of momentum
involving the masses and velocities given in the question is still missing.

System is isolated, hence ∑ 𝐅⃗ext = ⃗𝟎⃗

1.5
This implies conservation of momentum: 𝑚𝑢 = 𝑚𝑣 + 𝑀𝑉
Teacher’s commentary. Student expresses complete understanding that as the system is isolated, the
net force acting on the system is zero. This zero net force implies conservation of momentum of the
system. In terms of the given masses and velocities, he/she expresses the equation for conservation
of momentum.

Question 5
The question tests on students’ understanding of the conservation of angular momentum, particularly Marks
on the choice of system and the reference point used. The concept is slightly more difficult than the
conservation of linear momentum due to its dependence on the reference point.

∑ ⃗𝛕⃗ext = ⃗𝟎⃗ 0.5

Teacher’s commentary. Student expresses that the net external torque is zero, but he/she does not
show sufficient conceptual understanding on the reason as he/she does not link it back to the
question. The reason on why the total torque is zero is also missing.

∑ ⃗𝛕⃗ext = ⃗𝟎⃗ → 𝐋
⃗ = constant 1.0
Teacher’s commentary. Student expresses the required condition for conservation of angular
momentum. As there is no net torque acting on the system, the angular momentum of the system is
conserved. However, he/she does not describe why the total torque is zero, nor the mathematical
formalism of the conservation of angular momentum.

System is isolated, hence ∑ ⃗𝛕⃗ext = ⃗𝟎⃗

Taking the rod’s centre to be the reference point, 1.5
1
𝑚𝑢𝑥 = 𝑚𝑣𝑥 + 𝑀ℓ2 𝜔
12
Teacher’s commentary. Student expresses complete understanding on the concept. Firstly, the
system of the particle and rod is isolated as given in the questions, and hence, the net torque acting
on it is zero. This zero net torque implies conservation of angular momentum of the system. Since
the rod’s centre is taken as the reference point, the translational angular momentum of the rod
vanishes and he/she correctly expresses the equation for conservation of angular momentum in terms
of the given variables and constants.

259
 Question 6
The question tests on students’ understanding of the conservation of angular momentum. The
question is a little tricky as the particle is excluded from the system and the rod’s centre is not taken Marks
as the reference point. It is indeed wrong to say that the system is isolated as the particle would exert
an external force onto the rod.

∑ ⃗𝛕⃗ext = ⃗𝟎⃗ 0.5

Teacher’s commentary. Student expresses that the net external torque is zero, but he/she does not
show sufficient explanation to this mathematical statement as well as how it links to the question.

∑ ⃗𝛕⃗ext = ⃗𝟎⃗ → 𝐋
⃗ = constant 1.0
Teacher’s commentary. Student expresses the required condition for conservation of angular
momentum. As there is no net torque acting on the system, the angular momentum of the system is
conserved. But again, no sufficient explanation is given as to why the net torque is zero. The equation
for conservation of angular momentum is also missing.

System is not isolated, but since the point of collision is taken as a reference, ∑ ⃗𝛕⃗ext = ⃗𝟎⃗
This implies conservation of angular momentum of the rod about the point of collision.
Taking the point of collision to be the reference point, 1.5
1 12𝑉𝑥
𝑀𝑉𝑥 − 𝑀ℓ2 𝜔 = 0 → 𝜔 = 2
12 ℓ
Teacher’s commentary. Student expresses complete understanding on the concept. Firstly, knowing
that the particle is excluded from the system, the net external force cannot be zero. But since the
point of collision is taken as a reference, the net external torque acting on the system is still zero,
implying conservation of angular momentum. Student also understands that the angular momentum
of the particle should be excluded from the equation; the equation should only contain the
translational angular momentum and rotational angular momentum of the rod which have to cancel
each other as the rod is initially at rest. The choice of the reference point is crucial in this case.
Should another reference point is taken, the net external torque may not be zero and angular
momentum would not be conserved.

Question 7
Marks
This is a conceptual question on elastic collision.

∆𝐾 = 0 0.5
Teacher’s commentary. Student expresses that kinetic energy should not change in this collision.
Although this equation can be applied for the given situation, but it lacks of understanding as to why
it can be applied.
1 1 1 1
2
𝑚𝑢2 is distributed to 2 𝑚𝑣 2, 2 𝑀𝑉 2 and 2 𝐼𝜔2 1.0
Teacher’s commentary. Student expresses a certain level of understanding of the physical quantity
that should be constant, particularly in this collision, but it still lacks of understanding as he/she does
not mention the reason why kinetic energy should be conserved.

1 1 1 1
elastic → 𝐾initial = 𝐾final → 𝑚𝑢2 = 𝑚𝑣 2 + 𝑀𝑉 2 + 𝐼𝜔2
2 2 2 2 1.5
2 2 2
1 2 2
𝑚𝑢 = 𝑚𝑣 + 𝑀𝑉 + 𝑀ℓ 𝜔
12

260
Teacher’s commentary. Student expresses complete understanding that the kinetic energy of the
system is conserved only if the collision is elastic. Moreover, he/she also expresses the mathematical
formalism for the conservation of energy based on the given variables and constants.

Question 8
This questions tests students’ understanding on how the energy transfer takes place to maximise the
angular velocity of the rod after the collision. A very common alternative conception on this is to Marks
think that the particle has to stop moving in order to transfer all its kinetic energy. The easiest way
to approach this question is by looking at the collision in the centre of mass frame.

𝑣 = 0 does not maximise 𝜔 0.5

Teacher’s commentary. Student gives the correct initial indication that the particle should not come
to rest in order to maximise the angular velocity of the rod after the collision. However, he/she does
not describe the method of maximisation used as well as how the final linear velocities of the particle
and rod are related.

𝑣 = 𝑉 in order to maximise 𝜔 1.0

Teacher’s commentary. Student expresses the correct relation of the final velocities of the particle
and rod after the collision in order to maximise the rod’s angular velocity. Nevertheless, the
reasoning as to why they must be equal is missing.

In the centre of mass frame, the total linear momentum of the system must be zero.
To maximise, 𝜔, both the particle and rod must come to rest after the collision in the centre of
mass frame.
This implies 𝑣 = 𝑉 in the original frame of reference. 1.5
Conservation of linear momentum:
𝑚𝑢 = 𝑚𝑣 + 𝑀𝑉
𝑚
𝑣=𝑉= 𝑢
𝑚+𝑀
Teacher’s commentary. Student expresses the full explanation required. Looking at the collision
from the centre of mass frame allows him/her to deduce that the particle and rod must move together
after the collision in order to produce the maximum angular velocity of the rod. Moreover, he/she
also expresses this final linear velocity obtained by applying the principle of conservation of linear
momentum.

Question 9
This questions tests students’ understanding on how the energy transfer takes place to maximise the
angular velocity of the rod after the collision. Similar to Question 8, another foreseen alternative Marks
conception on this is to think that the rod should not translate at all after the collision, such that the
rotational kinetic energy of the rod can be maximised.

𝑉 = 0 does not maximise 𝜔 0.5

Teacher’s commentary. Student gives the correct initial indication that the rod’s centre should not
stay at rest in order to maximise its angular velocity after the collision. However, he/she does not
describe the method of maximisation used as well as how the final linear velocities of the particle
and rod are related.

𝑣 = 𝑉 in order to maximise 𝜔 1.0

Teacher’s commentary. Student expresses the correct relation of the final velocities of the particle
and rod after the collision in order to maximise the rod’s angular velocity. Nevertheless, the
reasoning as to why they must be equal is missing.

261
 In the centre of mass frame, the total linear momentum of the system must be zero.
To maximise, 𝜔, both the particle and rod must come to rest after the collision.
This implies 𝑣 = 𝑉 in the original frame of reference.
Conservation of linear momentum: 1.5
𝑚𝑢 = 𝑚𝑣 + 𝑀𝑉
𝑚
𝑣=𝑉= 𝑢
𝑚+𝑀
Teacher’s commentary. Student expresses the full explanation required. As described in Question 8,
the analysis in the centre of mass frame is beneficial for the analysis of this questions. The particle
and rod must move together after the collision in order to produce the maximum angular velocity of
the rod. In addition, the student also expresses the final linear velocity of both the particle and rod
obtained by applying the conservation of linear momentum.

Question 10
This questions tests students’ understanding on how the energy transfer takes place to maximise the
angular velocity of the rod after the collision. Similar to Questions 8 and 9, another foreseen Marks
alternative conception on this is to think that both the particle and rod should not translate at all after
the collision, such that the kinetic energy of the particle is wholly transformed to the rotational kinetic
energy of the rod.

𝑣 = 𝑉 = 0 does not maximise 𝜔 0.5

Teacher’s commentary. Student gives the correct initial indication that both the particle and rod’s
centre should not be at rest after the collision in order to maximise the rod’s angular velocity.
However, he/she does not describe the method of maximisation used as well as how the final linear
velocities of the particle and rod should be related.

If 𝑣 = 0 then 𝑉 cannot be zero, and vice versa, if 𝑉 = 0 then 𝑣 cannot be zero. 1.0
Teacher’s commentary. Student describes a good analysis on the final velocities of the particle and
rod. Nevertheless, he/she does not describe the reason of his/her mathematical argument.

From conservation of linear momentum,

𝑚𝑢 = 𝑚𝑣 + 𝑀𝑉 1.5
If 𝑣 = 𝑉 = 0, conservation of momentum is violated as 𝑢 ≠ 0.
Teacher’s commentary. Student describes the required mathematical argument using physics
concept. If both the particle and rod come to rest after the collision, then conservation of momentum
is violated.

Question 11
This is the ultimate question in this assessment as the answer is counterintuitive to most physical Marks
intuitions. The question tests on whether students are able to use their mathematical thinking to
approach the given problem involving their conceptual understanding on conservation laws.

ℓ
If 𝑥 = , 𝜔 is not maximised. 0.5
2
Teacher’s commentary. Student gives a correct reasoning that the rod’s angular velocity will not be
maximised if the point of collision is at the rod’s end. Unfortunately, it is still far from the required
full explanation as the mathematical or physical reasoning on why it is so is missing.

262
 ℓ
Let 𝑥 = , and then observe the following equations for conservation laws,
2
𝑚𝑢 = 𝑚𝑣 + 𝑀𝑉
1 1.0
𝑚𝑢2 = 𝑚𝑣 2 + 𝑀𝑉 2 + 𝑀ℓ2 𝜔2
12
Hence, letting 𝑣 = 𝑉 does not maximise 𝜔.
Teacher’s commentary. Students makes a good argument by using conservation of linear momentum
and conservation of energy. Letting the particle and rod move together after the collision is also a
good step. However, no sufficient reasoning is found as to why the angular velocity is not
maximised. This reasoning can be completed by using the conservation of angular momentum. One
should be able to see a contradicting mathematical statement if the point of collision is at the rod’s
end.

From the conservation laws and by letting 𝑣 = 𝑉,

𝑚
𝑚𝑢 = 𝑚𝑣 + 𝑀𝑉 = (𝑀 + 𝑚)𝑉 → 𝑣 = 𝑉 = 𝑢
𝑀+𝑚
12𝑉𝑥 1 1 12𝑉𝑥 2
𝜔 = 2 → 𝑚𝑢2 = 𝑚𝑣 2 + 𝑀𝑉 2 + 𝑀ℓ2 𝜔2 = 𝑚𝑣 2 + 𝑀𝑉 2 + 𝑀ℓ2 ( 2 )
ℓ 12 12 ℓ
2
12𝑥 2 𝑚𝑢 2
𝑚𝑢 = [𝑀 (1 + 2 ) + 𝑚] ( )
ℓ 𝑀+𝑚
12𝑥 2 12𝑥 2
𝑀2 + 2𝑀𝑚 + 𝑚2 = 𝑀𝑚 (1 + 2 ) + 𝑚2 → 𝑀 + 𝑚 = 2 𝑚 1.5
ℓ ℓ
ℓ 𝑀
𝑥= √1 +
2√3 𝑚
Since it is known that 𝑀 < 2𝑚,
ℓ 𝑀 ℓ
𝑥= √1 + <
2√3 𝑚 2
Teacher’s commentary. Student presents the full analysis required to approach the question. Using
all the conservation laws, the location of the point of collision that will maximise the rod’s angular
velocity can be derived. Although this method is not unique, but the mathematical analysis is
sufficient to bring him/her to the correct answer and reason. Moreover, he/she uses the fact that 𝑀 <
2𝑚 to show that this point of collision is still within the rod.

Question 12
The question tests on students’ understanding of the conservation of linear momentum. Although
the concept is simple, but the existence of rotational motion adds another layer of difficulty. In this Marks
case, students should pay attention to the existence of a pivot that causes the system to be non-
isolated.

The existence of a pivot causes nonzero change in the system’s momentum. 0.5
Teacher’s commentary. Although it is correct to say that the change in the system’s momentum is
due to the existence of a pivot, but this student does not describe the physics concept that bridges the
two statements. Newton’s law is required to show that the nonzero change in the system’s
momentum is due to the existence of an external force from the pivot.

𝑝initial = 𝑚𝑢
𝑝final = 𝑚𝑣
1.0
𝑣 ≠ 𝑢 → 𝑝final ≠ 𝑝initial
There is a nonzero change in the momentum of the system.

263
Teacher’s commentary. The mathematical formalism is sufficient to show that the momentum of the
system does change after collision occurs. This is shown by the fact that the velocity of the particle
has to change after the collision with the rod. The only thing missing in this argument is the Newton’s
law to show that the change in the system’s momentum is due to the external force from the pivot.

System is not isolated due to the existence of the pivot, hence ∑ 𝐅⃗ext ≠ ⃗𝟎⃗
This implies a nonzero change in the system’s momentum. 1.5
Hence, conservation of momentum cannot be used.
Teacher’s commentary. The argument is well described. Newton’s law dictates that the external
force from the pivot causes a nonzero change in the system’s momentum. Hence, in this scenario,
which also applies for subsequent questions, conservation of momentum cannot be applied.

Question 13
The question tests on students’ understanding of the conservation of angular momentum, particularly Marks
on the choice of system and the reference point used. The concept is slightly more difficult than
Question 5 due to the existence of a pivot.

∑ ⃗𝛕⃗ext = ⃗𝟎⃗ 0.5

Teacher’s commentary. Student expresses that the net external torque is zero, but he/she does not
show sufficient conceptual understanding on the reason as he/she does not link it back to the
question. The reason on why the total torque is zero is also missing.

∑ ⃗𝛕⃗ext = ⃗𝟎⃗ → 𝐋
⃗ = constant 1.0
Teacher’s commentary. Student expresses the required condition for conservation of angular
momentum. As there is no net torque acting on the system, the angular momentum of the system is
conserved. However, he/she does not describe why the total torque is zero, nor the mathematical
formalism of the conservation of angular momentum.

System is not isolated due to the existence of a pivot, hence ∑ 𝐅⃗ext ≠ ⃗𝟎⃗.
However, since the rod’s centre is taken as the reference point, ∑ ⃗𝛕⃗ext = ⃗𝟎⃗.
This implies conservation of angular momentum only about that point. 1.5
Taking the rod’s centre to be the reference point,
1
𝑚𝑢𝑥 = 𝑚𝑣𝑥 + 𝑀ℓ2 𝜔
12
Teacher’s commentary. Student expresses complete understanding on the concept. Firstly, the
system of the particle and rod is not isolated due to the existence of a pivot, and hence, the net force
acting on it is nonzero. However, this force acts on the rod’s centre; as the rod’s centre is taken as
the reference point, the net torque acting on the system is still zero. This zero net torque implies
conservation of angular momentum of the system. No term corresponding to translational angular
momentum of the rod exists in the equation as the rod’s centre is fixed. In addition, student correctly
expresses the equation for conservation of angular momentum in terms of the given variables and
constants.

Question 14
Marks
This is a conceptual question on elastic collision.

∆𝐾 = 0 0.5

264
Teacher’s commentary. Student expresses that the kinetic energy should not change in this collision.
Although this equation can be applied for the given situation, but it lacks of understanding as to why
it can be applied.

The initial kinetic energy of the particle is distributed to its final kinetic energy and the rotational
kinetic energy of the rod.
Mathematically, 1.0
1 1 1
𝑚𝑢2 = 𝑚𝑣 2 + 𝐼𝜔2
2 2 2
Teacher’s commentary. Student expresses a certain level of understanding of the physical quantity
that should be constant, particularly in this collision, but it still lacks of understanding as he/she does
not mention the reason why kinetic energy should be conserved and why there is no term
corresponding to the translational kinetic energy of the rod.

elastic → 𝐾initial = 𝐾final

1 1 1
→ 𝑚𝑢2 = 𝑚𝑣 2 + 𝐼𝜔2 as 𝑉 = 0
2 2 2
Simplifying, 1.5
1
𝑚𝑢2 = 𝑚𝑣 2 + 𝑀ℓ2 𝜔2
12
Teacher’s commentary. Student expresses complete understanding that the kinetic energy of the
system is conserved only if the collision is elastic. He/she also expresses the mathematical formalism
for the conservation of energy based on the given variables and constants. Moreover, he/she explains
that the translational kinetic energy of the rod is missing as the rod’s centre is fixed.

Question 15
This questions tests students’ understanding on how the energy transfer takes place to maximise the Marks
angular velocity of the rod after the collision. The existence of a pivot makes the analysis a little
different with Question 8.

𝑣 = 0 maximises 𝜔 0.5
Teacher’s commentary. Student gives the correct initial indication that the particle should come to
rest in order to maximise the angular velocity of the rod after the collision. However, he/she does
not describe the method of maximisation used.

Since 𝑉 = 0, 𝑣 must also be zero in order to maximise 𝜔 1.0

Teacher’s commentary. Student expresses the correct statement with regards to the final velocities
of the particle and rod after the collision in order to maximise the rod’s angular velocity.
Nevertheless, it still lacks of clarity as the reasoning is still missing.

The initial kinetic energy of the particle is distributed between its final kinetic energy and the final
rotational kinetic energy of the rod. In this case, the translational kinetic energy of the rod is zero
as the rod’s centre is fixed.
Mathematically, 1.5
1
𝑚𝑢2 = 𝑚𝑣 2 + 𝑀ℓ2 𝜔2
12
To maximise, 𝜔, it is clear that 𝑣 must be zero.
Teacher’s commentary. Student expresses the full explanation required. Conservation of energy
dictates the final velocity of the particle to be zero in order to maximise the rod’s angular velocity.

265
 Question 16
Together with Question 11, this is the ultimate question in this assessment as the answer is
counterintuitive to most physical intuitions. The question tests on whether students are able to use Marks
their mathematical thinking to approach the given problem involving their conceptual understanding
on conservation laws.

ℓ
If 𝑥 = , 𝜔 is not maximised. 0.5
2
Teacher’s commentary. Student gives a correct reasoning that the rod’s angular velocity will not be
maximised if the point of collision is at the rod’s end. Unfortunately, it is still far from the required
full explanation as the mathematical or physical reasoning on why it is so is missing.

By observing the conservation of energy,

1
𝑚𝑢2 = 𝑚𝑣 2 + 𝑀ℓ2 𝜔2 1.0
12
𝑣 = 0 maximises 𝜔.
Teacher’s commentary. Students makes a good argument by using conservation of energy and a
correct statement that the particle should come to rest to maximise the angular velocity of the rod.
However, the reasoning that connects it back to the question is missing. No argument is made
whether the rod should be hit at its end for this case.

Applying conservation laws,

1 1
𝑚𝑢𝑥 = 𝑚𝑣𝑥 + 𝑀ℓ2 𝜔 → 𝑚𝑥(𝑢 − 𝑣) = 𝑀ℓ2 𝜔
12 12
1 1
𝑚𝑢2 = 𝑚𝑣 2 + 𝑀ℓ2 𝜔2 → 𝑚(𝑢 − 𝑣)(𝑢 + 𝑣) = 𝑀ℓ2 𝜔2 → 𝑢 + 𝑣 = 𝜔𝑥
12 12
ℓ 4𝑢 𝜔ℓ 𝑀
𝑥= →𝜔= ,𝑣 = (1 − )
2 𝑀 4 3𝑚
(1 + )ℓ
3𝑚
Since it is known that 𝑀 < 3𝑚, this results in nonzero value of 𝑣, which means some of the kinetic
energy is taken by the particle. And this should not be the condition for the maximum angular
speed of the rod. 1.5
1 3𝑚 𝑢
𝑚𝑢𝑥 = 𝑀ℓ2 (2√ )
12 𝑀 ℓ

ℓ 𝑀
𝑥= √
2 3𝑚
Since it is known that 𝑀 < 3𝑚,
ℓ 𝑀 ℓ
𝑥= √ <
2 3𝑚 2
Teacher’s commentary. Student presents the full analysis required to approach the question. Firstly,
he/she shows that hitting the rod at its end does not maximise its angular velocity as the kinetic
energy will be partially taken by the particle after the collision. Secondly, using all the conservation
laws, he/she derives the location of the point of collision that will maximise the rod’s angular
velocity. Thirdly, he/she uses the fact that 𝑀 < 3𝑚 to show that this point of collision is still within
the rod.

266
6.3.2.5.Analysis of Interviews

It is expected that students’ conceptual understanding will be captured not only in

their answer and reason, but also in the fifth tier in the form of equivalent mathematical

language (refer to Table 6.13). Additionally, the students’ mathematical thinking is excerpted

from the interviews that serve as a triangulation of the findings.

Below are partial transcripts of the interviews. Students 21 and 22 were examples of

students who were awarded partial marks for Question 1 (0.5 and 1.0 respectively out of 1.5

marks). Though both showed some understanding, but Student 21 showed very limited

understanding when it comes to mathematical expression as he solely expressed the

relationship between torque and force. During the interview, the interviewer asked the

student to elaborate on his explanation so as to get more detailed excerpts of his

understanding.

Interviewer: “How does your equation lead to your answer and reason?”

Student 21: “I saw that there is a force and there should be a torque because the

force is applied at a certain distance away from the centre, so the rod should rotate.”

Interviewer: “The question actually asks if the centre of the rod will move and your

answer is true. May I know why?”

Student 21: “Because there is a net force acting on the whole object.”

Interviewer: “Am I right to say that there should be further steps to explain your

answer?”

Student 21: “Yeah…”

Interviewer: “What kind of equation?”

267
 Student 21: “𝐹 = 𝑚𝑎… it is Newton’s law.”

Interviewer: “The equation also asks if the rod rotates about its centre… so what

is the equivalent equation you can apply here?”

ℓ
Student 21: “Torque equals to 𝐹 times 𝑟, which is 2, and after that this torque

equals to 𝐼 times 𝛼… and because there is torque, so there will be angular

acceleration, so the rod will rotate about the centre of mass.”

Although Student 22’s mathematical explanation is worth more marks, further elaboration is

still required. In the interviews, he managed to present the full explanation after a little

guidance from the interviewer. He even managed to perform an immediate analysis for a

further question related to the problem.

Interviewer: “Your mathematical explanation does not really lead to your choice

of answer there. The question is asking first of all whether the rod will move, right?

Can you explain to me why?”

Student 22: “Because of Newton’s second law, I wrote 𝐹 = 𝑚𝑎. Since there are

no opposing forces, and F is the only force acting on this rod, I can conclude that the

rod will move across table… and then the torque will cause it to rotate.”

Interviewer: “… What is the mathematics behind the rotation then? Like how fast

will it rotate?”

Student 22: “Depending on how long you apply the force. If you apply the force

for longer period of time then the rod will rotate faster.”

268
 Interviewer: “How would you be able to determine the angular acceleration then?”

Student 22: “About the centre of mass then we use the formula: torque equals to

the moment of inertia times the angular acceleration.”

Interviewer: “What will happen if the force is applied at the centre of the rod?”

Student 22: “The centre of mass will still move because Newton’s second law still

applies. And if you are pushing at the centre, by symmetry argument the rod can’t

rotate either way. ”

In contrast, Student 23 was able to show his thorough conceptual understanding while

explaining what he wrote as the mathematical explanation for Question 1. He even managed

to answer a further question from the interviewer without showing any sign of alternative

conceptions.

Student 23: “… there is a torque in the system, and since 𝐼 is a nonzero value, that

means 𝛼 is nonzero, so it will start rotating. For 𝐹 = 𝑚𝑎, it’s because there is a net

force on the system, so there will also be an acceleration in a similar manner. So

force 𝐹 acts as an external force that also produces an external torque…”

Interviewer: “What if the force is applied at the centre of the rod, for example?”

Student 23: “It will be moving forward but it will not be rotating, because there is

no net torque on the system but there is still a force.”

269
 Below are some other transcripts for the interviews related to Question 3. Students

24 and 25’s mathematical explanations are worth 0.5 and 1.0 marks respectively out of 1.5.

Both students managed to arrive at the correct answer and reason, but Student 24 displayed

very limited understanding in his mathematical formalism for the question. However, during

the interviews both students could actually perform the full analyses for this question and

even performed excellent analysis for the additional question after being prompted.

Student 24: “… when you increase the length, both the torque and inertia will

increase. As you can see that the torque is proportional to length, while the inertia is

to length squared. In other words, when you increase the length, the torque increases

at a slower rate than the inertia, the angular acceleration will be slower.”

Interviewer: “Okay, that is because you are assuming that the force is constant.

How much force is required to produce the same angular acceleration then?”

Student 24: “Okay, when you increase the length by a factor of 2, the torque

increases by a factor of 2 and the moment of inertia increases by a factor of 4. In

order to maintain the same ratio, you must increase the torque by 4. So I would say

twice the force is required.”

Student 25 was awarded more marks as he performed a little further analysis on the test paper

as compared to Student 24. He managed to perform equivalent analysis during the interview.

Student 25: “When we double the length of the rod and the mass stays constant,

the Newton’s second law for translation does not change, but for rotation the distance

270
 from the edge of the rod to the centre of the rod increases, so the torque about the

centre of the rod increases by a factor of 2. However, the moment of inertia about the

centre of the rod quadruples since it has length squared. So the angular acceleration

decreases by a factor of 2. Therefore, we know that even though the amount of torque

increases, rotating the rod will be more difficult since it has higher moment of

inertia.”

Interviewer: “So I guess the equations you wrote there are insufficient, you could

actually write further according to what you said, such as the moment of inertia

quadruples as the length is multiplied by a factor by 2. So how much force is needed

if you want to produce the same amount of angular acceleration?”

Student 25: “Double the force… you need to double the force to cancel the factor

of 2 in the expression of acceleration.”

Interviewer: “So that the torque will be quadrupled as well?”

Student 25: “Yes.”

In contrast, Student 26 managed to elaborate very well on the mathematical formalism for

Question 3 in the test paper and he was awarded full marks (1.5 marks). As expected, he was

also able to provide equivalent analysis during the interview, including for a further question.

Student 26: “Since the length of the rod is doubled, and the force still acts on the

rod’s end, the torque by the force increases by a factor of 2. Since the moment of

inertia increases by a factor of 4, so it increases by a greater factor than the torque,

so the angular acceleration will decrease by a factor of 2.”

271
 Interviewer: “So how much force is needed to if we want to produce the same

amount of angular acceleration?”

Student 26: “I would say two times the force.”

Interviewer: “May I know why?”

Student 26: “Because now with double the length the torque has increased by a

factor of 2, so we need to double the force in order to have the same amount of

angular acceleration.”

6.3.2.6.Analysis of Correlations

A strong positive correlation was found between students’ performance in the answer

tier and their performance in the reason tier (𝑟(124) = .735, 𝑝 < .001), indicating that

students with stronger reasoning ability are more likely to attain high scores for the answer

tier. Furthermore, participants’ scores in the fifth tier would be more informative to showcase

their mathematical thinking. A closer look at Table 6.11 reveals that the percentage of

students who are able to perform mathematical thinking, or translate their answer/reason to

mathematical expressions is quite low, much lower than the percentage of students who are

able to obtain correct answer and/or reason. However, the findings still reveal that there are

significant positive correlations between the participants’ scores in the mathematical

expression tier and their scores in the answer tier (𝑟(124) = .447, 𝑝 < .001), as well as that

between the participants’ scores in the mathematical expression tier and their scores in the

reason tier (𝑟(124) = .594, 𝑝 < .001). The moderate strength of these correlations suggests

that students’ ability to perform mathematical thinking may be a good indicator of their

272
ability to achieve correct answers and reasons in a physics test. This could be a useful input

for physics teachers in general to focus on enhancing students’ mathematical thinking ability

in order to advance their conceptual understanding, especially when dealing with high-

performing students. However, variability in the test scores can still be observed due to other

factors, such as students’ overreliance on scientific intuition as well as their inability to

demonstrate mathematical thinking.

6.3.2.7.Analysis of Alternative Conceptions in Rotational Mechanics

Table 6.14 shows a list of 11 alternative conceptions on rotational mechanics

documented in our study. The mean confidence for the alternative conceptions ranges from

3.0 to 4.7. Based on the classification scheme of Caleon & Subramaniam (2010b), only six

of these alternative conceptions are spurious while the rest are genuine. The largest

percentages of students showing indication of an alternative conception occurs for Questions

11 and 16, with percentages of 58.7% and 59.5%, respectively.

Table 6.14

Alternative Conceptions on Rotational Mechanics

% sample
Question with Mean
No. Alternative conception
number alternative confidence
conception

A force acting on a rigid body not at its centre of

1 Q1 BD 10.3 4.5
mass will only cause rotation and not translation.

Rotational inertia of a rigid body remains constant Q3 AA 30.2 4.1

2
if its translational inertia is constant. Q3 BC 14.3 3.5

273
 In a rigid body collision, conservation of linear
3 Q4 AA 11.1 4.4
momentum holds only when the collision is elastic.

Q4 BD 45.2 4.2
Q5 BD 36.5 3.9
Q6 BD 33.3 3.5
Q8 AA 11.1 3.4
In a rigid body collision, linear momentum can be
4 Q9 BA 11.9 3.4
transformed/transferred into angular momentum.
Q10 AA 10.3 3.6
Q12 BD 49.2 4.2
Q13 BD 35.7 3.8
Q15 AA 23.0 3.6

In a rigid body collision, conservation of angular Q5 AE 11.9 3.8

5
momentum holds with no required conditions. Q13 AE 11.9 3.5

A system is isolated as long as no net torque acts

6 Q6 AB 11.9 3.4
on the system.

Conservation of kinetic energy in a system holds Q7 AB 15.1 4.7

7
only when the system is isolated. Q14 AB 11.9 3.6

In a rigid body collision, the best case scenario for Q8 AB 16.7 3.4
8 kinetic energy transfer is when one body comes to Q9 BB 12.7 3.4
rest after the collision. Q10 AB 12.7 4.0

Q8 BC 12.7 3.6
In an elastic rigid body collision, there will always Q9 BC 15.9 4.1
9 be relative motion between the centres of mass of Q10 BD 28.6 3.6
the two bodies after the collision. Q15 BC 12.7 3.6
Q15 BD 18.3 3.0

Hitting a free rod at its end causes the rod to rotate Q11 AE 58.7 4.1
10
with maximum angular speed. Q16 AE 59.5 4.3

The existence of a pivot does not contribute to an

11 Q13 AB 14.3 4.2
external force on the system.

As some of the alternative conceptions listed in Table 6.14 are greatly overlapping in terms

of their underlying concepts, we have categorised these items into six classes. Many of these

concepts are usually presumed by physics students, especially high-performing ones, but this

study reveals that these concepts need special attention.

274
(1) Newton’s second law for translation and rotation (Item No. 1)

The concept of Newton’s second law needs to be understood for both translation and rotation.

For a rigid body, translation is defined as the motion of its centre of mass, and rotation is

defined as the motion about its centre of mass. The centre of mass of a rigid body will move

with an acceleration as long as there is a net external force acting on the body. Likewise, the

body will rotate with an angular acceleration as long as there is a net torque acting on the

body about its centre of mass. In this manner, a force may cause both translation and rotation,

and both of them should be treated separately. It is an alternative conception to think that: A

force acting on a rigid body not at its centre of mass will only causes rotation and not

translation. This alternative conception was harboured by 10.3% of the participants with

mean confidence of 4.5, indicating a genuine alternative conception.

(2) Factors affecting rotational inertia (Item No. 2)

Translational inertia is different from rotational inertia. Translational inertia, which is better

known as “mass”, is a property of a physical body which measures an object’s resistance to

changing its state of motion when a force is applied. By definition, the centre of mass of a

system, which is located at the average of the weighted position vector of all constituent

masses, can be used as a representative of the whole rigid body. In other words, the

translational motion of the body is represented by the motion of its centre of mass. However,

it is an alternative conception to think that rotational inertia of a rigid body remains constant

if its translational inertia is constant, also known as “moment of inertia”. This is because the

moment of inertia of a rigid body equals the sum of the moments of inertia of all its

275
constituent particles and each particle may have different distances to the rotational axis, i.e.

the centre of mass. In the case of a rod, doubling its length will quadruple its moment of

inertia, thus affecting its angular acceleration as well. About 30.2% of the students (with

mean confidence 4.1) mislooked this concept and 14.3% of the students (with mean

confidence 3.5) failed to see the dependence of angular acceleration on the rod’s length.

(3) Required conditions for conservation laws (Items No. 3, 5-7, 11)

It was observed that between 11.1% and 14.3% of the participants had difficulties

understanding the concepts related to conservation laws, with regards to the respective

conditions that need to be met. For instance, in Question 4, 11.1% of the participants (with

mean confidence 4.4) think that in a rigid body collision, conservation of linear momentum

applies only when the collision is elastic, which is clearly an alternative conception. For both

Questions 5 and 13, 11.9% (with mean confidence 3.8) of the participants thought that in a

rigid body collision, conservation of angular momentum applies with no required conditions,

indicating that this alternative conception was harboured by the same group of students.

Conservation of linear momentum requires the system to be either isolated, such as

when there are no external forces acting on the system, or as long as the net external force

acting on the system is zero. In the case of a rigid body collision, this law is to be applied for

the translational motion of the bodies, i.e. the motion of the centre of mass of the rigid bodies.

The presence of a pivot provides an external force on the system, and hence causes the linear

momentum of the system to change during the collision. Likewise, conservation of angular

momentum requires the system to be either isolated, such as when there are no external

torques acting on the system, or as long as the net torque acting on the system is zero. The

276
latter could happen if, for instance, the point of reference is taken to be at the point of

collision, such that the external force acting on the system does not produce a torque. About

33.3% of the participants failed to see that conservation of angular momentum could be

applied on the rod (without the particle) about the point of collision. This failure could be

due to the rod being at rest initially but moving after the collision. Finally, only in a perfectly

elastic collision can conservation of energy be applied. This has nothing to do with the system

being isolated as energy can still be conserved even when the body is pivoted.

(4) Linear momentum vs. angular momentum (Item No. 4)

Although the participants could be classified as high-performing students with good

conceptual understanding in physics, having learnt the topic of rotational mechanics does not

guarantee the absence of simple alternative conceptions. Up to 49.2% of the participants

(with mean confidence 4.2) think that the linear momentum of an object could be transformed

or transferred to the angular momentum of another object. Hence, in the case of a projectile

hitting a stationary rod, a high percentage of students harboured the alternative conception

that in a rigid body collision, linear momentum can be transformed/transferred into angular

momentum.

Linear momentum is defined mathematically as the product of mass and velocity

⃗⃗ = 𝑚𝐯⃗⃗; for a rigid body, the velocity of its centre of mass will be used instead. In
vector, 𝐩

contrast, angular momentum of a point mass is defined mathematically as the vector cross

⃗ = 𝐫⃗ × 𝐩
product of position vector and linear momentum vector, 𝐋 ⃗⃗. For a rigid body, it

⃗ =𝐑
consists of the translational and rotational terms, 𝐋 ⃗⃗⃗ × 𝐏
⃗⃗ + 𝐼CM 𝛚 ⃗⃗, 𝐏
⃗⃗⃗⃗, where 𝐑 ⃗⃗, 𝐼CM and 𝛚
⃗⃗⃗⃗

277
denote the centre of mass position vector, the centre of mass momentum vector, moment of

inertia of the body about its centre of mass and the angular velocity vector of the body about

its centre of mass respectively. Thus, linear momentum and angular momentum are two

totally different physical quantities with different dimensions, and it is impossible for one to

be converted to the other. In the abovementioned case, the linear momentum of the projectile

can only be partially or fully transferred to the linear momentum of the rod, and the angular

momentum of the projectile can only be partially or fully transferred to the angular

momentum of the rod.

(5) Relative motion between two rigid bodies vs. kinetic energy transfer (Items No. 8-9)

Viewing a collision from the centre of mass frame is often a good way to understand the

concept of collision. In a collision between two particles, for instance, the two particles would

appear to approach each other before collision in the centre of mass frame, as the total

momentum of the system has to be zero. If the collision is perfectly elastic, both particles

will bounce off after collision, maintaining their original speeds, as this is the only way to

conserve energy while maintaining the system’s total momentum as zero. In this way, the

magnitude of the relative velocity between the two particles is preserved in an elastic

collision, both in the centre of mass frame and in the original frame, as relative velocity is

invariant across inertial frames. If the collision is perfectly inelastic, in the centre of mass

frame both particles will stop moving after collision, as all energy will be dissipated while

maintaining zero total momentum. This means that both particles will stick together and

move with equal velocity after collision in the original frame. Equivalently, the concept of

coefficient of restitution can be used for a more direct analysis (Ricardo & Lee, 2015).

278
 However, in a rigid body collision the analysis is a little different, and it is not correct

to think that in an elastic rigid body collision, there will always be relative motion between

the centres of mass of the two bodies after the collision, an alternative conception harboured

by up to 28.6% of the participants (with mean confidence 3.6). This is because translational

kinetic energy can be transformed into rotational kinetic energy while conservation of

angular momentum provides another constraint. In fact, for a collision between a projectile

and a stationary rod, to produce maximum rotational angular speed of the rod, translational

kinetic energy needs to be transformed as much as possible to the rotational one. This

happens when the centre of mass of each body comes to a stop after collision, as viewed in

the centre of mass frame. In the original frame, the velocity of the centre of mass of both

bodies would be identical. Though this analysis could be considered as higher order thinking,

as it requires multiple layers of thinking, the concepts involved are quite basic and high-

performing students are expected to be able to perform such analyses. If one wants to use the

concept of coefficient of restitution, it has to be redefined as the conditions are different from

the collision between two point masses (Ricardo, 2016).

Another common alternative conception, harboured by up to 16.7% of the

participants, is to think that the projectile has to come to rest after the collision with a

stationary rod, in order for the rod to have maximum angular speed after the collision. This

will either violate the conservation of linear momentum or not maximise the translational

kinetic energy conversion to the rotational one. The former occurs when the rod’s final linear

momentum is not equal to the projectile’s initial linear momentum, and the latter occurs when

they are equal. However, viewing the latter case from the system’s centre of mass frame

279
would make it clear that this is not the best-case scenario as there is still relative velocity

between the projectile and the rod’s centre of mass.

(6) Maximum torque (Item No. 10)

This last class of alternative conception is one harboured by most students, even in the pilot

study. While it is conceptually correct to think that maximum torque is required to rotate a

rod such that its angular speed is maximum, in a collision between a projectile and a

stationary rod, it is an alternative conception to think that hitting a free rod at its end causes

the rod to rotate with maximum angular speed. This has been shown by the above analysis.

In Question 11, 58.7% of the students (with mean confidence 4.1) harboured this alternative

conception, and in Question 16, 59.5% of the students (with mean confidence 4.3) harboured

the same alternative conception. This result confirms the case presented by Ricardo (2016)

as intuition or qualitative analysis would commonly fail to analyse such a case, as compared

to mathematical analysis.

6.4. Teaching Intervention in Promoting Mathematical Thinking

As elaborated in the previous chapters, the teaching intervention was done to promote

mathematical thinking in physics conceptual learning. Should there be conceptual changes

taking place, it would be an additional benefit in the use of mathematical thinking in physics.

The intervention was packaged as a special revision. Although it was framed as a revision,

the instructor used the framework of teaching intervention strategies for promoting

mathematical thinking in all the steps. At the beginning of the lesson, the instructor asked the

students to recall what they remembered about the topic discussed, and most students were

280
able to recall their previously-learnt concepts, as expected. The relatively small size of the

class created a supportive environment for students’ learning and the instructor did not have

difficulties engaging the whole class in general.

The class size for the intervention on rotational mechanics was slightly larger than

that for the study on momentum/collision; and hence, better classroom management was

needed throughout. Nevertheless, the participants in this class were attentive for the entire

two-hour lesson, due to their slightly older age. Students were also quite excited throughout

the lesson as the revision would be useful in their physics exam at the end of the semester. A

number of students did ask questions in the middle of the lesson to clarify some concepts.

At the end of the lesson, students were asked to form groups of 3-4 students (for the

study on momentum/collision) or 4-5 students (for the study on rotational mechanics) for

group discussions over some pre-assigned practice questions. The discussion was also

expected to enhance students’ mathematical thinking skills through explanations, arguments

and presentations of their answers and reasoning. Whiteboarding was one of the pedagogies

that was embedded in the lesson intervention, especially during the group discussions.

Students were expected to express their train of mathematical thoughts on the whiteboards

provided for each group. All group members could easily refer to and argue the expressions

written on the whiteboards, and during the group presentations every group could easily show

to other groups their group answers written on the whiteboard in mathematical form.

6.4.1. Students’ Reflections on Lesson Intervention

After the lesson intervention, participants in the experimental group were asked to

complete a survey on the lesson. A small number of them were also requested to write some

281
commentaries and/or reflections on their views about the lesson. The comments were

generally positive, confirming the benefits of the teaching intervention strategy in teaching

high-performing physics students. Table 6.15 shows the categories extracted from their

feedbacks with sample excerpts. The key phrases and sentences captured indicated the

strength of the teaching intervention. From the excerpted phrases and sentences, it was clear

how mathematical thinking benefited the participants, not only in their test performance but

also in their perception to perform deeper analyses for other physics problems they would

encounter in the future in school. While mathematics may decrease students’ interest in

physics, in the population of high-performing students in this study it does not seem to do so.

High performers were able to appreciate the significance of even the simplest mathematical

formalism of physics concepts and to form a close connection between the qualitative and

mathematical representation of the concept. It has to be noted that engaging in mathematical

thinking also involves creative thinking and innovative ways to address a physics problem.

The excerpted phrases and sentences showed the strong benefits that the experimental group

of students experienced after the teaching intervention. Apart from improving their test

performance, it was clear that creativity and confidence were two main aspects that have been

enhanced by this intervention.

282
 Table 6.15

Excerpts from Students’ Reflections on Teaching Intervention (𝑵 = 𝟏𝟏)

283
284
285
6.4.2. Survey Analysis

6.4.2.1.Study on Momentum/Collision

For the study on momentum/collision, out of the 59 participants in the experimental

group, 37 of them (62.7%) responded to the survey. The reliability (Cronbach’s alpha) of the

survey form was .71, indicating that it was well above the acceptable range of ≥ .7 (Nunnaly,

1978). Table 6.16 summarises the responses of the students to the survey statements.

The results showed that the students were positive about their experiences, with all

participants indicating the importance of mathematical thinking in physics, especially in the

study of momentum/collision, and that they could understand the topic better if mathematics

is involved. About 91.9% of the respondents indicated that they would like to explore more

problems on momentum/collision if mathematics is involved. About 97.3% of the

respondents considered the lesson intervention to be useful, and the same percentage would

consider the discussion led by the teacher to be helpful for the application of mathematical

thinking in problem-solving. Interestingly, the same percentage of students, i.e. 97.3%,

indicated that their confidence level in using mathematical thinking in physics had improved

after the teaching intervention.

286
Table 6.16

Responses to Survey Statements on the Topic of Momentum/Collision (𝑵 = 𝟑𝟕)

Overall SA/A N D/SD

No. Items
mean ± 𝑆𝐷 (%) (%) (%)

On the topic of momentum and collision

I can understand the topic of momentum / collision better if

1 4.6 ± .5 100.0 .0 .0
mathematics is involved.

Mathematical analysis is crucial in the study of momentum /

2 4.7 ± .5 100.0 .0 .0
collision in mechanics.

Mathematical analysis helps me to understand a number of

3 4.6 ± .6 94.6 5.4 .0
concepts on momentum / collision better than using intuition.

I like to explore more problems in momentum / collision with

4 4.4 ± .6 91.9 8.1 .0
the involvement of mathematics.

I prefer to learn the topic of momentum / collision

5 3.6 ± 1.0 10.8 27.0 62.2
qualitatively without any use of mathematics.

On the lesson intervention

I find the intervention lesson on momentum / collision to be

6 4.5 ± .6 97.3 2.7 .0
useful.

I can now better appreciate why mathematics is needed for

7 4.5 ± .6 94.6 5.4 .0
understanding concepts on momentum / collision.

The practice problems gave me valuable experience in using

8 4.4 ± .6 94.6 5.4 .0
mathematical thinking to understand momentum / collision.

The discussions led by the teacher have helped me to apply

9 mathematical thinking in analysing problems in momentum 4.5 ± .8 97.3 .0 2.7
/ collision.

The discussions with peers on the practice problems in

10 momentum / collision have helped me to apply mathematical 4.0 ± .9 73.0 21.6 5.4
thinking in analysing these problems.

After this session, I am more confident of using

11 mathematical thinking in analysing problems in momentum 4.3 ± .5 97.3 2.7 .0
/ collision.

287
 For the free response section in the survey form, generally the students’ responses

were not varied enough to merit embarking on an extensive coding scheme. The feedback

was on the programme, which was a common experience, and there were limited components

that they could describe. Some participants in the experimental group did not complete the

survey form. Although both hardcopy and softcopy versions of the survey were provided for

them, it is likely that fatigue could have set in as they seem to have some difficulties in

expressing their thoughts.

The first free response question is “What is it about the topic of momentum / collision

that makes it difficult to understand?” This question strongly focuses on the topic of

discussion. The type of responses can be categorised as follows:

(1) Vagueness or abstractness of the concept without mathematical tools

Students might have imagined the concept and tried to build a connection with real world

contexts. A few students might even confuse momentum with energy or force. When an

object moves fast, its momentum and kinetic energy are high. When an object collides with

another object, the momentum and kinetic energy get transferred. Hence, it is quite common

that students try to relate the two physical quantities as they have learnt energy much earlier

than momentum. Moreover, momentum is not a visible quantity that can be measured directly

like speed or acceleration.

288
(2) Failure of intuitive or qualitative reasoning

This response was expected as students were exposed to a very useful tool, i.e. mathematical

thinking. As expected from high-performing students, one respondent wrote “The intuition

or physical thinking doesn’t really feel natural to me. I would often prefer to try to solve

equations than to convince myself this or that would happen.”, emphasising how crucial

performing mathematical thinking in physics is.

(3) Analysing momentum in different frame

Having learnt the concept of relative velocity, performing physical analysis in a different

inertial frame of reference is still non-trivial for students. Although working in the centre-of-

mass frame would help them make sense of elastic and inelastic collisions, many would

probably prefer to perform the analysis in the original frame.

The second free response question is “What do you like or dislike about the use of

mathematical thinking on the topic of momentum / collision?” This question addresses their

experience in the assessment as well as the lesson intervention. The responses were generally

positive. The type of responses could be categorised as follows:

(1) Better conceptual understanding

It is intriguing to see how mathematical thinking enhances one’s conceptual understanding.

One student noted “I like that mathematical thinking can make understanding the physical

concepts easier rather than visualising it.” and another student noted “Mathematical

thinking allows me to understand the topic better and remember the understanding every

289
time I review the topic.” This means, using mathematical thinking makes the discussion on

different types of collisions more holistic and even intuitive after they are used to it.

Mathematics helps one to see what conditions would change when one or more variable(s)

change(s).

(2) Less memorisation

Contrary to popular belief, to high-performing students, using mathematical thinking in

approaching a physics concept does not require more effort. In fact, one student noted “There

is less to memorise if mathematical thinking is applied.” and another noted “Being able to

solve problems fully mathematically is useful due to the fact that it can be generalised into

many situations including different dimensions.” In other words, when mathematical

thinking was involved in conceptual learning, students considered it as having less to

memorise as a lot of things could be derived by performing mathematical thinking.

(3) Increased confidence and creativity

Mathematics is indeed a reliable tool as compared to our own intuition, as indicated in a

student’s remarks, “I think mathematical thinking is a good way to analyse momentum as

collision as it is a more accurate way of solving the problem rather than intuition which may

not always be correct.” It can be used to get the result as precisely as possible, to deeply and

thoroughly analyse a physical situation, and to expand the analysis to more general situations.

As a direct consequence, students’ confidence with regards to the topics learnt would

improve, as noted by another student, “I like that it gives confidence to one's answer and

express terms quantitatively.” Apart from these, the use of mathematical thinking also

290
enhances one’s creativity as they may use different mathematical tools they have learnt, or

even exhaust them, to have multiple approaches to solve a question.

However, some students also opined that performing mathematical thinking could be

unsatisfying when the mathematics gets too complicated and cannot be translated back to

their qualitative understanding. One student noted “It sometimes does not tally with my

conceptual understanding and I do find it frustrating when I am unable to debug the error

on time.”

6.4.2.2.Study on Rotational Mechanics

For the study on rotational mechanics, out of the 65 participants in the experimental

group, 43 of them (66.2%) responded to the survey. Similar to the other main study, the

reliability (Cronbach’s alpha) of the survey for the lesson intervention was 0.89, indicating

that it was well above the acceptable range of ≥ .7 (Nunnaly, 1978). Table 6.17 summarises

the responses of the students to the survey statements.

In general, students responded positively to the new approach in this revision lesson.

About 88.4% of the respondents gained confidence in analysing physics problems after the

use of mathematical thinking in the topic, and 76.7% would like to explore more problems

involving mathematical thinking. There were only 32.6% of the participants who preferred

to learn the topic qualitatively without any use of mathematics, though all of them (100%)

were able to appreciate why mathematics was needed for conceptual understanding in

relation to rotational mechanics.

291
Table 6.17

Responses to Survey Statements on the Topic of Rotational Mechanics (𝑵 = 𝟒𝟑)

Overall SA/A N D/SD

No. Items
mean ± 𝑆𝐷 (%) (%) (%)

On the topic of rotational mechanics

I can understand the topic of rotational mechanics better if

1 4.2 ± .9 90.7 2.3 7.0
mathematics is involved.

Mathematical analysis is crucial in the study of rotational

2 4.2 ± .7 90.7 4.7 4.7
mechanics in mechanics.

Mathematical analysis helps me to understand a number of

3 4.0 ± 1.1 81.4 9.3 9.3
concepts on rotational mechanics better than using intuition.

I like to explore more problems in rotational mechanics with

4 4.0 ± 1.0 76.7 11.6 11.6
the involvement of mathematics.

I prefer to learn the topic of rotational mechanics

5 3.2 ± 1.2 32.6 16.3 51.2
qualitatively without any use of mathematics.

On the lesson intervention

I find the intervention lesson on rotational mechanics to be

6 4.3 ± .8 88.4 4.7 7.0
useful.

I can now better appreciate why mathematics is needed for

7 4.4 ± .5 100.0 .0 .0
understanding concepts on rotational mechanics.

The practice problems gave me valuable experience in using

8 4.2 ± .8 88.4 7.0 4.7
mathematical thinking to understand rotational mechanics.

The discussions led by the teacher have helped me to apply

9 mathematical thinking in analysing problems in rotational 4.3 ± .6 90.7 9.3 .0
mechanics.

The discussions with peers on the practice problems in

10 rotational mechanics have helped me to apply mathematical 3.8 ± .9 74.4 14.0 11.6
thinking in analysing these problems.

After this session, I am more confident of using

11 mathematical thinking in analysing problems in rotational 4.1 ± .8 88.4 7.0 4.7
mechanics.

292
 Similar to the analysis for the study on momentum/collision, for the free response

section in the survey form, we classified the responses into a few categories. Some

participants in the experimental group did not fill up the survey form. Although both

hardcopy and softcopy versions of the survey were provided to them, it was likely that fatigue

may have set in or that these students had some difficulties in expressing their thoughts

clearly.

The first free response question is “What is it about the topic of rotational mechanics

that makes it difficult to understand?” This question strongly focuses on the topic of

discussion. The type of responses could be categorised as follows:

(1) The concept of rotational mechanics itself is not so intuitive

Student pointed out that mathematics is very crucial for understanding the concept of

rotational mechanics, even the simplest one. This includes vector analysis, visualisation of

all physical vector quantities in three dimensions as well as understanding the definitions of

physical quantities involved, e.g. angular momentum. Hence, the analysis of the system’s

behaviour is usually not very intuitive.

(2) Confusion with translational mechanics

Some students managed to understand the concept of rotational motion but confused it with

the translational one. For example, students might associate “linear momentum” and “angular

momentum” in their mind, even though these two physical quantities are not equivalent.

(3) Hard to visualise and relate to real world contexts

293
To many students, even after solving a question on the topic of rotational mechanics, it is

still hard for them to relate it back to real life situations that they can visualise easily. For

instance, it is easier to visualise “mass” as compared to its equivalence in rotation – “moment

of inertia”.

The second free response question is “What do you like or dislike about the use of

mathematical thinking on the topic of rotational mechanics?” This question addresses their

experience in the assessment as well as on the lesson intervention. The type of responses

could be categorised as follows:

(1) Logical and reliable approach towards solving the problem

Based on the excerpts of some students’ remarks, “I like that it (mathematical thinking)

provides a logical approach to solve these problems.” and “It (Mathematical thinking) made

the prediction of the motion more convincing.” To this pool of students, mathematics

provides more logical approaches towards problem-solving as it enables one to analyse the

physical situation mathematically. The results obtained are more reliable as compared to

qualitative analysis.

(2) Mathematics makes it easier to understand the concept and simplifies the problem

One student noted “Maths (Mathematics) makes concepts much easier to understand” and

another noted “I like the use of mathematical thinking on the topic because it is rather

straightforward.” Mathematics provides more comprehensive understanding of the concepts

and simplifies problems on rotational mechanics. Avoiding mathematics in learning this

294
concept should therefore be discouraged, especially in the sample of high-performing

students.

(3) Mathematics enables one to analyse different cases

Not only does one’s understanding become more comprehensive, mathematics also provides

one with the alternative analysis of a given situation as well as the ability to generalise the

case. This, subsequently, will enable one to analyse for different cases. This was shown in

one of the students’ remarks, “I like it because with mathematical thinking I can analyse

different cases of rotational mechanics answers and get more precise answers rather than

just understanding concepts like ‘the speed decreases’ or ‘angular frequency decreases’.”

The only dislike in the students’ responses is when the mathematics becomes complicated as

sometimes multiple equations need to be solved concurrently, e.g. equations for translation

and rotation. Failure to connect it back to qualitative understanding would encourage students

to simply memorise the formulas. This was shown in the following students’ remarks, “There

are too many equations to use when working on a question using mathematical approach.”

and “Solving multiple mathematical equations may be a bit too difficult.”

6.4.3. Post-Test Analysis

6.4.3.1.Study on Momentum/Collision

Out of 123 participants, 64 of them were placed in the control group (8 females; mean

age ± 𝑆𝐷 = 16.47 ± .71) while 59 of them were put in the experimental group (8 females;

mean age ± 𝑆𝐷 = 16.56 ± 1.34). The mean ± 𝑆𝐷 pre-test score for the control group is 34.0%

295
± 17.1% while that for the experimental group is 39.3% ± 15.8%. The pre-test scores

comparison reveals that the mean test scores (combined scores for answer, reason and

mathematical expression tiers) of both groups are not significantly different (𝑡(121) = 1.78,

𝑝 = .08).

From the results of the pre-test, we have seen how the test questions in the Chain-

Collision Assessment are generally difficult (mean FI = .62 for the answer tier, .36 for the

answer and reason tiers) and discriminating (mean DI = .39 for the answer tier, .38 for the

answer and reason tiers), even for the high-performing physics students. The results of the

reliability test based on Cronbach Alpha also reveals acceptable values for the internal

consistency of the assessment (𝛼 = .81 when total scores were considered, 𝛼 = .94 for the

confidence ratings). All these results show the suitability of using the Chain-Collision

Assessment to measure changes in scores due to the teaching intervention.

Table 6.18 presents the results of the post-test scores for both the experimental and

control groups as well as the statistical comparison between the performances of both groups.

When only answer tier is considered, 25.0% of the control group participants were able to

score 70% and above as compared to 94.9% of the experimental group participants. When

both answer and reason tiers are considered, 12.5% of the control group participants were

able to score 70% and above as compared to 88.1% of the experimental group participants.

This shows significant improvements in the scores of the answer and reason tiers for the

experimental group after the teaching intervention. With the inclusion of all tiers, none of the

control group participants were able to score 70% and above. In contrast, a large number of

296
the experimental group participants (79.6%) were able to do so. As mentioned in the pre-test

scores analysis, the scores in the fifth tier indeed affected the whole performance greatly.

The t-test analysis reveals a significant difference between the control group post-test

scores and experimental group post-test scores. Comparing only the scores for the answer

tier, 𝑡(121) = 12.54, 𝑝 < .001, indicates a significantly higher performance of the

experimental group as compared to the performance of the control group. When both the

answer and reason tiers are considered, 𝑡(121) = 15.74, 𝑝 < .001, it also indicates a

significant difference between the performance of both groups. Significant difference

between the control and experimental groups were also found for the total score (𝑡(121) =

19.47, 𝑝 < .001), in favour of the experimental group, as expected.

297
 Table 6.18

Cognitive Scores for Chain-Collision Assessment (Post-Test)

298
 Tables 6.19 and 6.20 present the data for the confidence ratings analysis of the control

group and experimental group, respectively. The mean ± 𝑆𝐷 confidence of the students in

the control group for the post-test was 4.5 ± .2 out of 6.0, which was approximately equal to

the mean ± 𝑆𝐷 confidence of the students in their pre-test performance, i.e. 4.5 ± .4

(𝑡(28) = −.0097, 𝑝 = .99. The mean ± 𝑆𝐷 confidence of the students in the experimental

group for the post-test was 5.4 ± .2 out of 6.0, indicating significant difference from the mean

confidence of the students in the control group (𝑡(28) = 10.10, 𝑝 < .001). This suggests to

us an increase in the confidence level of students after undergoing the lesson intervention, on

top of the improvement in their test scores. The same trend is also observed, even when

individual tiers was considered.

The mean ± 𝑆𝐷 CFC for the control group was 4.9 ± .5 while that for the

experimental group was 5.4 ± .2, indicating significant differences between them (𝑡(28) =

4.66, 𝑝 < .001). Similarly, the mean ± 𝑆𝐷 CFW for the control group was 4.3 ± .3 while

that for the experimental group was 4.9 ± .5, indicating significant differences between them

(𝑡(28) = 3.79, 𝑝 < .01). The same trend was also observed, even when individual tiers was

considered. These high values suggest to us that after the lesson intervention, the students in

the experimental group became more confident over concepts they understood but

unfortunately also over concepts they did not understand. Nonetheless, for the experimental

group, the CB yielded very small values – close to zero for all questions, thus indicating

students’ better accuracy in terms of knowing what they knew and what they did not know.

The CB for the control group, however, yielded larger values, on average, as compared to

299
the pre-test. This shows students’ overconfidence in the accuracy of their responses, which

is a common trend, as indicated in Boekaerts & Rozendaal (2010).

CDQ values for the experimental group yield a negative value for Question 11

whereas for the control group, it yields negative values for Questions 12 and 15. As negative

CDQ means students are more confident when they are wrong than when they are correct,

this finding indicates that a few conceptual changes have occurred after the lesson

intervention. However, there is no significant differences between the control group’s and

experimental group’s mean CDQ for the whole test (𝑡(28) = .60, 𝑝 = .55).

300
 Table 6.19

Analysis of Cognitive and Confidence Measures per Item (Chain-Collision Assessment, Control Group Post-Test)

301
 Table 6.20

Analysis of Cognitive and Confidence Measures per Item (Chain-Collision Assessment, Experimental Group Post-Test)

302
 Promoting mathematical thinking in physics lessons was seen to be effective in

helping students develop better understanding. As such, students in the experimental group

were able to at least engage in deeper reasoning when approaching a physics question. The

experimental group participants attempted to perform mathematical thinking on 66.7% of the

test paper. Although the number is still not close to 100%, it is much better than the control

group participants who only attempted to perform mathematical thinking on 22.1% of the

test. This finding can be attributed to the teaching intervention.

Table 6.21 shows the control group and experimental group’s performances in the

mathematical expression tier, respectively, and how the scores for this tier affected the total

score for each question. For the control group, it can still be observed that the lowest

percentages of students who showed some understanding in their mathematical formalism

occurred for Questions 13, 14 and 15, similar to the pre-test result. In contrast, the scores for

these questions improved significantly for the experimental group, indicating that some

conceptual changes have occurred after the teaching intervention. These conceptual changes

will be elaborated in the next section.

303
Table 6.21

Students’ Performance in Mathematical Expression Tier across Questions (Chain-Collision

Assessment, Control Group vs. Experimental Group Post-Test)

% students who obtained Average score (%) Average total score (%)
Average combined score
Question non-zero score in the obtained in the obtained in the answer,
(%) obtained in the answer
number mathematical expression mathematical expression reason and mathematical
and reason tiers
tier tier expression tiers

Control Experimental Control Experimental Control Experimental Control Experimental

Group Group Group Group Group Group Group Group

1 12.5 78.0 12.0 76.8 68.8 93.2 44.4 86.2

2 3.1 44.1 2.1 42.9 76.6 96.6 44.6 73.6

3 7.8 76.3 6.3 74.0 57.0 92.4 35.3 84.5

4 3.1 66.1 3.1 65.5 35.9 76.3 21.9 71.7

5 12.5 86.4 9.9 84.2 48.4 89.0 31.9 86.9

6 15.6 88.1 12.5 87.6 58.6 96.6 38.8 92.7

7 9.4 88.1 5.7 88.1 17.2 87.3 12.3 87.7

8 17.2 93.2 13.0 93.2 28.9 94.9 22.1 94.2

9 10.9 71.2 9.9 70.6 49.2 76.3 32.4 73.8

10 14.1 86.4 13.0 85.9 46.1 96.6 31.9 92.0

11 15.6 84.7 14.1 81.4 51.6 89.0 35.5 85.7

12 14.1 79.7 2.1 68.9 34.4 79.7 25.4 78.9

13 3.1 72.9 2.1 68.9 28.9 78.0 17.4 74.1

14 3.1 69.5 2.1 68.9 20.3 83.9 12.5 77.5

15 .0 52.5 .0 50.8 10.2 65.3 5.8 59.1

Average 9.5 75.8 7.2 73.9 42.1 86.3 27.5 81.2

Referring to Table 6.22, the analysis of the control group performance shows that the

same alternative conceptions as discovered in the pre-test were still largely present in the

304
students during the post-test, signifying that the students did not gain any conceptual changes

without any meaningful intervention. In contrast, the students in the experimental group

cleared up most of the alternative conceptions harboured in the pre-test. Out of the

previously-listed 10 alternative conceptions, only 3 alternative conceptions on the concept of

momentum/collision remained for the experimental group. The mean confidence for these

alternative conceptions ranges from 4.3 to 5.5. This shows that these 4 alternative

conceptions are persistent alternative conceptions as they all can be categorised as genuine

ones, i.e. higher than 10% of the students in the group, based on Caleon & Subramaniam

(2010b). Nevertheless, a large reduction in the percentage of students harbouring these

alternative conceptions can be observed from the table, showing the efficacy of the

intervention. These alternative conceptions indeed require special attention by physics

teachers and students, as high-performing students still harbored these alternative

conceptions even after acquiring the habit of performing mathematical thinking.

For the first category of alternative conceptions, which is on conservation of

momentum, 11.9% of the students in the experimental group still thought that the equation

for conservation of momentum applies only when the collision is elastic. Although the

percentage is lower than the one in the pre-test analysis for the experimental group, i.e.

16.9%, it shows that these students had been taking principle of conservation of momentum

for granted, without thoroughly understanding its conceptual underpinnings. Likewise, for

the second category of alternative conceptions, which is on the concept of centre of mass –

10.2% of the students attained the correct answer but with incorrect reasoning. Students

understood that the total momentum in the centre of mass frame equals to zero, but they failed

to realise that this concept is true even for two-dimensional or three-dimensional motions.

305
This percentage, however, is lower as compared to 16.9% in the pre-test analysis for the

experimental group.

The third category of alternative conceptions, which is the concept of relative velocity

in elastic and inelastic collisions, had been cleared after the teaching intervention. The

mathematical thinking performed by students was expected to play an important role in

clearing this alternative conception. This translated to clearing of the fourth category of

alternative conception subsequently, which covers a number of questions in the Chain-

Collision Assessment, and students were able to analyse momentum transfer and kinetic

energy transfer mathematically to a good extent. Via thorough mathematical analysis,

students were expected to conclude that momentum transfer and kinetic energy transfer are

constrained by the equations of conservation of momentum and that the system’s kinetic

energy should not increase.

Although the students might have understood about elastic collision being the best

case scenario for momentum transfer or kinetic energy transfer, the physical situation

presented in Question 15 requires another layer of thinking since another block is inserted in

between the two blocks. It is very interesting to note that 20.3% of the students still used

100% as a reference for maximum amount of kinetic energy transfer. However, there is a

significant decrease in the percentage from 45.8% in the pre-test analysis for the experimental

group.

306
 Table 6.22

Alternative Conceptions on Momentum/Collision (Chain-Collision Assessment Pre-Test vs. Post-Test)

307
6.4.3.2.Study on Rotational Mechanics

Table 6.23 presents the relevant findings for the study on rotational mechanics. Out

of 126 participants, 65 students were randomly selected to be in the experimental group (9

females; mean age ± 𝑆𝐷 = 17.08 ± 1.00) and the remaining 61 students in the control group

(12 females; mean age ± 𝑆𝐷 = 17.25 ± 1.23). The mean ± 𝑆𝐷 pre-test score for the control

group is 32.5 ± 13.6% while that for the experimental group is 32.5 ± 12.4%. The mean pre-

test scores (sum of scores for the answer, reason and mathematical expression tiers) for both

groups are not significantly different (𝑡(124) = 0.0092, 𝑝 = .99).

This main study on rotational mechanics shows similar trends as the other study on

momentum/collision. The analysis of the pre-test shows that the questions in the Rigid Body

Collision Assessment are generally difficult (mean FI = .64 if only the answer tier was

included, .33 when the answer and reason tiers were included) and discriminating (mean DI

= .32 if only the answer tier was included, .34 when the answer and reason tiers were

included) even for the high-performing physics students. The results of the reliability test,

based on Cronbach Alpha, reveals acceptable values for the internal consistency of the test

instrument (𝛼 = .88 when total scores were considered, 𝛼 = .92 for the confidence ratings).

Hence, this test is a suitable assessment to excerpt students’ understanding as well as measure

the effect of the teaching intervention.

Since the control group and experiment group did not perform significantly

differently in the pre-test, from a statistical significance standpoint we could compare their

performance of both groups directly in the post-test (see Table 6.23) for the intervention.

308
When only answer tier is considered, only 23.0% of the control group participants were able

to score 70% and above, as compared to 72.3% of the experimental group participants. When

both answer and reason tiers are considered, only 8.2% of the control group participants were

able to score 70% and above as compared to 75.4% of the experimental group participants.

This shows significant improvements in the scores of the answer and reason tiers for the

experimental group after the teaching intervention. Similar to the pre-test scores analysis, the

scores in the fifth tier indeed affected the overall performance greatly. With the inclusion of

all tiers, 1.7% of the control group participants were able to score 70% and above, in contrast

to a large number of the experimental group participants (66.2%).

309
 Table 6.23

Cognitive Scores for Rigid Body Collision Assessment (Post-Test)

310
 The t-test analysis reveals that the experimental group participants performed

significantly better as compared to the control group participants for the post-test. Comparing

only the scores for the answer tier, 𝑡(124) = 11.07, 𝑝 < .001, indicates a significant

difference between the performances of both groups. When both answer and reason tiers are

considered, 𝑡(124) = 12.25, 𝑝 < .001, again significant difference between the

performances of both groups was seen. With the inclusion of the mathematical expression

tier, a significant difference between scores of the control and experimental groups was also

found (𝑡(124) = 19.11, 𝑝 < .001), as expected.

Tables 6.24 and 6.25 present the respective confidence ratings analysis for the control

group participants and experimental group participants. The mean ± 𝑆𝐷 confidence of the

students in the control group for the whole test was 4.1 ± .3 out of 6.0, which was

approximately equal to the mean confidence of the students in their pre-test performance, i.e.

4.1 ± .4 (𝑡(30) = −.16, 𝑝 = .88). For the experimental group, the mean ± 𝑆𝐷 confidence

for the post-test was 4.8 ± .4 out of 6.0, indicating a significant difference from the mean

confidence of the students in the control group (𝑡(30) = 5.97, 𝑝 < .001). Similar to the main

study on momentum/collision, the lesson intervention not just led to an improvement in the

test scores, but also suggested an increase in confidence level of students. The same trend is

also observed even when individual tiers was considered.

The mean ± 𝑆𝐷 CFC for the control group for the post-test was 4.4 ± .6 and for the

experimental group, it was 5.0 ± .3, indicating significant differences between them

(𝑡(30) = −3.84, 𝑝 < .001). Similarly, the mean CFW for the control group was 3.8 ± .2

311
and for the experimental group, it was 4.5 ± .4, indicating significant differences between

them (𝑡(30) = 5.33, 𝑝 < .001). The same trend was also observed, even when individual

tier was considered. These high values suggest an increase after the intervention in the

students’ confidence level – not only in the concepts they understood but also those they did

not understand. Interestingly, the CB for the experimental group also yields very small

values, close to zero, for all questions, indicating students’ better accuracy in terms of

knowing what they knew and what they did not know. The CB for the control group yields

larger values, on average, as compared to those in the pre-test. This shows students’

overconfidence in the accuracy of their responses, which is a general trend, as indicated in

Boekaerts & Rozendaal (2010). The same trend was also observed in the main study on

momentum/collision.

Interestingly for the post-test, the CDQ for the control group still yields a negative

value for Questions 16, which is the main higher order thinking question in this assessment,

and the value had become positive for the experimental group. As negative CDQ means

students are more confident when they are wrong than when they are correct, this finding

indicates some conceptual changes have occurred after the lesson intervention. However,

there is no significant difference between the control group’s and experimental group’s mean

CDQ for the whole test (𝑡(30) = .10, 𝑝 = .92).

312
 Table 6.24

Analysis of Cognitive and Confidence Measures per Item (Rigid Body Collision Assessment, Control Group Post-Test)

313
 Table 6.25

Analysis of Cognitive and Confidence Measures per Item (Rigid Body Collision Assessment, Experimental Group Post-Test)

314
 Working on another topic, i.e. rotational dynamics to promote mathematical thinking

in physics enables students in the experimental group to further develop the habit of

performing deeper analysis of physical situations using mathematical thinking. This is

reflected by their performance in the post-test.

Table 6.26 shows the control group’s and experimental group’s performances in the

fifth tier respectively, and how the scores for this tier affected the total score for each

question. For the control group, it can still be observed that the lowest percentages of students

who showed some understanding in their mathematical formalism occurred for Questions 13,

14 and 15, similar to the pre-test result. In contrast, the scores for these questions improved

significantly for the experimental group, indicating some conceptual changes have occurred

after the teaching intervention. These conceptual changes will be elaborated in the next

section.

315
Table 6.26

Students’ Performance in Mathematical Expression Tier across Questions (Rigid Body

Collision Assessment, Control Group vs. Experimental Group Post-Test)

% students who obtained Average score (%) Average total score (%)
Average combined score
Question non-zero score in the obtained in the obtained in the answer,
(%) obtained in the answer
number mathematical expression mathematical expression reason and mathematical
and reason tiers
tier tier expression tiers

Control Experimental Control Experimental Control Experimental Control Experimental

Group Group Group Group Group Group Group Group

1 3.3 86.2 3.3 82.6 79.5 94.6 46.8 89.5

2 19.7 76.9 18.0 74.4 91.8 80.8 60.2 78.0

3 16.4 87.7 14.2 82.1 42.6 85.4 30.4 84.0

4 3.3 90.8 3.3 90.8 32.8 98.5 20.1 95.2

5 3.3 87.7 3.3 86.7 37.7 86.9 23.0 86.8

6 3.3 76.9 3.3 75.9 32.0 85.4 19.7 81.3

7 0.0 64.6 0.0 63.6 72.1 83.9 41.2 75.2

8 3.3 72.3 2.2 68.2 42.6 73.9 25.3 71.4

9 1.6 75.4 1.6 70.8 59.8 76.9 34.9 74.3

10 3.3 76.9 3.3 76.9 59.0 87.7 35.1 83.1

11 1.6 67.7 1.6 59.5 14.8 52.3 9.1 55.4

12 1.6 83.1 1.6 83.1 48.4 87.7 28.3 85.7

13 3.3 83.1 3.3 81.0 37.7 83.9 23.0 82.6

14 1.6 56.9 1.6 55.9 63.1 75.4 36.8 67.0

15 1.6 56.9 1.6 55.4 36.9 53.9 21.8 54.5

16 3.3 64.6 1.6 55.4 13.1 49.2 8.4 51.6

Average 4.5 75.5 4.0 72.6 47.8 78.5 29.0 76.0

316
 The control group performance showed a strong indication of alternative conceptions

in the post-test, and these were largely the same alternative conceptions as displayed in the

pre-test. For the experimental group, most of these alternative conceptions were no longer

there. However, there were still some persistent alternative conceptions, as shown in Table

6.27. Comparing it with the pre-test alternative conception analysis, it can be seen that this

is a significant improvement. The mean confidence for the alternative conceptions ranges

from 3.3 to 5.2. Interestingly, based on the five categories of alternative conceptions

identified in the pre-test, the experimental group of students cleared two of them. These

alternative conceptions indeed require special attention by physics teachers and students, as

high-performing students still showed these alternative conceptions even after acquiring the

habit of performing mathematical thinking.

Out of the six categories of alternative conceptions identified in the pre-test, the

experimental group of students significantly cleared three of them after developing the habit

of mathematical thinking through the teaching intervention. Firstly, is on the application of

Newton’s law in translation and rotation. Secondly, is on the factors affecting moment of

inertia of a rigid body. Thirdly, is on the confusion between linear momentum and angular

momentum. These conceptual changes were indeed remarkable as there were in total 49.2%

of the participants harbouring the confusion between linear momentum and angular

momentum in the pre-test. However, they managed to overcome this alternative conception

by looking at the dimensions of both physical quantities.

The required conditions for conservation laws are among the persistent alternative

conceptions that still existed even for the experimental group. About 18.5% of the students

in this group still thought that kinetic energy of the system would be conserved only if the

317
system was isolated. Whereas 10.8% of them thought that the energy would still be conserved

even in an inelastic collision as now the energy loss could be transformed to rotational kinetic

energy. About 13.8% of the participants could not differentiate between “isolated system”

and “non-isolated system where the external force does not produce a net torque”. Although

these two statements yield an identical implication – that is, conservation of angular

momentum –, they refer to two different physical conditions.

Regarding kinetic energy transfer, 12.3% of the experimental group of participants

thought that maximum transfer would occur if the particle stops moving after the collision so

as to give off its entire kinetic energy. As elaborated before, this is clearly an alternative

conception as there would still be relative motion between the particle and the rod. Also,

18.5% of the participants still thought that to exert maximum torque on the rod, the particle

has to hit the rod at one of its ends. This is a drastic drop in percentage as compared to 59.5%

in the pre-test with the same alternative conception. All these persistent alternative

conceptions should be addressed by physics teachers, especially during discussion of higher

order thinking problems.

318
 Table 6.27

Alternative Conceptions on Rotational Mechanics (Rigid Body Collision Assessment Pre-Test vs. Post-Test)

319
6.5. Chapter Summary

In summary, our study showed several interesting results. The NUS High School

students’ performance in the FCI, MBT and CSEM revealed that, in general, they could be

classified as high-performing students that were suitable for us as subject participants for our

main studies. The two main studies generally reported similar results using the newly-

constructed instruments. The instruments were indeed able to excerpt students’ mathematical

thinking in relation to their conceptual understanding. High-performing students were shown

to have limitations in performing mathematical thinking and expressing their mathematical

thoughts. However, their ability to perform mathematical thinking correlates strongly with

their conceptual understanding. This finding was also supported by the interviews, where

students’ mathematical thinking largely showed the depth of their understanding, and

students’ alternative conceptions also surfaced out. Lastly, our strategy for teaching

intervention was seen to be effective for this group of high-performing students. This strategy

strongly promotes also the use of mathematical thinking in physics conceptual learning. The

effectiveness of this strategy was shown by students’ improvement in the fifth tier scores

which led to the improvement in the post-test results over the pre-test. This effectiveness was

also shown in the excerpts of students’ reflections and survey feedback. An additional benefit

of the teaching intervention could be observed in the analysis of alternative conceptions in

the post-test vs. the pre-test.

320
 CHAPTER 7

DISCUSSION

To study mathematical thinking in physics, several stages are involved in this

research. The theoretical explorations done in Chapters 2 and 3 as well as the educational

studies done in Chapters 5 and 6 were to address our research questions, as it will be

elaborated below. In this chapter, all the findings from our study as well as answers to the

research questions will be expounded.

As the first step of our educational study was to assess the suitability of our sample,

several conceptual surveys were administered to students in the NUS High School of

Mathematics and Science in Singapore. Our study revealed that the students in our sampling

frame could be categorised as high-performing students as they excelled in several physics

conceptual surveys, i.e. Force Concept Inventory (FCI), Mechanics Baseline Test (MBT) and

Conceptual Survey in Electricity and Magnetism (CSEM). The administration of these test

instruments was done to four batches of students and the consistency in the results enabled

us to generalise it to the whole school population. The students’ performance was indeed

highly skewed to the left and as such, their test performance might not be able to give us any

interesting data, for example, correlation between their test scores and another cognitive

aspect might be attenuated. Ceiling effect was also foreseen should the same test instruments

be used to measure any improvement due to an intervention.

The excerpts of students’ performance in FCI, MBT and CSEM motivated us to

investigate possible reasons why although our student sample scored considerably highly,

many of these high-performing students did not manage to gain the maximum score when

321
there were some questions that even the top scorers found difficult. As suggested by Beichner

(1994), interviews were then conducted with some of the participants to provide stronger

arguments to support the findings obtained. Through the interviews, we realised that only a

few of them actually made errors related to alternative conceptions, while most of the errors

were due to carelessness (e.g. the student misread the question or the value) and negligence

(e.g. the student did not realise that the variable needed to solve the problem was calculable

or the student did not think through thoroughly – indicating lack of ability to perform

mathematical thinking). As such, even prior to the execution of the main study, interestingly

we found a consistent pattern where students’ performance would have been better if they

had developed the habit of performing mathematical thinking and did not rely too much on

their scientific intuition. Admittedly, this finding was not robust, and this needed to be

supported by a comprehensive study design for the main study. In conclusion, the interviews

showed that students were able to identify their own mistakes and make amendments to the

mistakes, suggesting that they actually had sufficient conceptual understanding to answer the

questions correctly. However, the excerpts of students’ mathematical thinking could hardly

be analysed using the abovementioned instruments and hence, they were not sufficient to

address our research questions.

The MCQ format in the abovementioned conceptual surveys had its own limitations

if it were to be adopted in our main study instruments. Firstly, “guessing” might contribute

to the error variance and reduce the reliability of the test, and, as such, an assumption that all

participants were engaged and conscientious throughout the test had to be made. Secondly,

the options presented in MCQ format did not provide deep insights into students’ thinking or

conceptual understanding. As there were usually limited options in an MCQ test, students

322
were prevented from constructing, organising and presenting their own answers (Chang et

al., 2010). With all these limitations, together with the prevalent ceiling effect in the students’

performance in conceptual surveys, it was mandatory that we did not use the same test

instruments nor the same format for our constructed test instruments in our main studies.

7.1. Development of Five-Tier Instrument to Excerpt Students’ Mathematical

Thinking in Physics

This section will elaborate on our study results in relation to our first research

question.

Research Question 1: “How can a five-tier diagnostic instrument that builds on

previous formats of two-tier, three-tier and four-tier instruments be developed for

the specific purpose of this study, and what are the issues in its development?”

As the excerpt of students’ mathematical thinking was an important aspect explored

in the educational aspect of this study, higher order thinking physics questions were preferred

over the usual problem-solving questions. This was because higher order thinking questions

necessitated a more thorough thinking process which was to be excerpted. Also, in our study

we would like to see the correlation between the degree of students’ conceptual

understanding and their ability to perform mathematical thinking. As such, a suitable format

needed to be adopted for our test instrument in order to excerpt students’ mathematical

thinking and, at the same time, their conceptual understanding. The available conceptual

323
surveys in the literature were short on this aspect and a new instrument format needed to be

constructed for the main studies.

The 5-tier test instrument was developed to overcome the limitations of the previous

formats of 2-tier, 3-tier and 4-tier instruments. Firstly, we acknowledged the advantage of

MCQ format over free-response format used in the pilot study as the marking could be done

consistently and more objectively over a large number of participants. Although free-

response format was used by previous researchers (Andersson & Karrqvist, 1983; Colin et

al., 2002; Langley et al., 1997; Palacios et al., 1989; Ronen & Eylon, 1993; Wittman, 1998)

as it gave adequate scope for participants to think and write about their ideas; however,

evaluating the results was not easy without a rigorous marking scheme (Al-Rubayea, 1996).

After several rounds of feedback, a multi-tier format was considered better as a range of

measures could be excerpted from the analysis – for example, test reliability, confidence

ratings analysis, test statistics, correlation tests across tiers, alternative conceptions, etc.

These benefits provided robustness in the analyses, for instance, guessing could be minimised

by the analysis of the reason tier, which was usually done by interviews (Hrepic, 2004; Tamir,

1989). Furthermore, students’ alternative conceptions could be explored from their mistakes

not only from the answer tier, but also from the reason tier, or both.

We also acknowledged the advantages of 3-tier and 4-tier tests over 2-tier test in

minimising false positive and false negative results. These formats had elevated the reliability

of a test instrument as students could be ranked more accurately via their responses in the

confidence ratings. Students with good conceptual understanding would generally be the

ones who gave correct answer and correct reason with high confidence ratings. Students with

324
alternative conceptions would be the ones who generally gave wrong answer and/or wrong

reason with a high confidence rating.

The main crux of the newly-constructed 5-tier instruments is in the additional tier it

provides on top of the existing 4 tiers. While the answer, reason and confidence tiers have

been used by some researchers in a range of topics in science education, none of these formats

were able to extract aspects of students’ cognition (related to mathematical thinking)).

Indeed, it is a very difficult task to make all the options in the answer and reason tiers

mathematically-oriented. If this were to be done, students would have a tendency to choose

formulas they might have memorised or they are familiar with. As such, students’ responses

might not be a true reflection of their mathematical thinking. Therefore, another tier is

necessary to be added to capture and unpack the information on students’ mathematical

thinking, and the free-response format was more suitable for this additional tier. This 5-tier

test enables the researcher to perform more robust analyses that include all the analyses that

can be done in a 4-tier test plus in the open-ended section of the fifth tier. As such, the fifth

tier, which is the mathematical expression tier, is novel and is the strength of the newly-

constructed instrument that was used in the two studies. This tier requires students to

articulate their answer and/or reason mathematically for each question. In this way, the

students’ mathematical thinking could be seen and analysed.

The results of our main studies showed various analyses that could be performed via

the excerpt of students’ responses in the fifth tier. First of all, it could be seen how students’

level of conceptual understanding tallied with the mathematical formulation of their thinking.

Secondly, we could also observe students’ mathematical thinking process in higher order

325
thinking questions. These results could be further triangulated through interviewing the

students with regards to their mathematical thinking.

As the 5-tier instrument was built upon previous formats of 2-tier, 3-tier and 4-tier

instruments, all the limitations in these multi-tier instruments are also manifested in this

format as well. To address these limitations to some extent, the design of our main study test

instruments was carefully looked into. Sreenivasulu and Subramaniam (2013) surfaced that

students’ choice of response in the answer tier may influence their choice of response in the

reason tier. To overcome this issue, some questions in the test instrument on

momentum/collision were designed to be in True/False format for the answer tier. All

questions in the test instrument on rotational mechanics were also in True/False format in the

answer tier. Moreover, five options were provided in the reason tier for each question in the

instruments with one additional option of “Other reason:”, where students were allowed to

use their own reasoning should they not agree with any of the five options provided. This

approach would minimise the possibility of the options in the reason tier leading to the correct

option in the answer tier. Furthermore, the test duration was expected to be longer than all

the previous formats as there were five aspects the participants were required to respond to

for each question. To address this issue, the number of questions in each instrument was kept

to an acceptable number, i.e. 15 questions for the first test instrument and 16 questions for

the second one.

Apart from the development of the 5-tier test instruments, the administration and

analysis of the instruments were not without limitations and issues. In both main studies, it

could be seen that many student participants were not able to respond especially for the fifth

tier, indicating gaps in translating their conceptual understanding to mathematical thinking.

326
At the same time, the disconnect between students’ mathematical thinking and qualitative

explanation was also observed as some failed to translate their mathematical thinking back

to qualitative descriptions. The length of a 5-tier instrument imposed another issue as the

cognitive and metacognitive processes that took place when answering the questions were

very demanding. As such, fatigue might take place during the test. Lastly, this study, like

other previous studies, had its own issues in which missing data were more likely found in

the students’ responses. In any multi-tier test, failure to respond to one of the tiers would

invalidate the students’ responses in the other tiers and thus affect analyses. However, this

issue was handled well in our current study as frequent reminders were given throughout the

administration of the test instruments. Some other researchers also asked test participants to

transfer all their responses to a separate page, such that any missing responses could be seen

and addressed easily (Seoh, 2017).

In summary, it is suggested that the 5-tier test format is a useful extension of the

multi-tier test format used in the literature. The fifth tier is indispensable for the students to

articulate their mathematical thinking with respect to their conceptual understanding in the

answer and reason tiers. It is better for this format to be open-ended so that the expression of

mathematical thinking is a cognitive act. Of course, one disadvantage of this fifth tier is the

analyses of the responses using a suitable marking scheme. In this study, the marking scheme

is deliberately kept simple, as indicated in the methodology section.

7.2. Use of Mathematical Thinking in Physics

This section will elaborate on our study results in relation to our second research

question.

327
 Research Question 2: “To what extent do high-performing students demonstrate

mathematical thinking when solving higher order thinking questions in selected

topics in physics?”

The constructed higher order thinking problems were expounded into two test-

instruments using 5-tier format: Chain-Collision Assessment and Rigid Body Collision

Assessment. These two instruments were used in our main studies to ascertain the extent to

which high-performing students demonstrated mathematical thinking when solving physics

questions. From the results of the main studies, the fifth tier (the mathematical expression

tier) allowed us to observe that mathematical thinking did not come easily, even for high-

performing students, with average score for this tier being 36.5% for the first study and 32.5%

for the second study. This revealed to us that although these high-performing students were

usually able to perform very well on basic conceptual questions, it did not really translate

into their ability to perform mathematical thinking. In other words, basic conceptual

understanding in physics has limited bearing on the students’ ability to perform mathematical

thinking. This may imply that the students’ high performance in examinations, assessments,

or even conceptual surveys (as reported in the preliminary study) could possibly be due to

other factors but not really mathematical thinking ability. Also, guessing, memorising,

elimination of unlikely options in the distracters and qualitative analysis could have

contributed in some way to their performance. Students’ scores in a physics assessment could

be seen to decline significantly when the students’ mathematical thinking was evaluated as

compared to their scores when only their answers were evaluated. While some students who

328
failed to obtain correct answer and/or correct reason provided irrelevant mathematical

explanations, many of the students still failed to provide full mathematical explanations after

obtaining correct answer and/or correct reason for a question. For the latter, the problem

might not lie in their mathematical thinking ability, but in the articulation of mathematical

thinking. In contrast, it was also found in this study that even when students could perform

mathematical thinking in the right direction, many of them were unable to articulate it back

into qualitative expressions. The results of both main studies were similar and could be

potentially generalised to other topics in physics, based on the teaching experience of the

author.

The fifth tier was marked carefully using a fine-grained marking scheme, where

student could be awarded 0.0, 0.5, 1.0 or 1.5 marks for their mathematical explanations in

each question (refer to Tables 6.6 and 6.13). Other tiers (that is, answer and reason), were

marked as is typically done for a 4-tier test. The results show that there is a significant positive

correlation (with moderate strength) between students’ ability to perform mathematical

analyses and their performance in the answer and reason tiers (𝑟 > .4, 𝑝 < .001 for both

studies). The basis for this correlation is possibly due to direct causation between students’

mathematical thinking and their performance in assessment involving higher order thinking

questions. Moreover, the association between students’ mathematical thinking and

conceptual understanding was also manifested in the interviews. From an educational

perspective, this provided motivation to explore teaching interventions that promote and

enhance students’ mathematical thinking in physics conceptual learning in order to advance

their performance in this kind of assessments.

329
 The correlation between mathematical thinking ability and degree of physics

conceptual understanding cannot be underestimated. There has been a growing interest to

understand the interrelationships between physics and mathematics education, although these

relationships have been the subject of study by philosophers, mathematicians and physicists

since antiquity (classical era) (Uhden et al., 2012). The two subjects have been going hand-

in-hand intimately (Bailly & Longo, 2011) in a way that mathematics carries its role as an

essential tool for physics (Wagh & Deshpande, 2012) while physics has been described as a

rich source of inspiration and insight in mathematics (Atiyah, 1990). Galileo Galilei (1564-

1642), being an individual whose knowledge spans a number of subjects, expressed that “The

book of nature is written in the language of mathematics” (as cited in Forinash et al., 2000).

This provides us with a rich insight that physics, as a study of nature, is inseparable from

mathematics as its language.

Remarks such as “A physicist is also a mathematician, and vice versa” might not be

peculiar in the past. Many historical giants like Isaac Newton, Johann Carl Friedrich Gauss,

Albert Einstein and others are hard to classify as pure mathematicians or pure physicists.

However, nowadays such remarks may cause some dispute even among mathematicians and

physicists themselves. Not only mathematicians and physicists differ in their scope of study,

but also in the thinking approach taken. The scope of physics is reality, the universe, and

natural phenomena, whereas the scope of mathematics is everything one can imagine or could

be imagined, whether or not it has any connection to reality. The real difference is that physics

is a science, and mathematics is not. Physics therefore works on the principle of formulating

a hypothesis, testing it, and discarding it if disproven. Mathematics works on the principle of

playing around, sometimes with an aim (perhaps to prove something one suspects) or

330
sometimes without. These differences affect one’s thinking pattern as he/she learns the

subject.

Based on interactions with other physics teachers, many of them subscribe to the

school of thought that “A physics concept should be understood qualitatively first, before the

mathematical representation of it comes into place”. However, our results showed that the

use of mathematical thinking may be able to strengthen one’s conceptual understanding. In

contrast to traditional physics class, we strongly believe that mathematics is not just a

necessary skill, but it should be seen as a way of representation of physics concepts involving

the thought process and analysis, and that it is a crucial part in the study of physics. In

rotational mechanics, for instance, it may not be obvious whether it is easier to rotate a longer

rod or a shorter rod given the amount of applied force and the mass of the rod being fixed.

Representing this problem mathematically allows us to see that as the length of the rod

doubles, the amount of torque doubles too, yet the moment of inertia quadruples, resulting in

a lower angular acceleration. Hence, it is easier to rotate a shorter rod in this case.

Previous studies have explored the use of mathematical thinking in the subject of

Mathematics itself (Dubinsky, 2002; Dreyfus, 2002; Jacobs et al., 2010; Tall, 2002) as well

as in relation to creativity testing (Ervynck, 2002; Haylock, 1997; Leikin, 2007), computer

science (Wing, 2008; Bar & Stephenson, 2011; Henderson et al., 2001; Gries et al., 2001),

psychology (Resnick & Ford, 2012; Hannula, 2006), neuroscience (Dehaene et al., 1999;

Sohn et al., 2004; Goswami, 2004) and others. Champagne et al. (1980) identified that

mathematical skills is one of the main variables contributing to student’s success in learning

physics. Cohen et al. (1978) found, that SAT mathematics scores correlate highly both with

the level of the physics course and with final grades in the course. Though final course grades

331
may not be the best measure of actual achievement in physics, it gives us an underlying idea

about how important mathematics is for physics students to achieve bigger success. Similar

results have also been reported by others (Blumenthal, 1961; Cohen et al., 1978), i.e. there is

a positive correlation between physics course grades and mathematics scores of college

entrance exams. Interestingly, another study by Meltzer (2002) reported that students’

normalised learning gains on a physics concept test are not correlated to their pre-test scores,

but they are significantly correlated with their pre-instruction mathematics skill. This is

possibly due to strong dependence of higher level physics on mathematics.

There is indeed a branch of Mathematics, called Mathematical Physics, which deals

with the development of mathematical methods suitable for applications and for the

formulation of physical theories. But above and beyond applying mathematical methods, it

has to be noted that mathematical thinking is a cognitive skill and habit that needs to be

implemented and developed since the early years. However, very limited resources were

found on the use of mathematical thinking in exploring high school physics problems in the

literature. This formed the cornerstone of the first part of this thesis study, where much

thought was put on two topics in mechanics: momentum/collision and rotational mechanics.

Theoretical explorations making use of mathematical thinking were done and elaborated in

Chapters 2 and 3. Mathematics was extensively used to create higher order thinking questions

in the two topics, as several layers of thinking need to be performed to approach each

question. Although the topics are not new or difficult – they are based on high school physics

syllabus –, the novelty in these explorations lies in the higher order thinking problems

generated. Without mathematical thinking, both physics students and teachers with strong

basic conceptual understanding on the topics may perform inaccurate analyses in attempting

332
to answer the questions. This fact was reinforced in our results as well when the extract of

students’ reflections showed the prevalence of alternative conceptions; this was due to

overreliance on scientific intuition, thus concurring with the study by Lemos (2008).

Students’ alternative conceptions could also be analysed from their responses in the answer

and reason tiers using the classification scheme done by Caleon and Subramaniam (2010a).

Although the student participants had learnt the topics tested prior to participating in this

study, it could be seen that they still showed conceptual mistakes in higher order thinking

questions. Moreover, the analysis of the fifth tier as well as interviews strongly suggest that

these alternative conceptions were related to their mathematical thinking. In other words,

students who had not developed the habit of performing mathematical thinking had a

tendency to make mistakes in higher order thinking questions. These alternative conceptions

were not manifested in other conceptual surveys used as the test instruments only tested basic

conceptual understanding with very little mathematical thinking involved.

Unfortunately, the habit of performing mathematical thinking was not fully

encouraged in many high school physics classes. This was also observed in the way our

student participants responded to physics questions, where other approaches would be

resourced before mathematical thinking was performed. This lack of mathematical thinking

in the learning of physics is mainly due to the fear of both teachers and students. Oftentimes,

there is fear that the use of mathematics would demotivate students from learning physics as

they would require students to have a certain level of mathematical skills or else they would

have to devote more effort to harness their mathematical skills. In this respect, mathematical

thinking as an important aspect in physics is viewed as a stumbling block by many students

(Klüver & Klüver, 2019). Interestingly, while this may be true among average-performing

333
groups of students, the trend was not seen for high-performing students in the results of our

study.

7.3. Use of Mathematical Thinking in Physics Conceptual Learning in Improving

Students’ Performance in Higher Order Thinking Questions

This section will elaborate on our study results in relation to our third research

question.

Research Question 3: “How do high-performing students respond to teaching

interventions that seek to promote mathematical thinking on higher order thinking

questions in the selected topics in physics?”

Most successful teachers subscribe to the mindset that it is far more crucial to teach

people how to think, and not just what to think (Shure, 2001). The joy and feeling of

accomplishment is often achieved when teaching becomes more than just transferring

knowledge or even spoon-feeding students but stretching their capacity to develop good

understanding and utilise existing concepts to expand and create new knowledge. The

teaching interventions explored in this thesis study can help high-performing students acquire

better conceptual understanding as well as develop higher order thinking skills.

Our approach to introduce mathematical thinking into the learning process is by

exploring a new teaching strategy with the framework elaborated in Chapter 4. This strategy

involves encouraging students to mathematise physical situations and perform mathematical

thinking during the process of problem-solving. Mathematical notations, variables, formulas

334
as well as mathematical tools (e.g. algebra, differential equation, trigonometry) are highly

involved, and the mathematical thinking process is expected to help students leverage their

performance in an assessment, especially one that involves higher order thinking.

As mentioned in Chapter 1, prior to this thesis study, I had adopted this intervention

strategy for small groups of students undergoing Physics Olympiad training, both in school

and at national level. This teaching approach was used even for teaching simple physics

concepts and it had resulted in significant improvements in students’ performance when

higher order thinking questions were involved, although the results were not robust since the

number of students were relatively small as compared to cohort sizes in schools. However,

the results of this thesis study have shown that the same intervention strategy could be applied

to a wider pool of high-performing students, and not exclusively for Physics Olympiad

students.

Good questioning techniques have long been regarded as a fundamental tool of

effective teachers (Way, 2008). Unfortunately, research shows that 93% of teacher questions

are “lower order” knowledge-based questions focusing on recall of facts (Daines, 1986). The

right type of questioning, i.e. “mathematical questioning”, is definitely required to stimulate

the mathematical thinking that can arise from engagement in physics problems and

investigations. In teaching the concept of momentum to solve a collision problem, for

instance, an example of common teacher questioning would be “Why do you think you can

or cannot apply the law of conservation of momentum?” and the expected answer would be

“We can apply it because the system is isolated.” or “We can apply it because there is no net

external force acting on the system.” In this strategy, teachers could instead use mathematical

335
questioning, such as “Since the relation between the net external force, ∑ 𝐅⃗ext , and the

𝑑𝐩
⃗⃗
⃗⃗, is ∑ 𝐅⃗ext =
momentum of the system, 𝐩 ⃗⃗ is constant here?” and the
, do you think 𝐩
𝑑𝑡

⃗⃗.” In essence, either teacher

⃗⃗ is constant because ∑ 𝐅⃗ext = 𝟎
expected answer would be “Yes, 𝐩

questioning above does not have superiority of one over another in teaching this simple

concept, as they both provoke the same level of conceptual depth. However, it is expected

that students who are more mathematically inclined would benefit from the latter approach

as their mathematical thinking is provoked, thus confirming the results of a previous study

(Meltzer, 2002).

In our main study, the students in the experimental group were asked to attend a

lesson intervention that was framed as a revision lesson on the selected topic. The lesson was

unique in that mathematical thinking was used extensively, emphasised and promoted

throughout the lesson. The lesson does not teach to the test as the contexts used were

different. From students’ reflections and survey results, it could be concluded that

mathematics was one important factor in their conceptual understanding in physics. More

than 75% of the participants agreed that mathematics was a crucial element in the study of

physics. More than 88% of the participants also felt benefited by the new teaching approach

as the habit of performing mathematical thinking enabled them to perform logical thinking

and deeper analysis for a given physical situation. This fact was also confirmed by the

significant difference in the post-test results between the control group and the experimental

group (𝑝 < .001 for both studies). Though the teaching intervention cannot strictly be called

as ‘conceptual change intervention’, the process of mathematical thinking also led to

reduction in the number of alternative conceptions for the experimental group. Although

336
there were still some persistent alternative conceptions, the percentage of students with these

alternative conceptions generally decreased. This indicates there were conceptual changes

taking place when they carefully approached a given physical intuition using mathematical

thinking and mathematical representations. Although this might go against their own

intuition sometimes, it provided them with stronger basis for their answers. Consequently,

the students’ performance in the post-test in both the Chain-Collision Assessment and the

Rigid Body Collision Assessment improved significantly even for the answer and reason

tiers, as compared to their performance in the pre-test. This emphasises the significance of

the teaching intervention that took place. Unfortunately, very often physics teachers tend to

delink the phrase “physics concept” from mathematics. Many of them feel that “teaching

concepts is more important than the mathematics”. They may be afraid of bringing up the

mathematical thinking element when analysing a physical situation as it may discourage

students to learn more. This tendency affects students’ attitudes towards physics. As such,

students are not used to representing ideas, thoughts, reasoning, or thinking mathematically.

Previous studies have focused on teaching approaches that are suitable for average-

performing students, special needs students and high-performing students (e.g. Wells et al.,

1995; Brewe et al., 2009; Colvin et al., 2014; Asiksoy & Ozdamli, 2016). A good teacher

would be able to use differentiated instructions when dealing with students with various

levels of ability (Tomlinson, 2001). When dealing with high-performing students, one cannot

assume that the introduction of mathematics in physics would push students away from their

interest in pursuing physics or engineering in higher education. For example, the subject

participants in our study found the lesson intervention to be interesting and enjoyable, as

reflected in the survey and reflections. This concurs with previous studies that found passion

337
for mathematics to be a driving factor for high-performing students to study higher physics

(McDonnell, 2005; Ornek et al., 2008; Holmegaard et al., 2010, Oon & Subramaniam, 2011).

A study by de Jong and Fergusion-Hessler (1991) shows that ‘experts’ demonstrate

increased ability to translate between representations when asked to reproduce problems.

Being able to translate between representations, they are able to use a variety of tools at their

disposal to attempt to solve the problem. Indeed, there is a threshold level of ability in the

combination of representations necessary for solving a given disciplinary problem (Airey &

Linder, 2009). In the study, the students commented that particular representations made

solving the problem easier even though they wouldn’t have used that representation if they

had the choice (de Jong & Fergusion-Hessler, 1991). This concurs with the reflections from

our student participants who found that mathematical representations really help in analysing

physical situations.

The findings in this study may be a potential remedy for the declining interest among

physics students that has been observed in many countries (Woolnough, 1994; Politis et al.,

2007; Oon & Subramaniam, 2010; Oon & Subramaniam, 2011). This is a cause for concern.

It is very much related to the declining performance in assessment, as good grades motivate

the students (Oon & Subramaniam, 2011; Barnes, McInerney, & Marsh, 2005; Lyons, 2006;

Smithers & Robinson, 2008; Woolnough, 1993) and, likewise, lower grades demotivate

students. The same study by Oon & Subramaniam (2011) also reveals that most physics

teachers in Singapore schools agree that mathematics is required for students to do well in

physics. This is a double-edged sword for while mathematics is crucial for physics, it also

puts students off (Oon & Subramaniam, 2011; Politis et al., 2007; Khalijah et al., 1995). Most

students view heavy math as the main reason to drop physics (Williams et al., 2003;

338
Woolnough, 1994). As mentioned before, mathematical thinking was seen as a stumbling

block for many students (Klüver & Klüver, 2019). This fear has been developed not only

among physics students but also among physics teachers. Physics teachers tend to avoid

mathematical thinking in their teaching instruction, in the belief that students will continue

to take science in their further study. However, this creates avoidance among physics teachers

towards using mathematical analysis which could potentially create a negative impact

towards students’ learning (Politis et al., 2007). Schools with high-performing students are

of no exception. The same trend is observed among students who are interested in physics

and are still majoring in physics in the upper years. This happens especially as the assessment

consists of increasingly challenging problems. Interestingly, this same group of high-

performing students actually have no problem with mathematics and are able to apply a wide

range of mathematical tools (e.g. algebra, series, vectors, functions, calculus, etc.). However,

teachers’ tendency to avoid mathematical thinking for the introductory level could have

disillusioned these students in that there should not be so much overlapping between the two

subjects, i.e. Mathematics and Physics. This, in turn, could have caused a lack in ability to

express their conceptual understanding in a topic context in another representation – that is,

mathematical language. In other words, students’ lack of representational fluency using

mathematical language may be due to unfamiliarity in performing the thinking and not the

lack of using the skills. The fear of using mathematical thinking builds a false belief among

high performers that physics is just a set of key words and formulas to be memorised, and

this belief should be dispelled. Ironically, the same study by Politis et al. (2007) stressed that

teachers’ views on subjects are important in developing students’ subject interest to study

because teachers are adjudged to have a wealth of experience in relation to the factors

339
influencing subject choice, as well as considerable insights into the mindsets of the students.

On the flip side, high-performing students, in general, view physics as a boring subject as

they simply have to memorise concepts, facts and formulas (Reid & Skryabina, 2002, 2003).

The students’ reflections and surveys showed a very positive impression about using

mathematical thinking in physics conceptual learning. Important key phrases or sentences

were captured, such as “magnificence of a very simple mathematical concept”, “reduce

possible errors due to false intuitions”, “more effective and accurate when it comes to solving

the especially tough and complicated physical systems”, and so on. In summary, the

mathematical thinking intervention approach in this study is a good strategy to guide the

students to think in a more logical way and even develop better scientific intuition to foresee

the state of a physical system. It also led to conceptual changes for these students. It indeed

showed that the assumption that using mathematical thinking in physics would put students

off is a misbelief.

The results of this thesis study is indeed promising and ground-breaking. It is shown

that promoting mathematical thinking in physics conceptual learning helps one improve in

physics assessment as well as their confidence level. Higher assessment scores would imply

higher confidence in the subject, and higher confidence would eventually lead to higher

interest in learning deeply. In other words, the improvement in students’ performance can

boost their motivation and confidence in pursuing physics at higher education.

Interestingly, previous studies have reported a positive correlation between physics

course grades and mathematics scores of college entrance exams (Blumenthal, 1961; Cohen

et al., 1978). Another study by Meltzer (2002) reported very interesting findings: students’

normalised learning gains on a physics concept test are not correlated to their pre-test scores,

340
but they are significantly correlated with the students’ pre-instruction mathematics skill. This

suggests to us that the two subjects are not supposed to be standalone subjects, but that they

should support each other instead. In learning physics concepts, for instance, mathematical

thinking plays a significant role in giving a more thorough conceptual understanding.

Moreover, Sadler and Tai (2001) have found that students who had high school physics

courses that spent more time on fewer topics and focused more on deeper mathematical

techniques and methods in the problem-solving process, performed much better in college

physics than those who raced through more content. This finding is interesting and should be

seriously taken into consideration by curriculum planners. It suggests the significance of

students’ mathematical ability in the learning of physics. However, prior to this thesis study,

not much was known on how students’ physics conceptual understanding was associated

with their mathematical thinking and how a given physics problem could be approached by

performing mathematical thinking. This formed the cornerstone of our study, as presented in

this thesis. Overall, together with this thesis study, the findings of previous studies confirm

the importance of early exposure to mathematical thinking in physics.

341
 CHAPTER 8

CONCLUSION

The ability to use mathematical thinking in other areas is an important goal of

schooling in general (Stacey, 2006), as it will support a number of other subjects taught in

school as well as other aspects of life such as science, technology, economy, etc. In this

school of thought, mathematical thinking is crucial in large measure because it equips

students with the ability to use mathematics in several different areas of life, thus forming an

important outcome of schooling. Hence, it is important for students to have a taste of the

intellectual adventure that mathematical thinking can promote, including its use in physics.

Our current study explored the use of mathematical thinking in physics. This study

was unique in the way that it combined theoretical explorations as well as educational studies,

both in relation to mathematical thinking. For the theoretical explorations, it was shown that

mathematical thinking could be used to perform deeper analysis of a simple physics concept

for the purpose of creating higher order thinking questions. For our educational studies, firstly

we developed 5-tier instruments to excerpt students’ mathematical thinking in physics. The

instruments allowed us to identify alternative conceptions, the extent of students’

mathematical thinking and the correlation between mathematical thinking ability and

conceptual understanding in physics. We found that mathematical thinking does not come

easily even for high-performing students, that is, the depth of students’ conceptual

understanding might not translate to ability to perform and articulate their mathematical

thinking. It was also found that a teaching intervention that promotes mathematical thinking

was effective for high-performing students. Mathematical thinking was beneficial for them

342
to perform logical thinking and deep analyses of physics problems.

8.1. Study Limitations

Having explored the usefulness of mathematical thinking in the study of physics, this

thesis study is not without limitations. Firstly, the theoretical explorations and constructions

of higher order thinking questions were only done for a few topics in mechanics. There are

other topics in physics that would need to be explored as the results of this study are not

supposed to be topic-specific. However, even applying mathematical thinking into a single

physics topic would require a long time. If the same educational study is to be repeated for

other topics, a new test instrument would need to be constructed as there are no suitable

instruments in the literature. The construction of the instrument, administration on the

students, interviews, teaching interventions and data analyses are time-consuming. As such,

this task is left for future studies to investigate the relevance of incorporating mathematical

thinking to other topics in physics.

Secondly, even within a single topic of mechanics, only a very limited aspect of the

topic could be explored. In the topic of momentum/collision, for instance, the theoretical

explorations focused on one-dimensional chain-collision, as we would like to analyse the

momentum transfer as well as energy transfer from one block to another during a collision.

There are other cases that can be explored, such as impulse-momentum theorem, explosion

case, maximising energy in an explosion, two- and three-dimensional collisions, bouncing

balls, and others. In our second topic of exploration, i.e. rotational mechanics, we focused

only on the concept of Newton’s laws of translation and rotation as well as rigid body

collision, as we would like to test on the concept of angular momentum and rotational kinetic

343
energy. There are other subtopics that can be explored, such as rolling without slipping,

angular impulse-angular momentum theorem, pulley-system, and others. Unlike other

conceptual surveys that can test a variety of aspects on a single topic, owing to the relatively

large number of questions, our constructed instruments could only test limited aspect due to

the small number of questions involved.

Thirdly, as the participants in our educational study were expected to perform

mathematical thinking in physics, the assessments were only administered to high-

performing students as we did not want the lack of basic physics concepts to be the stumbling

block to what we wanted to explore. Our educational study was thus confined to only one

school in Singapore, i.e. NUS High School of Mathematics and Science. The participants in

the school performed very well in several conceptual surveys in physics, thus putting them

in the category of high-performing students. The school also puts much emphasis on

mathematics and sciences in its curriculum. Moreover, the school has a relatively small

population, i.e. around 170 students per batch. This also puts a limitation on the sample size

for both our main studies. We should be careful in generalising the results obtained to other

groups of students or students from other schools in Singapore. These students might not

have the same exposure to mathematics and physics as well as sufficient competencies in the

two subjects.

Fourthly, as each instrument uses 5-tier format, the administration of each test

instrument took quite long and fatigue might take place. This was because there were five

responses required from each student participant for each question. In this aspect, an

assumption that students were attentive throughout the test was necessary. It was also

assumed that all student participants gave a frank indication of the confidence ratings for

344
each question, thus reflecting their true metacognitive skills. To overcome this limitation, it

is desirable that future studies be conducted over a period of time.

Lastly, on the teaching interventions of this thesis study, it was not ideal that the

researcher was also the teacher who conducted the intervention. This can potentially induce

an element of experimenter bias. As the research study dealt with high-performing students

and higher order thinking problems in physics, it was indeed imperative that the research,

including the teaching intervention, be conducted by someone with adequate expertise.

Social studies researchers would face difficulty if they were to conduct this thesis study.

Nevertheless, this bias was minimised in a number of ways – for example, triangulating

findings from the study using a number of approaches such as interviews, having a 5-tier

instrument to robustly probe for students’ understanding, checking for students’ cognitive

progression not only on the basis of cognitive scores but also metacognitive measures as well

as change in percentage of students harbouring alternative conceptions from pre-test to post-

test and ensuring that in the intervention the teacher did not teach to the test but used different

examples and contexts than those in the test instrument.

Further studies should also investigate the effect on students’ performance in

assessment when the new teaching intervention is used in teaching physics continuously from

basic to advanced concepts. Moreover, the effect of prolonged intervention is yet to be

explored as interventions in this study were only completed in one session. We expect to see

better improvements in students’ performance when students are exposed to such physics

lessons continually as they would have more time to develop the habit of applying

mathematical thinking in their approach.

345
8.2. Future Works

For future works, we would like to see further theoretical aspects as well as

educational studies that could be performed in relation to this thesis study. For the theoretical

exploration, another aspect of mathematical thinking will be explored in the topic of moment

of inertia, which is construction of new methods. This interesting exploration could be

extended to frame stimulating physics questions in the future.

8.2.1. Theoretical Exploration on Moment of Inertia

In the previous chapters, we have seen how mathematical thinking was deeply applied

for conceptual learning and problem-solving. The study could be extended to another aspect

of mathematical thinking, which is construction of new methods for determining a physical

property called moment of inertia. In this section, we explore the use mathematical thinking

in exploring new methods to compute the moment of inertia of rigid body. Two new elegant

methods, namely scaling method (Ricardo, 2015) and squashing method (Wang & Ricardo,

2019), to compute moment of inertia of regular and symmetrical objects will be elaborated.

These new methods can also be used in physics class as a way to promote mathematical

thinking.

8.2.1.1.Introduction

The moment of inertia of a rigid body is an essential physical property in the study of

rotational dynamics. The concept of moment of inertia is analogous to mass in the linear

motion. Although it can be derived theoretically from the mass distribution and geometrical

parameters of an object, the cookie-cutter approach involves substantial integration and may

346
be intimidating to high school students or even undergraduates without prior backgrounds in

calculus. Many schemes (e.g. integration, differentiation (Saravanan & Ricardo, 2020;

Rizcallah, 2015; Andersen, 2007) and clever manipulations of symmetry (Hong & Hong,

2013; Streib 1989)) have been devised to calculate the moments of inertia of simple objects.

And now in this chapter, alternative ways bypassing calculus to compute moments of inertia

of rigid bodies of regular shape using their symmetrical property will be elaborated. These

approaches will be very useful and preferred for teaching rotational dynamics at the high

school as well as undergraduate level, as it only deals with minimal integration – augmenting

an introductory physics student’s repertoire.

A classic high school or undergraduate demonstration (IOP Physics World, 2014) is

a race between diverse objects (e.g. cylinders, hollow cylinders, spheres and hollow spheres

of varying dimensions and densities), released simultaneously from a given starting point,

which then roll down an inclined plane. This experiment is often used to illustrate the

disparate “rotational inertia” of objects, alluding to the notion of the moment of inertia.

However, one particularly surprising result is that the centres of uniform cylinders (their

centres travel transversely to their cylindrical axes, along the slope) – with identical radii and

masses but different longitudinal lengths – cross the finishing line at the same time! Based

on this scenario, several guiding questions can be mathematically constructed:

(1) How will the moment of inertia of a solid cylinder scale if the cylinder is squashed

into a disk of the same mass?

(2) Will your answer change if the length of the cylinder is doubled?

347
(3) Will your answer change if the radial mass distribution of both the cylinder and disk

is not uniform?

The common explanation of the above observation is that the moment of inertia of a

uniform cylinder (with a fixed radius and mass) about its cylindrical axis is independent of

its length. When further inquired about the reason behind this unusual property, one often

turns to the fact that the moment of inertia is only dependent on the perpendicular distances

of mass elements to an axis – rendering the axial coordinates of mass elements and hence the

extent of an object along the axial direction irrelevant.

That said, we can view this situation from another perspective. By considering the

limiting case where the length of the uniform cylinder tends to zero, the afore empirical result

implies that the moment of inertia of a uniform cylinder about its cylindrical axis is equivalent

to that of a uniform circular disk (with an identical radius and a surface mass density equal

to the volume mass density of the cylinder multiplied by its length) about an axis passing

perpendicularly through its centre. Most notably, the surface mass density of this disk is

independent of the cylinder that was “compressed”, as long as the cylinder’s radius and mass

are fixed, since the volume mass density of such a cylinder must scale inversely with its

length. Therefore, if we show that all such cylinders can be reduced to this common disk, we

would have proven the claim that this particular moment of inertia is independent of the

length of such cylinders. However the property highlighted in the last sentence of the

previous paragraph precisely suggests so! Since the extents of the cylinders along the axial

direction are irrelevant, the cylinders can be flattened into our desired disk – illustrating the

crux of the squashing method, one of the two methods that will be elaborated in this section.

348
 Now, consider a rigid rod of certain mass and length, for instance. And consider the

following guiding questions:

(1) How will the mass, length and moment of inertia of a uniform rod scale if we cut the

rod into half?

(2) How do we construct the moment of inertia of the original rod using the moment of

inertia of the two halves?

(3) Therefore, how can we derive the expression for the moment of inertia of a uniform

rod of mass m and length ℓ?

These questions are the ones that inspired the discovery of scaling method. Scaling

transforms the moment of inertia of an object into the moment of inertia of an equivalent

object with a reduced or enlarged size. Noticing that several of these equivalent objects can

compose the original object – or equivalently, this equivalent object can be composed by

several of the original objects – forms an equation that allows us to determine the moment of

inertia of the object. On the other hand, squashing transforms the moment of inertia of an

object into the moment of inertia of an equivalent object with a reduced dimension, which is

often simpler to determine. The combination of scaling or squashing and other methods (e.g.

the perpendicular axis theorem and the parallel axis theorem) can also be used to compute

complex moments of inertia with minimal integration.

349
8.2.1.2.Theory

Inertia is the resistance of any physical object to any change in its state of motion, i.e.

the magnitude and direction of its velocity. It is the tendency of objects to keep moving in a

straight line at constant velocity, which may be zero. In high school physics, some teachers

may define inertia as “the laziness of an object to change its velocity”. The principle of inertia

is one of the fundamental principles of classical physics that are used to describe

the motion of objects and how they are affected by applied forces. Inertia comes from the

Latin word, iners, meaning idle, sluggish. In linear motion, this inertia is expressed as mass,

which is a quantitative property of physical systems. Isaac Newton defined inertia as his first

law in his Philosophiæ Naturalis Principia Mathematica, which states: The vis insita, or

innate force of matter, is a power of resisting by which every body, as much as in it lies,

endeavours to preserve its present state, whether it be of rest or of moving uniformly forward

in a straight line (Newton, 1846).

Another form of inertia is rotational inertia or moment of inertia, which refers to the

“laziness of an object to change its angular velocity”. This is a property that is constant for a

rigid body rotating about a specific axis, and hence it also has practical consequences. For

example, a rolling ball will resist any change in its angular velocity unless there exists kinetic

friction with the ground acting on it when sliding occurs. For non-rigid body, the moment of

inertia is not a suitable property to describe the motion of the object as it may not remain

structurally intact.

The concept of moment of inertia was first introduced in 1673 by Christiaan Huygens

in his study on physical pendulum or compound pendulum (Mach, 1960). The term “moment

of inertia” was later introduced by Leonhard Euler in his book Theoria motus corporum

350
solidorum seu rigidorum in 1765 (Mach, 1960; Euler & Carsten, 1765) and it is incorporated

into Euler’s second law, which is an extension of Newton’s second law when applied in rigid

body instead of point mass.

The computation of moment of inertia is usually done in undergraduate physics

(Young et al., 2012), though the concept and the idea may have been introduced qualitatively

since high school. When it comes to quantitative problem solving, a table of moment of

inertia is usually provided. With this qualitative understanding, students are able to compare,

for instance, between the moment of inertia of a ring and a solid disk as they rotate about an

axis passing perpendicularly through their centre. However, the quantitative expressions of

the moment of inertia have not gotten enough appreciation from high school students as its

computation requires the knowledge and the application of calculus.

Figure 8.1

Point Mass Undergoing Circular Motion about Fixed Axis

⃗⃗ = 𝑚𝐯⃗⃗
𝐩

𝐫⃗

Consider a point mass 𝑚 rotating about a certain axis (Figure 8.1). Its instantaneous

position vector (measured from the rotational axis) and instantaneous linear momentum are

given by 𝐫⃗ and 𝐩
⃗⃗, respectively. The angular momentum of this point mass is defined as

351
 ⃗ ≔ 𝐫⃗ × 𝐩
𝐋 ⃗⃗ = 𝐫⃗ × (𝑚𝐯⃗⃗) = 𝑚𝐫⃗ × (𝛚
⃗⃗⃗⃗ × 𝐫⃗) = 𝑚(𝐫⃗ ∙ 𝐫⃗)𝛚
⃗⃗⃗⃗ − 𝑚(𝐫⃗ ∙ 𝛚
⃗⃗⃗⃗)𝐫⃗

In this discussion, we will consider only the case where the plane of motion is perpendicular

to the rotational axis, i.e. 𝐫⃗ ∙ 𝛚

⃗⃗⃗⃗ = 0, also known as fixed axis rotation. With this assumption,

moment of inertia can be taken as a scalar instead of a tensor (Tenenbaum, 2006; Kane &

Levinson, 1985), and the angular momentum will always be pointing to the direction of the

angular velocity

⃗ = 𝑚𝑟 2 𝛚
𝐋 ⃗⃗⃗⃗

⃗⃗ = 𝑚𝐯⃗⃗, where 𝐩
This expression is analogous to the expression for linear momentum, 𝐩 ⃗⃗

represents the linear momentum; 𝐯⃗⃗ represents the linear velocity; and 𝑚 represents the

⃗ represents the angular momentum; 𝛚

translational inertia (mass). Here, 𝐋 ⃗⃗⃗⃗ represents the

angular velocity; 𝐫⃗ represents the position vector of the point mass measured from the

rotational axis; and 𝑚𝑟 2 is defined as the rotational inertia (moment of inertia) 𝐼.

Now, consider a rigid body (Figure 8.2) rotating about the 𝑧-axis (perpendicular to

both 𝑥- and 𝑦-axes, passing through the origin). We assume that rigid bodies can be treated

as discrete collections of point masses. Since the angular velocity of every single constituent

point mass is identical, the angular momentum of this rigid body can be obtained by

⃗ = ∑(𝑚𝑖 𝑟𝑖 2 𝛚
𝐋 ⃗⃗⃗⃗) = (∑ 𝑚𝑖 𝑟𝑖 2 ) 𝛚
⃗⃗⃗⃗
𝑖 𝑖

and the term in the parentheses is defined as the moment of inertia of the rigid body 𝐼.

352
Figure 8.2

Rigid Body Rotating about 𝒛-Axis

𝑚𝑖

𝐫⃗𝑖
𝑥

At this point, one may have thought that it would be impossible to find the moment

of inertia of a rigid body without performing any integration. But the subsequent parts of this

section are going to elaborate on alternative methods to compute the moments of inertia of

regular and symmetric objects, utilising mathematical thinking. The concept of inertia tensor

will be avoided as it is not suitable for high school students.

(1) Parallel Axis Theorem

Consider a rigid body that is rotating about the 𝑧-axis (Figure 8.3), and let the moment

of inertia of the rigid body about another axis that is parallel to the 𝑧-axis but passing through

⃗⃗
its centre of mass be 𝐼CM . Let 𝐫⃗𝑖 be the position vector of a constituent point mass 𝑚𝑖 and 𝐑

be the position vector of the centre of mass of the rigid body.

353
Figure 8.3

Centre of Mass Position of Rigid Body

CM
𝐫⃗𝑖 ′
𝑚𝑖
⃗𝐑⃗

𝐫⃗𝑖
𝑥

With 𝑀 denoting the total mass of the rigid body, 𝑀 = ∑𝑖 𝑚𝑖 , we have

⃗⃗ = 𝐫⃗𝑖
𝐫⃗𝑖′ + 𝐑

∑𝑖 𝑚𝑖 𝐫⃗𝑖
⃗⃗ =
𝐑
𝑀

⃗⃗
∑ 𝑚𝑖 𝐫⃗𝑖 ′ = 𝟎
𝑖

The moment of inertia about the axis passing through the centre of mass is thus given by

𝐼𝐶𝑀 = ∑ 𝑚𝑖 𝑟𝑖 ′2
𝑖

and by combining the above equations, the moment of inertia about the 𝑧-axis is given by

⃗⃗⃗)
𝐼 = ∑ 𝑚𝑖 𝑟𝑖 2 = ∑ 𝑚𝑖 𝐫⃗𝑖 ∙ 𝐫⃗𝑖 = ∑ 𝑚𝑖 (𝑟𝑖 ′2 + 𝑅2 + 2𝐫⃗𝑖′ ∙ 𝐑
𝑖 𝑖 𝑖

⃗⃗⃗ ∙ ∑ 𝑚𝑖 𝐫⃗𝑖 ′
𝐼 = ∑ 𝑚𝑖 𝑟𝑖 ′2 + (∑ 𝑚𝑖 ) 𝑅2 + 2𝐑
𝑖 𝑖 𝑖

𝐼 = 𝐼CM + 𝑀𝑅2

(8.1)

354
Equation (8.1) is known as parallel axis theorem and is very useful in computing moments

of inertia of rigid bodies.

(2) Perpendicular Axis Theorem

𝑥𝑖
Consider a two-dimensional rigid body on the 𝑥-𝑦 plane (Figure 8.4). Let 𝐫⃗𝑖 = (𝑦 )
𝑖

be the position vector of a constituent point mass 𝑚𝑖 .

Figure 8.4

Two-Dimensional Rigid Body on 𝒙𝒚-Plane

𝑥𝑖 𝑚𝑖

𝑦𝑖
𝐫⃗𝑖
𝑥

The moment of inertia of the rigid body about the 𝑥-axis is given by

𝐼𝑥 = ∑ 𝑚𝑖 𝑦𝑖 2
𝑖

as 𝑦𝑖 is the distance between the point mass and the 𝑥-axis. Similarly, the moment of inertia

of the rigid body about the 𝑦-axis is given by

𝐼𝑦 = ∑ 𝑚𝑖 𝑥𝑖 2
𝑖

355
as 𝑥𝑖 is the distance between the point mass and the 𝑦-axis.

Using Pythagorean theorem (Sally, 2007), 𝑟𝑖 2 = 𝑥𝑖 2 + 𝑦𝑖 2 , the moment of inertia of

the rigid body about the 𝑧-axis can be expressed as

𝐼𝑧 = ∑ 𝑚𝑖 𝑟𝑖 2 = ∑ 𝑚𝑖 (𝑥𝑖 2 + 𝑦𝑖 2 ) = ∑ 𝑚𝑖 𝑥𝑖 2 + ∑ 𝑚𝑖 𝑦𝑖 2
𝑖 𝑖 𝑖 𝑖

𝐼𝑧 = 𝐼𝑥 + 𝐼𝑦

(8.2)

8.2.1.3.Scaling Method for Moment of Inertia Computation

Dimensional analysis is a method of analysing the relationships between different

physical quantities by identifying their fundamental dimensions (such as length, mass, time)

and units of measure (such as meters, kilograms, seconds) and tracking these dimensions as

calculations or comparisons are performed. It is usually used as a method for solving

problems in qualitative physics (Bhaskar & Nigam, 1990), when one does not care about the

constant of proportionality. In fact, the idea on which dimensional analysis is based on is

very simple: physical laws do not depend on arbitrarily chosen basic units of measurement.

An important conclusion can then be drawn from this simple idea: the functions that express

physical laws must possess a certain fundamental property, which in mathematics is called

generalised homogeneity or symmetry (Barenblatt, 1996).

Consider a symmetrical rigid body of regular shape. This object can have any number

of independent lengths depending on the dimensions of the shape. For example, a two-

dimensional object will have a maximum of two independent lengths: a square only has one

independent length (it can be its side, diagonal, etc.) as it has an additional constraint

356
compared to a rectangle. Using dimensional analysis, we know that moment of inertia is of

dimension mass × length2. If, for instance, a two-dimensional object of mass 𝑀 has

independent lengths given by 𝑎 and 𝑏, then its moment of inertia will be a polynomial

1⁄ 3⁄
consisting the following terms: 𝑀𝑎2 , 𝑀𝑏2 , 𝑀𝑎𝑏, 𝑀𝑎−1 𝑏3 , 𝑀𝑎 2 𝑏 2, … , with an

appropriate coefficient for each term. The list of terms is not exhaustive as the exponent

should cover the whole range of real numbers.

Hence, the moment of inertia of a one-dimensional object can be expressed as

𝐼 = 𝑘𝑀𝑑 2

where 𝑘 is a dimensionless constant to be determined; 𝑀 is the mass of the object; 𝑑 is the

dimension (length) of the object. And the moment of inertia of a two-dimensional object can

be expressed as

𝐼 = ∑ 𝑘𝛼 𝑀𝑑1 𝛼 𝑑2 2−𝛼 ∀𝛼 ; 𝛼∈ℝ

𝛼

where {𝑘𝛼 } are dimensionless constants to be determined; 𝑀 is the mass of the object; 𝑑1 and

𝑑2 are the two independent dimensions (lengths) of the object. Similarly, the moment of

inertia of a three-dimensional object can be expressed as

𝐼 = ∑ ∑ 𝑘𝛼,𝛽 𝑀𝑑1 𝛼 𝑑2 𝛽 𝑑3 2−𝛼−𝛽 ∀ (𝛼, 𝛽 ) ; (𝛼, 𝛽 ) ∈ ℝ2

𝛽 𝛼

where {𝑘𝛼,𝛽 } are dimensionless constants to be determined; 𝑀 is the mass of the object;

𝑑1 , 𝑑2 , 𝑑3 are the three independent dimensions (lengths) of the object.

If a one-dimensional object of uniform density is shrunk into a smaller shape

proportionally, or enlarged into a bigger shape proportionally (Figure 8.5(a)), then the

moment of inertia of the resulting object is given by

357
 𝐼 ′ = 𝑘(𝑛𝑀 )(𝑛𝑑 )2 ∀ 𝛼 ; 𝛼 ∈ ℝ

𝐼 ′ = 𝑛3 𝐼

(8.3)

Figure 8.5

(a) One-Dimensional and (b) Two-Dimensional Uniform Objects Scaled Proportionally by

Linear Factor 𝒏

𝑑 𝑛𝑑

(a)

𝑑 𝑛𝑑

(b)

If a two-dimensional object of uniform density is shrunk into a smaller shape proportionally,

or enlarged into a bigger shape proportionally (Figure 8.5(b)), then the moment of inertia of

the resulting object is given by

𝐼 ′ = ∑ 𝑘𝛼 (𝑛2 𝑀)(𝑛𝑑1 )𝛼 (𝑛𝑑2 )2−𝛼 ∀𝛼 ; 𝛼∈ℝ

𝛼

358
 𝐼′ = 𝑛4 ∑ 𝑘𝛼 𝑀𝑑1 𝛼 𝑑2 2−𝛼 ∀𝛼 ; 𝛼∈ℝ
𝛼

𝐼 ′ = 𝑛4 𝐼

(8.4)

where 𝑛 is the linear proportionality factor. This is true as {𝑘𝛼 } remain unchanged as the

resulting rigid body is similar to the original one. Similarly, if a three-dimensional object of

uniform density is shrunk into a smaller shape proportionally, or enlarged into a bigger shape

proportionally, then the moment of inertia of the resulting object is given by

𝐼′ = ∑ ∑ 𝑘𝛼,𝛽 (𝑛3 𝑀)(𝑛𝑑1 )𝛼 (𝑛𝑑2 )𝛽 (𝑛𝑑3 )2−𝛼−𝛽 ∀ (𝛼, 𝛽 ) ; (𝛼, 𝛽 ) ∈ ℝ2

𝛽 𝛼

𝐼′ = 𝑛5 ∑ ∑ 𝑘𝛼,𝛽 𝑀𝑑1 𝛼 𝑑2 𝛽 𝑑3 2−𝛼−𝛽 ∀ (𝛼, 𝛽 ) ; (𝛼, 𝛽 ) ∈ ℝ2

𝛽 𝛼

𝐼 ′ = 𝑛5 𝐼

(8.5)

This is true as {𝑘𝛼,𝛽 } remain unchanged as the resulting rigid body is similar to the original

one.

In this subsection, it will be demonstrated that dimensional analysis, combined with

the parallel axis theorem and perpendicular axis theorem, can be used to compute moments

of inertia of symmetric objects by inquiring on their symmetrical property. Although the

focus will be on one-dimensional and two-dimensional rigid bodies, the application of this

scaling method can always be extended to three-dimensional rigid bodies. For example, if a

1
one-dimensional rigid body is shrunk proportionally into half (𝑛 = ) of its original size, its
2

1 1
moment of inertia will be multiplied by a factor of = 8; and if a two-dimensional rigid
23

359
 1
body is shrunk proportionally into half (𝑛 = 2) of its original size, its moment of inertia will

1 1
be multiplied by a factor of 24 = 16.

When teaching the computation of moment of inertia in physics class, students can

be directed to think mathematically using the following guiding questions, in order for them

to grasp the method:

(1) How will the moment of inertia of a symmetrical rigid body scale if the body is
1
linearly shrunk by a factor of 2?

(2) How do we construct the original rigid body from identical smaller bodies having the

same shape as the original one?

(3) How is the moment of inertia of a rigid body related to the moment of inertia of its

constituent smaller bodies?

The following seven examples will explain how one can make use of these guiding

inquiry questions to determine the moment of inertia of rigid body of regular shape.

360
(1) Uniform Rod about Axis Passing Perpendicularly Through Centre of Mass

Figure 8.6

Uniform Rod about Axis Passing Perpendicularly through Centre of Mass

axis of rotation

Let a rigid rod of mass 𝑀 and length 𝐿 rotate about an axis passing perpendicularly

through its centre of mass (Figure 8.6). Its moment of inertia can be written as 𝐼 = 𝑘𝑀𝐿2 .
𝑀 𝐿
Now let the rod be split into two identical shorter rods with mass each and length 2 each
2

(Figure 8.7).

Figure 8.7

Shorter Rod about Axis Passing Perpendicularly through Centre of Mass

From equation (8.3), the moment of inertia of each shorter rod (shaded rod) about its
1
own axis passing perpendicularly through its centre of mass (solid arrow) is given by 8 𝑘𝑀𝐿2 .

Using parallel axis theorem, we can reconstruct the moment of inertia of the original rod

about the original axis (dashed arrow):

361
 1 𝑀 𝐿 2 1 1
𝐼 = 2 [ 𝑘𝑀𝐿2 + ( ) ] = 𝑘𝑀𝐿2 + 𝑀𝐿2 = 𝑘𝑀𝐿2
8 2 4 4 16

1
Solving this equation yields 𝑘 = 12, or

1
𝐼= 𝑀𝐿2
12

(8.6)

Alternatively, another rod with identical mass and length can be added to construct a

longer rod of mass 2𝑀 and length 2𝐿 (Figure 8.8).

Figure 8.8

Longer Rod about Axis Passing Perpendicularly through Centre of Mass

axis of rotation

From equation (8.3), the moment of inertia of this longer rod is 8 times the desired

result. Using parallel axis theorem,

2
𝐿 2
8𝑘𝑀𝐿 = 2 [𝑘𝑀𝐿2 +𝑀( ) ]
2

1 1
Solving this equation yields 𝑘 = 12, or 𝐼 = 12 𝑀𝐿2 , as obtained previously.

362
(2) Solid Rectangle about Axis Passing Perpendicularly Through Centre of Mass

Figure 8.9

Solid Rectangle about Axis Passing Perpendicularly through Centre of Mass

axis of rotation

Let a solid rectangle of mass 𝑀, length 𝑎 and width 𝑏, rotate about an axis passing

perpendicularly through its centre of mass (Figure 8.9). Its moment of inertia can be written

as 𝐼 = ∑𝛼 𝑘𝛼 𝑀𝑎𝛼 𝑏2−𝛼 . Repeating the same procedure, let the rectangle be split into four

𝑀 𝑎
identical smaller rectangles with mass each and side 2 each (Figure 8.10).
4

Figure 8.10

Smaller Rectangle about Axis Passing Perpendicularly through Centre of Mass

From equation (8.4), the moment of inertia of one small rectangle (shaded rectangle)
1
about a new axis passing perpendicularly through its centre of mass is given by 𝐼. Using
16

363
parallel axis theorem, we can reconstruct the moment of inertia of the original rectangle about

the original axis:

1 𝑀 𝑎2 𝑏2
𝐼 = 4 [ 𝐼 + ( + )]
16 4 16 16

1 1 1
𝐼 = 𝐼+ 𝑀𝑎2 + 𝑀𝑏2
4 16 16

1
𝐼= 𝑀 (𝑎 2 + 𝑏 2 )
12

(8.7)

Alternatively, a bigger rectangle of length 2𝑎 and width 2𝑏 could also be constructed

by adding three identical rectangles (Figure 8.11). From equation (8.4), the moment of inertia

of this bigger rectangle about a new axis parallel to its side passing through its centre of mass

is sixteen times the moment of inertia of the original rectangle (shaded rectangle). By parallel

axis theorem,

𝑎2 𝑏2
16𝐼 = 4 [𝐼 + 𝑀 ( + )]
4 4

12𝐼 = 𝑀(𝑎2 + 𝑏2 )
1
Solving this equation yields 𝐼 = 12 𝑀(𝑎2 + 𝑏2 ), as obtained previously.

364
Figure 8.11

Bigger Rectangle about Axis Passing Perpendicularly through Centre of Mass

This result can also be obtained clearly using perpendicular axis theorem. Since a
1
rectangle can be treated as a collection of infinite number of rods, 𝐼𝑥 = 12 𝑀𝑏2 and 𝐼𝑦 =

1 1 1
𝑀𝑎2 , thus 𝐼𝑧 = 𝐼𝑥 + 𝐼𝑦 = 12 𝑀(𝑎2 + 𝑏2 ). For a square, 𝑎 = 𝑏, 𝐼 = 6 𝑀𝑎2 .
12

365
(3) Solid Square about Axis Parallel to Plane Passing through Centre of Mass and

Tilted from Symmetry Axis

Figure 8.12

Solid Square about Axis Parallel to Plane Passing through Centre of Mass and Tilted from

Symmetry Axis

axis of rotation

Let a solid square of mass 𝑀 and side 𝑎 rotate about an axis parallel to its plane (𝑥𝑦-

plane) passing through its centre of mass and making an angle 𝜃 with the 𝑦-axis (Figure

8.12). Its moment of inertia can be written as 𝐼 = 𝑘𝑀𝑎2 , where we might expect 𝑘 to be a
𝑀
function of 𝜃. Now let the square be split into four identical smaller squares with mass 4

𝑎
each and side each (Figure 8.13).
2

From equation (8.4), the moment of inertia of one small square about a new axis
1
parallel to the original axis is given by 16 𝑘𝑀𝑎2 . Notice that the axes of two small squares

𝑎
need to be shifted by 𝑑1 = 4 (cos 𝜃 + sin 𝜃) and the axes of the other two small squares need

366
 𝑎
to be shifted by 𝑑2 = 4 (cos 𝜃 − sin 𝜃 ). Using parallel axis theorem, we can reconstruct the

moment of inertia of the original square about the original axis:

2 2
1 𝑀 𝑎 1 𝑀 𝑎
𝐼 = 2 [ 𝑘𝑀𝑎2 + ( (cos 𝜃 + sin 𝜃 )) ] + 2 [ 𝑘𝑀𝑎2 + ( (cos 𝜃 − sin 𝜃)) ]
16 4 4 16 4 4

1 1
𝐼= 𝑘𝑀𝑎2 + 𝑀𝑎2 = 𝑘𝑀𝑎2
4 16
1
Solving this equation yields 𝑘 = 12, or

1
𝐼= 𝑀𝑎2
12

(8.8)

which is surprisingly independent of 𝜃.

Figure 8.13

Smaller Square about Axis Parallel to Plane Passing through Centre of Mass and Tilted

from Symmetry Axis

𝑑1

𝑑2

367
 This result can also be obtained easily using perpendicular axis theorem, by aligning

the 𝑦-axis with the rotational axis. Since by symmetry (Figure 8.14), 𝐼𝑥 = 𝐼𝑦 , 𝐼𝑧 = 𝐼𝑥 + 𝐼𝑦 =

1
2𝐼𝑦 = 6 𝑀𝑎2 , hence

1
𝐼𝑦 = 𝑀𝑎2
12

Figure 8.14

Solid Square on 𝒙𝒚-Plane

368
(4) Solid Right-Angled Triangle about Axis Passing Perpendicularly through Centre of

Mass

Figure 8.15

Solid Right-Angled Triangle about Axis Passing Perpendicularly through Centre of Mass

2𝑏
3

𝑥
𝑏 axis of rotation
3

𝑎 2𝑎
3 3

Let a right-angled triangle of mass 𝑀, legs (catheti) 𝑎 and 𝑏, rotate about an axis

passing perpendicularly through its centre of mass (𝑧-axis) (Figure 8.15). By perpendicular

axis theorem, 𝐼𝑧 = 𝐼𝑥 + 𝐼𝑦 .

369
Figure 8.16

Solid Rectangle Constructed from Two Right-Angled Triangles

𝑦 𝑦′

𝑎 𝑎
3 2

𝑎 2 1
Using parallel axis theorem, 𝐼𝑦′ = 𝐼𝑦 + 𝑀 ( 6 ) = 𝐼𝑦 + 36 𝑀𝑎2 . Consider another

identical right-angled triangle (dashed triangle) (Figure 8.16). The two triangles form a
1 1
rectangle of mass 2𝑀, length 𝑎 and width 𝑏. Hence, 2𝐼𝑦′ = 12 (2𝑀 )𝑎2 , or 𝐼𝑦′ = 12 𝑀𝑎2 . This

1 1
yields 𝐼𝑦 = 18 𝑀𝑎2 . Symmetrically, 𝐼𝑥 = 18 𝑀𝑏2 . Therefore,

1
𝐼𝑧 = 𝐼𝑥 + 𝐼𝑦 = 𝑀 (𝑎 2 + 𝑏 2 )
18

(8.9)

370
(5) Equilateral Triangle about Axis Passing Perpendicularly through Centre of Mass

Let an equilateral triangle of mass 𝑀 and side 𝑎 rotate about an axis passing

perpendicularly through its centre of mass (Figure 8.17). Let the moment of inertia be

expressed as 𝐼 = 𝑘𝑀𝑎2 . Let the triangle be split into four identical equilateral triangles of
𝑀 𝑎
mass and side 2 (Figure 8.18).
4

Figure 8.17

Equilateral Triangle about Axis Passing Perpendicularly through Centre of Mass

𝑎
√3
3
𝑎

𝑎
√3
6

Figure 8.18

Smaller Equilateral Triangle about Axis Passing Perpendicularly through Centre of Mass

371
 From equation (8.5), the moment of inertia of a small triangle (shaded triangle) about
1
its own centre of mass is given by 16 𝑘𝑀𝑎2 . Only the axes of three triangles need to be shifted

𝑎
by 6 √3, since the middle triangle has the same axis as the original triangle. Reconstructing

the moment of inertia using parallel axis theorem,

1 𝑀 𝑎 2 1
𝐼 = 3( 𝑘𝑀𝑎2 + ( √3) ) + 𝑘𝑀𝑎2 = 𝑘𝑀𝑎2
16 4 6 16

1 1
𝑘𝑀𝑎2 + 𝑀𝑎2 = 𝑘𝑀𝑎2
4 16
1
Solving this equation yields 𝑘 = 12, or

1
𝐼= 𝑀𝑎2
12

(8.10)

(6) Hollow Sphere about Axis Passing through Centre of Mass

Consider a hollow sphere of mass 𝑀 and radius 𝑅 with its axis of rotation passing

through its centre of mass. The hollow sphere can be sliced horizontally into infinitesimally
𝑅
thin rings (let the total number of rings be 2𝑁 (𝑁 → ∞)) with equal vertical thickness 𝑁 each

(Figure 8.19). However, this vertical thickness is not the same as the thickness of the ring, as

the thickness of the ring is a small arc of the sphere. The further it is from the centre of the

sphere, the thicker the ring is, and its radius reduces proportionally. Hence, since the mass of

each ring depends linearly on its radius and its thickness, the mass of any ring is identical.
𝑀
Let the mass of the first ring at the centre of the sphere be 𝑚 = and its radius be 𝑅, the
2𝑁

𝑅
same as the radius of the sphere. Then any ring at position 𝑘𝑅 = 𝑛 𝑁 (𝑘 ∈ [−1,1] ; 𝑛 ∈

372
[−𝑁, 𝑁] , 𝑛 ∈ ℤ) from the first ring will have radius 𝑅√1 − 𝑘 2 and mass 𝑚, equal to the mass

of the first ring.

Figure 8.19

Slicing Hollow Sphere into Horizontal Infinitesimal Rings

If the moment of inertia of each ring about the 𝑥-axis is 𝐼𝑥 and about the 𝑧-axis is 𝐼𝑧 ,

then lim ∑ 𝐼𝑧 will give the moment of inertia of the hollow sphere about the 𝑧-axis, 𝐼. By
𝑁→∞

1
perpendicular axis theorem, 𝐼𝑥 = 2 𝐼𝑧 , and by parallel axis theorem one can shift the 𝑥-axis

to pass through the centre of the sphere, and the moment of inertia about this axis becomes

𝐼𝑥 + 𝑚(𝑘𝑅)2 . Since by symmetry, the moment of inertia of the hollow sphere about the 𝑥-

axis and the 𝑧-axis are identical, we have

∑[𝐼𝑥 + 𝑚(𝑘𝑅)2 ] = ∑ 𝐼𝑧

373
 1 2
𝑛 2
∑ 𝐼𝑧 + 𝑚𝑅 ∑ ( ) = ∑ 𝐼𝑧
2 𝑁
𝑁 𝑁
2𝑚𝑅2 2)
2𝑚𝑅2 2)
4𝑚𝑅2 𝑁(𝑁 + 1)(2𝑁 + 1)
∑ 𝐼𝑧 = ( ∑ 𝑛 = (2 ∑ 𝑛 = ( )
𝑁2 𝑁2 𝑁2 6
𝑛=−𝑁 𝑛=1

where we have counted the contribution from the top half and the bottom half of the sphere.
𝑀
Since 𝑚 = 2𝑁 , solving the above equation yields

2 3 1 2
𝐼 = lim 𝑀𝑅2 (1 + + 2
) = 𝑀𝑅2
𝑁→∞ 3 2𝑁 2𝑁 3

(8.11)

(7) Solid Sphere about an Axis Passing through Its Centre of Mass

Figure 8.20

Slicing Solid Sphere into Horizontal Infinitesimal Disks

374
 Consider a solid sphere of mass 𝑀 and radius 𝑅 with its axis of rotation passing

through its centre of mass. The solid sphere can be sliced into infinitesimally thin solid disks
𝑅
(let the total number of disks be 2𝑁 (𝑁 → ∞)) with equal thickness each (Figure 8.20).
𝑁

Hence the mass of each disk is proportional to the square of its radius. Let the mass of the

first disk at the centre of the sphere be 𝑚 and its radius be 𝑅, the same as the radius of the

𝑅
sphere. Any disk at position 𝑘𝑅 = 𝑛 𝑁 (𝑘 ∈ [−1,1] ; 𝑛 ∈ [−𝑁, 𝑁] , 𝑛 ∈ ℤ) from the first disk

will have radius 𝑅√1 − 𝑘 2 and mass 𝑚(1 − 𝑘 2 ).

If the moment of inertia of each disk about the 𝑥-axis is 𝐼𝑥 and about the 𝑧-axis is 𝐼𝑧 ,

then lim ∑ 𝐼𝑧 will give the moment of inertia of the solid sphere about the 𝑧-axis, 𝐼. By
𝑁→∞

1
perpendicular axis theorem, 𝐼𝑥 = 2 𝐼𝑧 , and by parallel axis theorem one can shift the 𝑥-axis

to pass through the centre of the sphere, and the moment of inertia about this axis becomes

𝐼𝑥 + 𝑚(1 − 𝑘 2 )(𝑘𝑅)2 . Since the moment of inertia of the solid sphere about the 𝑥-axis and

the 𝑧-axis are identical, we have

∑[𝐼𝑥 + 𝑚(1 − 𝑘 2 )(𝑘𝑅)2 ] = ∑ 𝐼𝑧

1 𝑛 2 𝑛 2
∑ 𝐼𝑧 + 𝑚𝑅2 ∑ ( ) (1 − ( ) ) = ∑ 𝐼𝑧
2 𝑁 𝑁

2𝑚𝑅2
∑ 𝐼𝑧 = ∑ 𝑛 2 (𝑁 2 − 𝑛 2 )
𝑁4

𝑛 2 𝑚
Since the mass of the solid sphere is 𝑀 = ∑ 𝑚(1 − 𝑘 2 ) = ∑ 𝑚 (1 − (𝑁 ) ) = 𝑁2 ∑(𝑁 2 −

𝑛2 ), it yields

2𝑀𝑅2 ∑𝑁 2 2 2
𝑛=−𝑁 𝑛 (𝑁 − 𝑛 ) 2𝑀𝑅2 2 ∑𝑁 2 2 2
𝑛=1 𝑛 (𝑁 − 𝑛 )
∑ 𝐼𝑧 = =
𝑁 2 ∑𝑁 2 2
𝑛=−𝑁(𝑁 − 𝑛 ) 𝑁 2 𝑁 2 + 2 ∑𝑁 2 2
𝑛=1(𝑁 − 𝑛 )

375
where we have counted the contribution from the top half and the bottom half of the sphere.

Solving the above equation yields

2 1 3 1 2 1 1 5 1 4 1 3 1
2𝑀𝑅2 2𝑁 (3 𝑁 + 2 𝑁 + 6 𝑁) − 2 (5 𝑁 + 2 𝑁 + 3 𝑁 − 30 𝑁)
∑ 𝐼𝑧 =
𝑁2 1 1 1
𝑁 2 + 2𝑁 3 − 2 (3 𝑁 3 + 2 𝑁 2 + 6 𝑁)

4 5 1 3 1
𝑁 − 3 𝑁 + 15 𝑁
∑ 𝐼𝑧 = 2𝑀𝑅 (15 2 )
4 5 1 3
3𝑁 − 3𝑁

5 1
2 1− +
𝐼 = lim 𝑀𝑅2 ( 4𝑁 2 4𝑁 4 ) = 2 𝑀𝑅2
𝑁→∞ 5 1 5
1−
4𝑁 2

(8.12)

8.2.1.4.Squashing Method for Moment of Inertia Computation

Now, we introduce the idea of squashing for moment of inertia computation. Suppose

we wish to determine the moment of inertia 𝐼𝑧 of a three-dimensional object, with mass

density 𝜌(𝑥, 𝑦, 𝑧), about the 𝑧-axis. Notice that the relevant integration can be performed over

the 𝑧-direction first.

𝐼𝑧 = ∬ (∫ 𝜌(𝑥, 𝑦, 𝑧) 𝑑𝑧) (𝑥 2 + 𝑦 2 ) 𝑑𝑥 𝑑𝑦 = ∬ 𝜎(𝑥, 𝑦)(𝑥 2 + 𝑦 2 ) 𝑑𝑥 𝑑𝑦

where

𝜎(𝑥, 𝑦) = ∫ 𝜌(𝑥, 𝑦, 𝑧) 𝑑𝑧

Effectively, we have compressed the body along the z-direction such that all points with the

same (𝑥, 𝑦) coordinates collapse into a single point (𝑥, 𝑦) in the 𝑥𝑦-plane, with a surface

mass density 𝜎 (𝑥, 𝑦) – a process illustrated by Figure 8.21. The desired moment of inertia 𝐼𝑧

376
is equal to that of this new two-dimensional object, characterised by 𝜎(𝑥, 𝑦). Note that this

operation trivially preserves the total mass as ∬ 𝜎(𝑥, 𝑦) 𝑑𝑥 𝑑𝑦 = ∭ 𝜌(𝑥, 𝑦, 𝑧) 𝑑𝑧 𝑑𝑥 𝑑𝑦.

Furthermore, this notion of squashing can easily be extended to reducing a two-dimensional

object into a one-dimensional object that is described by a linear mass density.

Figure 8.21

Squashing Operation along 𝒛-Direction

Here is a corollary: for an object with uniform density along the 𝑧-direction, since

𝜎(𝑥, 𝑦) = ∫ 𝜌(𝑥, 𝑦) 𝑑𝑧 = 𝜌(𝑥, 𝑦)𝐿 where 𝐿 is the object’s length in the 𝑧-direction, one can

conclude that the surface mass density of the squashed object will be exactly equal to the

surface mass density of a cross-section with the same total mass! Intuitively, one can also

visualise that the mass distribution on the flattened object is similar to that on an original

cross-section since all cross-sections were identical initially.

One possible strategy for moment of inertia computations is to combine the

perpendicular axis theorem and squashing method. This is because the major limitation of

the perpendicular axis theorem is its applicability to merely flat objects. However, the

squashing operation reduces a multi-dimensional object into an equivalent, lower-

377
dimensional one for moment of inertia calculations – providing an avenue to circumvent this

bottleneck. Therefore, applying the perpendicular axis theorem in tandem with squashing is

a potent strategy.

Using this simple corollary and the aid of the abovementioned theorems, one can

make use of squashing method to derive the moments of inertia of some regular objects, as

illustrated by the following examples.

(1) Uniform Cylinder about Cylindrical Axis

To illustrate the utility of squashing, we can assert that the moment of inertia 𝐼𝑧 of a

uniform cylinder – with mass 𝑀, radius 𝑅 and length 𝐿 – about its cylindrical axis, is equal

to that of a uniform circular disk of mass 𝑀 (because the total mass is unaffected by

squashing) and radius 𝑅, about an axis passing perpendicularly through its centre, since we

obtain a uniform disk by squashing a uniform cylinder along its cylindrical axis. Thus,

1
𝐼𝑧 = 𝑀𝑅2
2

(8.13)

for the cylinder. Equation (8.13) also explains the empirical observation discussed in the

introduction since it is clearly independent of the length of the cylinder, as long as the

cylinder’s mass and radius are unchanged. We shall accept this result first as it provides a

great aid to the rest of the examples. We will prove this result using circular squashing –

which is another great application of mathematical thinking in itself – in the last example.

(2) Cylinder with Sinusoidal Distribution about Cylindrical Axis

378
 For a slightly more complicated example, consider a cylindrical mass distribution

with radius 𝑅 and length 𝐿 parameterised by the volume mass density

𝜌(𝑟, 𝜃, 𝑧) = 𝜌0 |sin(𝜃 + 𝑘𝑧)|

2𝜋
where 𝑘 = and 𝜌0 is a constant. 𝑟, 𝜃 and 𝑧 are the conventional radial, azimuthal and axial
𝐿

coordinates in a cylindrical coordinate system. The origin is located at the centre of one

circular end of the cylinder while the centre of the other end is located along the 𝑧-axis with

a positive 𝑧-coordinate 𝐿. Thus, the above equation is only valid in the regime 0 ≤ 𝑟 ≤ 𝑅

and 0 ≤ 𝑧 ≤ 𝐿. To compute the moment of inertia 𝐼𝑧 of this object with respect to the 𝑧-axis,

we can squash it along the 𝑧-direction to obtain a planar circle with surface mass density
𝐿

𝜎(𝑟, 𝜃) = ∫ 𝜌0 |sin(𝜃 + 𝑘𝑧)| 𝑑𝑧

0

Adopting the substitution 𝑢 = 𝜃 + 𝑘𝑧 and 𝑑𝑢 = 𝑘 𝑑𝑧,

2𝜋+𝜃 2𝜋 2𝜋+𝜃
𝜌0 𝜌0
𝜎(𝑟, 𝜃) = ∫ |sin 𝑢| 𝑑𝑢 = (∫ |sin 𝑢| 𝑑𝑢 + ∫ |sin 𝑢| 𝑑𝑢)
𝑘 𝑘
𝜃 𝜃 2𝜋

2𝜋 𝜃 2𝜋 𝜋
𝜌0 𝜌0 2𝜌0
𝜎 (𝑟, 𝜃) = (∫ |sin 𝑢| 𝑑𝑢 + ∫|sin 𝑢| 𝑑𝑢) = ∫ |sin 𝑢| 𝑑𝑢 = ∫ sin 𝑢 𝑑𝑢
𝑘 𝑘 𝑘
𝜃 0 0 0

2𝜌0 𝐿
𝜎 (𝑟, 𝜃) =
𝜋

where we have exploited the facts that sin(2𝜋 + 𝑥 ) = sin 𝑥 and |sin(𝜋 + 𝑥 )| =

|sin(𝑥 − 𝜋)| = |− sin 𝑥| = sin 𝑥 for 0 ≤ 𝑥 ≤ 𝜋. Observe that after compressing the

2𝜌0𝐿
cylindrical distribution, we obtain a uniform circle with total mass 𝑀 = 𝜋𝑅2 = 2𝜌0 𝐿𝑅2 .
𝜋

Thus,

379
 1
𝐼𝑧 = 𝑀𝑅2 = 𝜌0 𝐿𝑅4
2

(8.14)

Incidentally, the idea of squashing also presents a vivid physical picture of why 𝐼𝑧 of this

cylindrical distribution should be identical to that of a uniform circle. Since the cylindrical

distribution into circular cross-sections of infinitesimal thickness 𝑑𝑧 at different 𝑧-

coordinates. Since the equation signifies that the cross-section at 𝑧-coordinate 𝑧 is basically

that at the origin, rotated by an angle 𝑘𝑧, and because 𝑧 ranges from 0 to 𝐿 such that 𝑘𝑧 varies

from 0 to 2𝜋, the squashing operation simply combines a continuous set of cross-sections

that progressively rotate for one complete revolution. Hence, the surface mass distribution

obtained is naturally a uniform circle.

(3) Uniform Ellipse about Major and Minor Axes

Next, the moments of inertia of a uniform ellipse – with mass 𝑀, semi-major axis 𝑎

and semi-minor axis 𝑏 – will be derived about its major and minor axes. Choose the origin

at the centre of the ellipse and orient the 𝑥- and 𝑦-axes along the major and minor axes
𝑀
respectively. For the sake of convenience, define 𝜎 = 𝜋𝑎𝑏 as the uniform surface mass density

ellipse
of the ellipse. To compute 𝐼𝑥 , squash the ellipse along the 𝑥-direction to obtain a rod

spanning from 𝑦 = −𝑏 to 𝑦 = 𝑏 with linear mass density

2
√𝑎 2 −𝑎 2 𝑦 2
𝑏
𝑎2 2
𝜆ellipse (𝑦) = ∫ 𝜎 𝑑𝑥 = 2𝜎√𝑎2 − 𝑦
𝑏2
𝑎2
−√𝑎 2 − 2 𝑦 2
𝑏

380
where we have noted from the standard equation describing the perimeter of an ellipse

𝑥2 𝑦2
(𝑎2 + 𝑏2 = 1) that for a given 𝑦-coordinate 𝑦, the 𝑥-coordinate of points on the ellipse ranges

𝑎2 𝑎2
from 𝑥 = −√𝑎2 − 𝑏2 𝑦 2 to 𝑥 = √𝑎2 − 𝑏2 𝑦 2 . Now, compare the above equation with the

linear mass density 𝜆circle (𝑦) of a different rod obtained from compressing a uniform circle

with radius 𝑏 and surface mass density 𝜎 along the 𝑥-axis that passes through its centre.

√𝑏2 −𝑦 2

𝜆circle (𝑦) = ∫ 𝜎 𝑑𝑥 = 2𝜎√𝑏2 − 𝑦 2

−√𝑏2 −𝑦 2

𝑎
Since 𝜆ellipse (𝑦) = 𝑏 𝜆circle (𝑦) and both rods span from 𝑦 = −𝑏 to 𝑦 = 𝑏, their moments of

ellipse
inertia 𝐼𝑥 and 𝐼𝑥circle must be related by

ellipse 𝑎 circle
𝐼𝑥 = 𝐼
𝑏 𝑥

because scaling the mass of a distribution by a factor 𝑘 also changes its moment of inertia by

a factor of 𝑘, given that its geometry is fixed. Applying perpendicular axis theorem, we obtain

the moment of inertia of a circle of radius 𝑏 about its diameter,

1
𝐼𝑥circle = 𝐼𝑦circle = (𝜎𝜋𝑏2 )𝑏2
4

Hence,

ellipse 𝑎 1 𝑀 1
𝐼𝑥 = ∙ ∙ (𝜋𝑏2 )𝑏2 = 𝑀𝑏2
𝑏 4 𝜋𝑎𝑏 4

An intuitive explanation of this result is that the ellipse is merely a circle of radius 𝑏 stretched
𝑎
along the 𝑥-axis by a factor of 𝑏 . Since the extent of an object along the 𝑥-direction is

ellipse
irrelevant when computing its moment of inertia about the 𝑥-axis, 𝐼𝑥 is naturally the

381
moment of inertia of a circle with radius 𝑏 and identical mass 𝑀 about a diameter. Moving

on, one can similarly conclude

ellipse 1
𝐼𝑦 = 𝑀𝑎2
4

about the minor axis of the ellipse. Finally, one can even apply the perpendicular axis theorem

to write

ellipse ellipse ellipse

𝐼𝑧 = 𝐼𝑥 + 𝐼𝑦

ellipse 1
𝐼𝑧 = 𝑀 (𝑎 2 + 𝑏 2 )
4

(8.15)

about the perpendicular axis passing through the centre of the ellipse.

(4) Uniform Pyramid about Perpendicular Axis through Apex

Consider a hollow pyramid with a uniform surface mass density and total mass 𝑀. If

we are interested in computing its moment of inertia 𝐼𝑧hollow about the 𝑧-axis that passes

through its apex and is perpendicular to its base, we can squash the pyramid in the 𝑧-direction

to obtain a uniform surface mass distribution that takes the shape of its base. Thus, 𝐼𝑧hollow of

any uniform hollow pyramid is simply the moment of inertia of its base – if it were to possess

a uniform surface mass density such that its total mass is identically 𝑀 – about a

perpendicular axis through its centroid.

Applying this result to a uniform hollow cone with mass 𝑀 and base radius 𝑅,

1
𝐼𝑧hollow = 𝑀𝑅2
2

382
as the moment of inertia of a uniform circle with mass 𝑀 and radius 𝑅, about a perpendicular

1
axis through its centre, is 2 𝑀𝑅2 . As another example, 𝐼𝑧hollow of a uniform and hollow regular

tetrahedron with mass 𝑀 and edge length 𝐿 is

1
𝐼𝑧hollow = 𝑀𝐿2
12

as its base is an equilateral triangle with edge length 𝐿. Finally, we can extend our analysis

to determine 𝐼𝑧solid (with a similar definition of the 𝑧-axis) of a uniform solid pyramid with

mass 𝑀. Firstly, suppose we have concluded that 𝐼𝑧hollow of its hollow counterpart with mass

𝑚 is

𝐼𝑧hollow = 𝑘𝑚ℓ2

where 𝑘 is a constant and ℓ is a length dimension. Now, notice that we can deem a uniform

solid pyramid as many thin shells of uniform hollow pyramids with similar shapes, enveloped

by one another, about a common 𝑧-axis. Let the length dimensions of these layers range from

ℓ = 0 to ℓ = 𝐿 where 𝐿 is the extreme value of the relevant length dimension in the solid

pyramid. The contribution of the shell – with mass 𝑑𝑚, thickness 𝑑ℓ and length dimension

ℓ – to 𝐼𝑧solid is thus 𝑑𝐼𝑧solid = 𝑘ℓ2 𝑑𝑚. Critically, observe that the area of a uniform-shell

scales with ℓ2 such that 𝑑𝑚 = 𝑎ℓ2 𝑑ℓ for some constant 𝑎. The total mass of the solid

pyramid is then
𝐿
1
𝑀 = ∫ 𝑑𝑚 = ∫ 𝑎ℓ2 𝑑ℓ = 𝑎𝐿3
3
0

while

383
 𝐿
1 3
𝐼𝑧solid = ∫ 𝑘ℓ2 𝑑𝑚 = ∫ 𝑎𝑘ℓ4 𝑑ℓ = 𝑎𝑘𝐿5 = 𝑘𝑀𝐿2
5 5
0

As such, for a uniform solid cone with mass 𝑀 and base radius 𝑅,

3 1 3
𝐼𝑧solid = ∙ 𝑀𝑅2 = 𝑀𝑅2
5 2 10

while

3 1 1
𝐼𝑧solid = ∙ 𝑀𝐿2 = 𝑀𝐿2
5 12 20

for a uniform and solid regular tetrahedron with mass 𝑀 and edge length 𝐿.

(5) Uniform Cuboid about Perpendicular Axis through Centroid

Consider a uniform cuboid with mass 𝑀 and edge lengths 𝑎, 𝑏 and 𝑐 parallel to the

𝑥-, 𝑦- and 𝑧-axes, respectively. Suppose we wish to determine the moment of inertia 𝐼𝑧 of

this cuboid about the 𝑧-axis passing through its centroid. We can first squash the cuboid along

the 𝑧-direction to obtain a uniform rectangle with edge lengths 𝑎 and 𝑏. The perpendicular

axis theorem implies 𝐼𝑧 = 𝐼𝑥 + 𝐼𝑦 where 𝐼𝑥 and 𝐼𝑦 are the moments of inertia of the rectangle

with respect to the 𝑥- and 𝑦-axes lying in the plane of the rectangle and passing through its

centroid. To compute 𝐼𝑥 , we can squash the uniform rectangle along the 𝑥-direction to obtain

a uniform rod of length 𝑏, for which 𝐼𝑥 will be computed about a perpendicular axis crossing

its centre. Remembering that the mass of this rod is still 𝑀 since the squashing operation
1
preserves the total mass, 𝐼𝑥 = 12 𝑀𝑏2 . Similarly, we can squash the rectangle along the y-

1
direction to conclude that 𝐼𝑦 = 12 𝑀𝑎2 . Hence,

𝐼𝑧 = 𝐼𝑥 + 𝐼𝑦

384
 1
𝐼𝑧 = 𝑀 (𝑎 2 + 𝑏 2 )
12

(8.16)

(6) Uniform Cylinder about Transverse Axis through Centroid

Next, we wish to determine the moment of inertia of a uniform solid cylinder of mass

𝑀, radius 𝑅 and length 𝐿 about the 𝑧-axis that is perpendicular to its cylindrical axis, passing

through its centroid O.

Referring to Figure 8.22, we can first squash the cylinder along the 𝑧-axis to obtain a

non-uniform mass distribution in the shape of a rectangle in the 𝑥𝑦-plane.

Figure 8.22

Squashing Cylinder along 𝒛-Direction

The moment of inertia of this non-uniform plate is tedious to determine. However, we can

now apply the perpendicular axis theorem 𝐼𝑧 = 𝐼𝑥 + 𝐼𝑦 where 𝐼𝑥 and 𝐼𝑦 are the moments of

inertia of the plate about the 𝑥- and 𝑦-axes passing through O (the 𝑦-axis is parallel to the

original cylindrical axis). To compute 𝐼𝑥 and 𝐼𝑦 , the plate can be squashed along the

corresponding directions again to obtain one-dimensional distributions. For the 𝑥-direction,

385
the squashed plate becomes a uniform rod of mass 𝑀 and length 𝐿, for which 𝐼𝑥 is computed

about a perpendicular axis passing through its centre. Hence,

1
𝐼𝑥 = 𝑀𝐿2
12

The moment of inertia of the resultant distribution (the non-uniform rod in the middle of

Figure 8.23) after squashing in the 𝑦-direction is less obvious.

Figure 8.23

Squashing Plane along 𝒚-Direction

That said, we can exploit the symmetry of the rod to claim that its moments of inertia

𝐼𝑦 ′ and 𝐼𝑧 ′ relative to the 𝑦- and 𝑧-axes, are identical.

𝐼𝑦 ′ = 𝐼𝑧 ′

Since the rod is equivalent to a uniform circle – of mass 𝑀 and radius 𝑅 in the 𝑥𝑧-plane,

squashed along the 𝑧-direction, 𝐼𝑧 ′ is equal to that of a uniform circle about a diameter.

1
𝐼𝑧 ′ = 𝑀𝑅2
4
1 1
Recalling that 𝐼𝑥 = 12 𝑀𝐿2 , 𝐼𝑦 = 𝐼𝑦 ′ = 𝐼𝑧 ′ = 4 𝑀𝑅2 ,

386
 1 1
𝐼𝑧 = 𝐼𝑥 + 𝐼𝑦 = 𝑀𝐿2 + 𝑀𝑅2
12 4

(8.17)

(7) Uniform Triangle about Axis through Vertex and Parallel to Opposite Edge

Referring to the left diagram in Figure 8.24, consider an arbitrary, uniform triangle

with mass 𝑀 and a 𝑧-axis that passes through a vertex O, parallel to the edge opposite O. The

altitude of the triangle from O is ℎ and we are interested in 𝐼𝑧 of the triangle.

Figure 8.24

Squashing Uniform Triangle along 𝒛-Axis

Squashing the triangle along the 𝑧-direction, we obtain a rod of total mass 𝑀 whose

linear mass density is directly proportional to the distance from the 𝑧-axis (by similar

triangles). Observe that the same rod will be generated after squashing, regardless of the

shape of the triangle that we started with! This is because all such rods have the same mass

𝑀 and the same linear variation of mass density. Thus, all triangles with mass 𝑀 (note that

their surface mass densities will generally differ) and altitude ℎ have the same 𝐼𝑧 . In

387
particular, we can determine 𝐼𝑧 of the right-angle isosceles triangle depicted on the right of

Figure 8.24. We choose this right-angle isosceles triangle for convenience.

To this end, apply the parallel axis theorem to this equivalent triangle,

2ℎ 2 4
𝐼𝑧CM = 𝐼𝑧 − 𝑀 ( ) = 𝐼𝑧 − 𝑀ℎ2
3 9

where 𝐼𝑧CM is the moment of inertia about an axis, parallel to the 𝑧-direction, crossing the

centroid of the right triangle. Similarly,

ℎ 2 1
𝐼𝑦CM = 𝐼𝑦 − 𝑀 ( ) = 𝐼𝑦 − 𝑀ℎ2
3 9

Due to symmetry, 𝐼𝑧CM = 𝐼𝑦CM , such that

1
𝐼𝑦 = 𝐼𝑧 − 𝑀ℎ2
3

Finally, if 𝐼𝑥 of the triangle (the positive x-axis is pointing out of the page under a

conventional right-handed coordinate system) can be determined, we are done since the

perpendicular axis theorem implies 𝐼𝑥 = 𝐼𝑦 + 𝐼𝑧 . There are many ways of calculating 𝐼𝑥 given

the simplicity of this equivalent set-up but one method is to conjoin the vertices O of eight

such triangles to form a square of total mass 8𝑀 and edge length 2ℎ about centroid O. Using

the previously-derived moment of inertia of a rectangle, the moment of inertia of this square
1 16
about a perpendicular axis through its centroid is (8𝑀)((2ℎ)2 + (2ℎ)2 ) = 𝑀ℎ2 .
8 3

Because this square comprises eight triangles that are of concern,

1 16 2
𝐼𝑥 = ∙ 𝑀ℎ2 = 𝑀ℎ2
8 3 3

for one triangle. Then,

𝐼𝑥 = 𝐼𝑦 + 𝐼𝑧

388
 2 1
𝑀ℎ2 = (𝐼𝑧 − 𝑀ℎ2 ) + 𝐼𝑧
3 3

1
𝐼𝑧 = 𝑀ℎ2
2

(8.18)

(8) Uniform Hollow Cone about Parallel Axis through Apex

Next, we wish to determine the moment of inertia 𝐼𝑧 of a hollow cone, with a uniform

surface mass density, about the z-axis (parallel to the base) shown in Figure 8.25. Its mass,

height and base radius are 𝑀, ℎ and 𝑅, respectively.

Figure 8.25

Squashing Hollow Cone along 𝒛-Direction

Squashing the cone along the 𝑧-direction, we obtain a non-uniform isosceles triangle in the

𝑥𝑦-plane. 𝐼𝑧 is identical to the moment of inertia 𝐼𝑧 ′ of this triangle about the 𝑧-axis. Applying

the perpendicular axis theorem, 𝐼𝑧 ′ = 𝐼𝑥 ′ + 𝐼𝑦 ′ where 𝐼𝑥 ′ and 𝐼𝑦 ′ are the moments of inertia

of this triangle about the 𝑥- and 𝑦-axes, respectively.

389
 Let us consider 𝐼𝑥 ′ first. Squashing the triangle along the x-direction, we obtain a non-

uniform rod whose moment of inertia is troublesome to determine. However, notice that we

can reverse the order of squashing to obtain the same rod – namely, to squash along the 𝑥-

direction before squashing along the 𝑧-direction. Compressing the original cone along the 𝑥-

direction yields a uniform circle, which is subsequently squashed along the 𝑧-direction.

Because the resultant rod lies entirely along the y-axis, its moments of inertia about the 𝑥-

and 𝑧-axes are identical. The former is 𝐼𝑥 ′ while the latter can be easily deduced from the
1
alternate order of squashing as 4 𝑀𝑅2 since it is simply the moment of inertia of a uniform

1
circle with mass 𝑀 and radius 𝑅, about a diameter. Thus, 𝐼𝑥 ′ = 4 𝑀𝑅2 .

To compute 𝐼𝑦 ′, squash the triangle along the 𝑦-direction to obtain another rod of

length ℎ. Revisiting the original cone, we realise that the squashing operations in the 𝑦- and

𝑧-directions have effectively compressed the circular cross-section of the cone at each 𝑥-

coordinate into a corresponding point on the 𝑥-axis. As the circumference and thus mass of

a cross-section scales linearly with its height from the apex O, the linear mass density of a

section on the final rod is directly proportional to its distance from the 𝑦-axis. Luckily, the

moment of inertia of such a rod was exactly computed in the previous section, from which
1
𝐼𝑦 ′ = 2 𝑀ℎ2 . Piecing the above information together,

1 1
𝐼𝑧 = 𝐼𝑧 ′ = 𝐼𝑥 ′ + 𝐼𝑦 ′ = 𝑀𝑅2 + 𝑀ℎ2
4 2

(8.19)

390
(9) Uniform Disk about Perpendicular Axis through Centroid

Finally, we can extend our analysis to a uniform disk by introducing the idea of

circular squashing, as opposed to the linear squashing hitherto. Observe from the definition

of the moment of inertia that the contribution of a particle is only dependent on its

perpendicular distance to the relevant axis and independent of its azimuthal coordinate.

Therefore, we can rotate all particles with identical perpendicular distances about the axis,

until they reach a single azimuthal position, without altering the moment of inertia of the

object since the perpendicular distance of each particle to the axis is maintained. The total

mass of the object is also evidently preserved. Applying this process to a uniform disk, with

mass 𝑀 and radius 𝑅, about the 𝑧-axis passing perpendicularly through its centroid, we obtain

a rod whose linear mass density varies linearly with distance from the original centroid. This

is because for every intermediate radius (0 ≤ 𝑟 ≤ 𝑅), the uniform points along the perimeter

of the concentric circle of radius 𝑟 are swept into a single point and this perimeter scales

linearly with 𝑟. More vividly, all radial lines on the circle are combined into a single radial

line – akin to closing a Chinese hand fan. Because the resultant rod possesses an identical

mass 𝑀 and spans from distance 𝑟 = 0 to 𝑟 = 𝑅 from the original centroid, our previous

result for the moment of inertia of a rod – whose linear mass density is directly proportional

to distance from the origin – implies that 𝐼𝑧 of the uniform disk is

1
𝐼𝑧 = 𝑀𝑅2
2

(8.20)

391
 Scaling method works in general, as long as the object can be sliced into a contracted

version of the original one, or a number of identical objects can be put together into an

enlarged version of the original one. The last two examples, particularly, show how one can

extend this property to determine the moment of inertia of three-dimensional objects, namely

hollow sphere and solid sphere. Squashing method also works in general, as long as the object

can be squashed into another object with reduced dimensions, whose moment of inertia can

be computed easily or at least easier with the application of the parallel axis or perpendicular

axis theorems. The last example shows how one can extend this method to do circular

squashing.

We conclude that there is a more straightforward, if not easier and preferred method

for the computation of regular and symmetric rigid bodies. The methods presented in this

chapter makes use of the symmetrical property of the objects as well as the application of the

two theorems. The approaches will be more appreciated by students and more easily

understood as the application of calculus is not required. The development of these two new

methods – scaling and squashing – are an excellent aspect of mathematical thinking that

could be explored further.

8.2.2. Educational Studies on Mathematical Thinking in Physics

Due to the limitations of the study, as elaborated in Section 8.1, further educational

studies can be done pertaining to mathematical thinking in physics. First and foremost, the

study can be replicated for other topics in physics, as the current study only explored two

specific topics. Prior to the construction of a suitable instrument, theoretical explorations

need to be done on the topic explored. The use of mathematical thinking must come strongly

392
in this exploration in order to create higher order thinking questions that can be used in the

assessment.

Secondly, should luxury of time be not an issue, the researcher may want to increase

the number of questions in the assessment to cover more breadth and depth in the specific

topic explored. In the study on momentum/collisions, for instance, more questions can be

constructed pertaining to impulse, explosion, two- and three-dimensional collisions, etc. In

the study on rotational motion, aspects such as rolling without slipping, Newton’s law for

translation and rotation, moment of inertia, and others can also be included in the assessment.

Having more questions in the assessment also allows us to cover more aspects of

mathematical thinking, in order to see more benefits on the use of mathematical thinking in

physics. In the current study, only the aspect of problem-solving was explored in higher order

thinking questions.

One of the limitations of the current study is the narrow pool of participants as they

were taken from only one school in Singapore, NUS High School of Mathematics and

Science. Future studies may want to explore the use of mathematical thinking in physics in

wider pool of participants to see the generalisability of the results. Admittedly, the

requirement that the study participants must be high performers in physics puts a constraint

in the study. Nevertheless, a researcher can always extend the study to physics students of

higher level, e.g. top junior colleges and university students, who have more exposure and

maturity in mathematical thinking.

Lastly, a more robust educational study can be performed to see the effect of teaching

intervention strategy on promoting mathematical thinking. In our current study, the teacher

who conducted the lesson intervention was the researcher himself. Although all precautions

393
were taken to ensure that the teacher does not teach to the test, having a different set of

persons conducting the research and teaching the lesson intervention would be preferable.

However, as mentioned in the earlier section, this kind of research could not be done by any

teacher or any social study researcher, as it is mandatory that both the researcher performing

the research and the teacher conducting the lesson intervention have sufficient depth of

knowledge in physics concepts and skills in mathematics.

8.3. Study Implications

The thesis study has several important implications on physics and physics education

in general. Mainly, it demonstrates that the teaching intervention that focuses on

mathematical thinking in physics conceptual learning is effective in improving students’

performance in physics assessment, especially where higher order thinking problems are

involved. This intervention could be implemented more in teaching physics to understand its

prolonged effect on students’ ability in conceptualising ideas and analysing approaches for

problem-solving. We hope that as students’ performance and analytical skills improve

through the use of mathematical thinking, long-lasting positive effects can be seen as more

high-performing students would find physics more interesting.

Overall the findings of this thesis study are believed to carry potential benefits, which

are discussed below:

(1) Mathematical thinking allows one to analyse a physical situation more deeply.

The hope is that the use of mathematics can be enhanced in physics teaching and

learning as well as in the analyses of natural phenomena in general, as opposed to the sole

394
use of qualitative analysis or even intuition based on common sense beliefs. This teaching

intervention could motivate students to build their rational ways of thinking and enable them

to perform the desired analysis to a physics problem. This improvement was clearly shown

in the way students approached the higher order thinking questions in the assessment after

the teaching intervention, both qualitatively and quantitatively. On the other hand, in the

teaching intervention, teachers could also encourage students to utilise mathematical

formulas, tricks and ability they have had to bridge connections between pieces of

information provided in a physical situation. This is in line with representational fluency –

that is, being able to toggle between different levels of representations for the same concept

can lead to meaningful understanding. The findings also show that this teaching intervention

tends to minimise one’s overreliance on intuition based on common sense beliefs (Kaiser et

al., 1986). The use of mathematics is expected to help students develop critical thinking when

analysing natural phenomena instead of immediately jumping into conclusions based on

common sense beliefs.

(2) Mathematical thinking in physics makes students realise the interconnections

between the two subjects, Mathematics and Physics, and to see how mathematics

sharpens their analytical skills in many situations, especially when applied to physical

phenomena.

Early and continuous introduction of mathematical thinking in physics learning

would help students to perceive applications of mathematics more positively, especially in

cross-subject applications. Many students have wrong perceptions when it comes to learning

mathematics. Even high-performing students in mathematics perceive Mathematics as a

395
standalone subject that has nothing to do with real life phenomena and no connection at all

with other subjects. On the other hand, many students perceive science, including physics, as

just some day-to-day phenomena with the addition of some process skills required to perform

analyses for a given science problem. These perceptions are greatly influenced by certain

teaching pedagogies subscribed by teachers, as the teachers’ beliefs about the subject may

influence students’ learning attitudes (Politis et al, 2007). If teachers have the confidence that

introducing mathematics would help the students perform better and are able to encourage

them to do so, we believe that students’ perceptions and attitudes when it comes to learning

and applying mathematics would change.

(3) Mathematical thinking helps students to perform better in physics assessment which

involves higher order thinking questions.

Students’ improvement in physics assessment should not be underestimated as they

have a prolonged effect towards students’ perceptions on the subject as well as students’

attitudes towards and aptitudes for the subject. The findings have shown that students in the

experimental group acquired better mathematical thinking skills as well as conceptual

changes after the teaching intervention. After being exposed to this newly-explored teaching

intervention, the number of alternative conceptions harboured and the percentage of students

with the alternative conceptions decreased significantly. We hope that students can show

better appreciation towards the beauty of physics – not just as theory and formulas to

memorise, or some process skills to acquire – it is the beauty of thoroughly analysing a simple

system using simple physics concepts which, in turn, brings them to understand the

complexities of that system.

396
 We hope that our findings in this study would encourage teachers to not avoid the use

of mathematics in teaching physics, especially when it comes to teaching high-performing

students who have high competency in mathematics. Indeed, mathematical thinking is

indispensable for promoting holistic understanding of physics concepts. Oftentimes, since

mathematical thinking is scarcely used, students are unable to unravel the beauty of thinking

mathematically in physics. For high performers, we believe that instead of diminishing

students’ interest in science, the extensive use of mathematics would increase their

motivation and curiosity to pursue the topics of interest even more. As the extensive use of

mathematical thinking in physics conceptual learning has not been explored earlier in the

educational literature, this study is of particular interest as it provides a new approach in the

study of physics and physics education, in general.

8.4. Contributions of This Study to New Knowledge

The novelty of our two newly-formulated concept inventories is expected to add value

to the library of physics assessments as, to the best of our knowledge, no other 5-tier

diagnostic tests have been developed pertaining to the extensive use of mathematical thinking

and specifically targeted for high-performing students. The instruments are unique in that

they test conceptual understanding and mathematical thinking at the same time. Moreover,

they do not just assess students’ cognitive ability in applying concepts to attain correct

answers, but also their reasoning as well as metacognitive skills via all the different tiers.

These concept inventories have been validated and tested, and they are readily-available for

397
use in replication studies in the future. It is hoped to encourage more teachers and researchers

to explore further aspects of mathematical thinking in physics.

The development and validation of the 5-tier instruments have also given us insights

on how to accommodate various cognitive and metacognitive aspects in a single instrument.

As elaborated in Section 7.1, in the construction process of the 5-tier instrument, it was found

necessary for the fifth tier to be in free-response format instead of multiple-choice. It is

indeed challenging if a multiple-choice format were to be used to excerpt students’

mathematical thinking as each option had to be written mathematically. And even if all the

options were mathematically-oriented, it would still not be able to capture much of students’

mathematical thinking. This 5-tier format is a new and unique format of our study

instruments.

This thesis study also breaks new ground in physics education research. Although

previous studies have identified mathematical thinking as an important factor in the study of

physics, no other researchers have evaluated correlations between conceptual understanding

and mathematical thinking using the same assessment instrument. In our teaching

intervention strategy, mathematical thinking was strongly promoted in order for students to

develop the habit of performing mathematical thinking as they approached physical

situations. This intervention strategy also has its novelty, as it focuses on mathematical

thinking, while most other teaching intervention strategies focus on identifying alternative

conceptions and addressing these by promoting cognitive conflict.

398
 REFERENCES

AACU 2010. American Association of Colleges and Universities. Quantitative Literacy

VALUE Rubric. Retrieved from http://www.aacu.org/value/rubrics/quantitative-literacy.

Adadan, E., & Savasci, F. (2012). An analysis of 16–17-year-old students' understanding of

solution chemistry concepts using a two-tier diagnostic instrument. International Journal of
Science Education, 34(4), 513-544.

Adodo, S. O. (2013). Effects of two-tier multiple choice diagnostic assessment items on

students’ learning outcome in basic science technology (BST). Academic Journal of
Interdisciplinary Studies, 2(2), 201.

Afif, N. F., Nugraha, M. G., & Samsudin, A. (2017, May). Developing energy and
momentum conceptual survey (EMCS) with four-tier diagnostic test items. In AIP
Conference Proceedings (Vol. 1848, No. 1, p. 050010). AIP Publishing.

Aghadiuno, M. C. K. (1992). Mathematics: history, philosophy and applications to science.

International Journal of Mathematical Education in Science and Technology, 23(5), 683-
690.

Ainsworth, S. (2006). DeFT: A conceptual framework for considering learning with multiple
representations. Learning and Instruction, 16(3), 183-198.

Airey, J., & Linder, C. (2009). A disciplinary discourse perspective on university science
learning: Achieving fluency in a critical constellation of modes. Journal of Research in
Science Teaching, 46(1), 27-49.

Aldrich, F., & Sheppard, L. (2000). ‘Graphicacy’: the fourth ‘R’? Primary Science Review,
64, 811.

Allmond, S., & Makar, K. (2010, July). Developing primary students’ ability to pose
questions in statistical investigations. In Proceedings of the 8th international conference on
teaching statistics. Voorburg, The Netherlands: International Statistical Institute.

Al-Rubayea, A. A. M. (1996). An analysis of Saudi Arabian high school students’

misconceptions about physics concepts in Yasin, K.(2004), a M. Sc (Doctoral dissertation,
dissertation thesis report. M. Sc. Thesis report, Middle East Technical University).

Andersen, W. L. (2007). Noncalculus treatment of steady-state rolling of a thin disk on a

horizontal surface. The Physics Teacher, 45, 430-433.

Anderson, C. W., & Smith, E. L. (1984). Children’s preconceptions and content-area

textbooks. Comprehension Instruction: Perspectives and Suggestions, 187-201.

399
Andersson, B., & Kärrqvist, C. (1983). How Swedish pupils, aged 12‐15 years, understand
light and its properties. European Journal of Science Education, 5(4), 387-402.

Applebee, A. N. (1986). Problems in process approaches: Toward a reconceptualization of

process instruction. The Teaching of Writing, 85, 95-113.

Aşıksoy, G., & Özdamlı, F. (2016). Flipped Classroom adapted to the ARCS Model of
Motivation and applied to a Physics Course. Eurasia Journal of Mathematics, Science &
Technology Education, 12(6).

Atiyah, M. (1990). On the work of Edward Witten. In Proceedings of the International

Congress of Mathematicians, Kyoto, Japan (pp. 31-35).

Aydın, Ö. (2007). Assessing tenth grade students’ difficulties about kinematics graphs by a
three-tier test (Master’s thesis, Middle East Technical University).

Aykutlu, I., & Şen, A. İ. (2012). Üç Aşamalı Test, Kavram Haritası ve Analoji Kullanılarak
Lise Öğrencilerinin Elektrik Akımı KonusundakiKavram Yanılgılarının Belirlenmesi.
Eğitim ve Bilim, 37(166).

Bailly, F., & Longo, G. (2011). Mathematics and the natural sciences: the physical
singularity of life (Vol. 7). World Scientific.

Ballesteros, A., Enciso, A., Herranz, F. J., & Ragnisco, O. (2008). Bertrand spacetimes as
Kepler/oscillator potentials. Classical and Quantum Gravity, 25(16), 165005.

Barenblatt, G. I. (1996). Scaling, self-similarity, and intermediate asymptotics: dimensional

analysis and intermediate asymptotics (Vol. 14). Cambridge University Press.

Barnes, G., McInerney, D. M., & Marsh, H. W. (2005). Exploring sex differences in science
enrolment intentions: An application of the general model of academic choice. The
Australian Educational Researcher, 32(2), 1-23.

Barnett, S. M., & Ceci, S. J. (2002). When and where do we apply what we learn?: A
taxonomy for far transfer. Psychological Bulletin, 128(4), 612.

Barr, V., & Stephenson, C. (2011). Bringing computational thinking to K-12: what is
Involved and what is the role of the computer science education community?. Inroads, 2(1),
48-54.

Basili, P. A., & Sanford, J. P. (1991). Conceptual change strategies and cooperative group
work in chemistry. Journal of Research in Science Teaching, 28(4), 293-304.

400
Beeth, M. E. (1998). Teaching for conceptual change: Using status as a metacognitive tool.
Science Education, 82(3), 343-356.

Beichner, R. J. (1994). Testing student interpretation of kinematics graphs. American Journal

of Physics, 62(8), 750-762.

Bender, C. M., Brody, D. C., & Meister, B. K. (2000). Quantum mechanical Carnot engine.
Journal of Physics A: Mathematical and General, 33(24), 4427.

Bennett, J. (2002). Using Research To Improve Teaching and Learning: What Are the
Issues?. Education in Science, 197, 18-19.

Benson, J., & Hocevar, D. (1985). The impact of item phrasing on the validity of attitude
scales for elementary school children. Journal of Educational Measurement, 22(3), 231-240.

Bhaskar, R., & Nigam, A. (1990). Qualitative physics using dimensional analysis. Artificial
Intelligence, 45(1-2), 73-111.

Bieda, K. N., & Nathan, M. J. (2009). Representational disfluency in algebra: Evidence from
student gestures and speech. ZDM, 41(5), 637-650.

Blumenthal, R. H. (1961). Multiple instruction and other factors related to achievement in

college physics. Science Education, 45(4), 336-342.

Boaler, J. (1997). Experiencing school mathematics: Teaching styles, sex and setting.
Buckingham, UK: OpenUniversity Press.

Boekaerts, M., & Rozendaal, J. S. (2010). Using multiple calibration indices in order to
capture the complex picture of what affects students’ accuracy of feeling of confidence.
Learning and Instruction, 20(5), 372-382.

Bowen, G. M., Roth, W. M., & McGinn, M. K. (1999). Interpretations of graphs by university
biology students and practicing scientists: Toward a social practice view of scientific
representation practices. Journal of Research in Science Teaching, 36(9), 1020-1043.

Brewe, E., Kramer, L., & O’Brien, G. (2009). Modeling instruction: Positive attitudinal shifts
in introductory physics measured with CLASS. Physical Review Special Topics-Physics
Education Research, 5(1), 013102.

Brown, D. E. (1992). Using examples and analogies to remediate misconceptions in physics:

Factors influencing conceptual change. Journal of Research in Science Teaching, 29(1), 17-
34.

401
Bryce, T., & Macmillan, K. (2005). Encouraging conceptual change: the use of bridging
analogies in the teaching of action–reaction forces and the ‘at rest’ condition in physics.
International Journal of Science Education, 27(6), 737-763.

Bukova, E. (2006). The development of new curriculum to overcome students’ difficulties in

perceiving the concept of limit and constructing the relationship between the concept of limit
and the other mathematical concepts (Unpublished doctoral dissertation, Dokuz Eylul
University Institute of Educational Science, İzmir).

Burton, L. (1999). Why is intuition so important to mathematicians but missing from

mathematics education?. For the Learning of Mathematics, 19(3), 27-32.

Caleon, I. S., & Subramaniam, R. (2008). Attitudes towards science of intellectually gifted
and mainstream upper primary students in Singapore. Journal of Research in Science
Teaching, 45(8), 940-954.

Caleon, I. S., & Subramaniam, R. (2010b). Do students know what they know and what they
don’t know? Using a four-tier diagnostic test to assess the nature of students’ alternative
conceptions. Research in Science Education, 40(3), 313-337.

Caleon, I., & Subramaniam, R. (2010a). Development and application of a three‐tier

diagnostic test to assess secondary students’ understanding of waves. International Journal
of Science Education, 32(7), 939-961.

Carey, S. (1986). Cognitive science and science education. American Psychologist, 41(10),
1123.

Carmichael, P., Driver, R., Holding, B., Phillips, I., Twigger, D., & Watts, M. (1991).
Research on Students’ Conceptions in Science: A Bibliography. Children’s Learning in
Science Research Group.

Carr, W., & Kemmis, S. (1986). Becoming critical: Education, knowledge and action
research (rev. edn). Burwood, VIC: Deakin University.

Carvalho, P. S., & e Sousa, A. S. (2005). Rotation in secondary school: teaching the effects
of frictional force. Physics Education, 40(3), 257.

Champagne, A. B., & Klopfer, L. E. (1977). A sixty-year perspective on three issues in

science education:(III) Reflective thinking and problem solving. Science Education, 61(4),
431-452.

Champagne, A. B., & Klopfer, L. E. (1984). Research in science education: The cognitive
psychology perspective. Learning Research and Development Center, University of
Pittsburgh.

402
Champagne, A. B., Klopfer, L. E., & Anderson, J. H. (1980). Factors influencing the learning
of classical mechanics. American Journal of Physics, 48(12), 1074-1079.

Chang, C. Y., Yeh, T. K., & Barufaldi, J. P. (2010). The positive and negative effects of
science concept tests on student conceptual understanding. International Journal of Science
Education, 32(2), 265-282.

Chang, H. P., Chen, J. Y., Guo, C. J., Chen, C. C., Chang, C. Y., Lin, S. H., ... & Chen, C. C.
(2007). Investigating primary and secondary students’ learning of physics concepts in
Taiwan. International Journal of Science Education, 29(4), 465-482.

Chapko, M. A., & Buchko, M. (2004). Math Instruction for Inquiring Minds: Two Principals
Explain Why They Turned From Conventional Math Instruction to an Approach that Focuses
on Understanding and Applying Math Concepts. Principal, 84(2), 30-33.

Chapman, S. (1960). Misconception concerning the dynamics of the impact ball apparatus.
American Journal of Physics, 28(8), 705-711.

Chen, C. C., Lin, H. S., & Lin, M. L. (2002). Developing a two-tier diagnostic instrument to
assess high school students' understanding-the formation of images by a plane mirror.
Proceedings-National Science Council Republic of China Part D Mathematics Science and
Technology Education, 12(3), 106-121.

Chi, M. T., De Leeuw, N., Chiu, M. H., & LaVancher, C. (1994). Eliciting self-explanations
improves understanding. Cognitive Science, 18(3), 439-477.

Chu, H. E., Treagust, D. F., & Chandrasegaran, A. L. (2009). A stratified study of students’
understanding of basic optics concepts in different contexts using two‐tier multiple‐choice
items. Research in Science & Technological Education, 27(3), 253-265.

Cinyere, N. M., & Madu, B. C. (2014). Effect of analogy teaching approach on students’
conceptual change in physics. Greener Journal of Educational Research, 4(4), 119-125.

Clerk, D., & Rutherford, M. (2000). Language as a confounding variable in the diagnosis of
misconceptions. International Journal of Science Education, 22(7), 703-717.

Cohen, H. D., Hillman, D. F., & Agne, R. M. (1978). Cognitive level and college physics
achievement. American Journal of Physics, 46(10), 1026-1029.

Coleman, E. B., & Shore, B. (1991). Problem-solving processes of high and average
performers in physics. Journal for the Education of the Gifted, 14(4), 366-379.

Colin, P., Chauvet, F. O., & Viennot, L. (2002). Reading images in optics: students’
difficulties and teachers’ views. International Journal of Science Education, 24(3), 313-332.

403
Colvin, K., Champaign, J., Liu, A., Fredericks, C., & Pritchard, D. (2014, July). Comparing
Learning in a MOOC and a Blended, On-Campus Course. In Educational Data Mining 2014.

Crank, J. (1979). The mathematics of diffusion. Oxford university press.

Crooks, T. J. (1988). The impact of classroom evaluation practices on students. Review of

educational research, 58(4), 438-481.

Crotty, M. (1998). The foundations of social research: Meaning and perspective in the
research process. Sage.

Daines, D. (1986). Are teachers asking higher level questions:. Education, 106(4).

Dalaklioğlu, S., Demirci, N., & Şekercioğlu, A. (2015). Eleventh grade students’ difficulties
and misconceptions about energy and momentum concepts. International Journal of New
Trends in Education and Their Implications, 6(1), 13-21.

Darmer, M. A. (1995). Developing transfer and metacognition in educationally

disadvantaged students: Effectiveness of the Higher Order Thinking Skills (HOTS) Program
(Doctoral dissertation, University of Arizona).

de Jong, T., & Ferguson-Hessler, M. G. (1991). Knowledge of problem situations in physics:

A comparison of good and poor novice problem solvers. Learning and Instruction, 1(4), 289-
302.

Dega, B. G., & Govender, N. (2016). Assessment of students’ scientific and alternative
conceptions of energy and momentum using concentration analysis. African Journal of
Research in Mathematics, Science and Technology Education, 20(3), 201-213.

Dehaene, S., Spelke, E., Pinel, P., Stanescu, R., & Tsivkin, S. (1999). Sources of
mathematical thinking: Behavioral and brain-imaging evidence. Science, 284(5416), 970-
974.

Delialioğlu, Ö., & Aşkar, P. (1999). Contribution of students' mathematical skills and spatial
ability of achievement in secondary school physics. Hacettepe Üniversitesi Eğitim Fakültesi
Dergisi, 16(16).

DeVellis, R. F. (2016). Scale development: Theory and applications (Vol. 26). Sage
publications.

Devlin, K. (2000). Finding your inner mathematician: As the abstraction turns. The
Education Digest, 66(4), 63.

404
Ding, L. (2007). Designing an energy assessment to evaluate student understanding of
energy topics (Doctoral dissertation, NC State University).

Ding, L., Reay, N. W., Lee, A., & Bao, L. (2008). Effects of testing conditions on conceptual
survey results. Physical Review Special Topics-Physics Education Research, 4(1), 010112.

DiSessa, A. A. (1993). Toward an epistemology of physics. Cognition and instruction, 10(2-

3), 105-225.

Diyanahesa, N. E. H. (2017). Pengembangan tes diagnostik miskonsepsi materi momentum

and impuls menggunakan isomorphic problem. SKRIPSI Jurusan Fisika-Fakultas MIPA
UM.

Dole, J. A. (2000). Readers, texts and conceptual change learning. Reading & Writing
Quarterly, 16(2), 99-118.

Dreyfus, T. (2002). Advanced mathematical thinking processes. In Advanced mathematical

thinking (pp. 25-41). Springer, Dordrecht.

Dubinsky, E. (2002). Reflective abstraction in advanced mathematical thinking. In Advanced

mathematical thinking (pp. 95-126). Springer, Dordrecht.

Duman, İ., Demirci, N., & Şekercioğlu, A. G. (2015). University students’ difficulties and
misconceptions on rolling, rotational motion and torque concepts. International Journal on
New Trends in Education and Their Implications, 6(1), 46-54.

Durkee, P. (1974). An Analysis of the Appropriateness and Utilization of TOUS with Special
Reference to High-Ability Students Studying Physics. Science Education, 58(3), 343-356.

Ebenezer, J. V., & Gaskell, P. J. (1995). Relational conceptual change in solution chemistry.
Science Education, 79(1), 1-17.

Einstein, A., & Davis, F. A. (2013). The principle of relativity. Courier Corporation.

Ervynck, G. (2002). Mathematical creativity. In Advanced mathematical thinking (pp. 42-

53). Springer, Dordrecht.

Eryılmaz, A. (2010). Development and Application of Three-Tier Heat and Temperature

Test: Sample of Bachelor and Graduate Students. Eurasian Journal of Educational Research
(EJER), (40).

Esposito, M., Kawai, R., Lindenberg, K., & Van den Broeck, C. (2010). Quantum-dot Carnot
engine at maximum power. Physical review E, 81(4), 041106.

405
Euler, L., & Karsten, W. J. G. (1765). Theoria motus corporum solidorum seu rigidorum, ex
primis nostræ cognitionis principiis stabilita et ad omnes motus, qui in hujusmodi corpora
cadere possunt, accommodata.[With preface by WJG Karsten.] MS. additions.

Fariyani, Q., Rusilowati, A., & Sugianto, S. (2017). Four-Tier Diagnostic Test to Identify
Misconceptions in Geometrical Optics. Unnes Science Education Journal, 6(3).

Ferrari, M., & Elik, N. (2003). Influences on intentional conceptual change. In Intentional
conceptual change (pp. 28-62). Routledge.

Fetherstonhaugh, T., & Treagust, D. F. (1992). Students' understanding of light and its
properties: Teaching to engender conceptual change. Science Education, 76(6), 653-672.

Feynman, R. P., Leighton, R. B., & Sands, M. (2013). The Feynman Lectures on Physics
(Desktop Edition Volume I, Vol. 1). Basic Books.

Foong, Y. Y., & Lam, T. L. (1988). Development and validation of the attitudes to science
instrument. Institute of Education.

Forinash, K., Rumsey, W., & Lang, C. (2000). Galileo’s Mathematical Language of Nature.
Science & Education, 9(5), 449-457.

George, D., & Mallery, P. (2003). SPSS for Windows step by step: A simple guide and
reference. 11.0 update (4th ed.). Boston: Allyn & Bacon.

George, E. A., Broadstock, M. J., & Abad, J. V. (2013, April). Learning energy, momentum,
and conservation concepts with computer support in an undergraduate physics laboratory. In
Fourth international conference of the learning sciences (pp. 2-3).

Georghiades, P. (2004). Making pupils' conceptions of electricity more durable by means of

situated metacognition. International Journal of Science Education, 26(1), 85-99.

Ginsburg, A., Cooke, G., Leinwand, S., Noell, J., & Pollock, E. (2005). Reassessing US
International Mathematics Performance: New Findings from the 2003 TIMSS and PISA.
American Institutes for Research.

Gobert, J. D., & Clement, J. J. (1999). Effects of student‐generated diagrams versus student‐
generated summaries on conceptual understanding of causal and dynamic knowledge in plate
tectonics. Journal of Research in Science Teaching: The Official Journal of the National
Association for Research in Science Teaching, 36(1), 39-53.

Goldhaber, S., Pollock, S., Dubson, M., Beale, P., & Perkins, K. (2009, November).
Transforming upper‐division quantum mechanics: Learning goals and assessment. In AIP
Conference Proceedings (Vol. 1179, No. 1, pp. 145-148). AIP.

406
Goldman, S. R. (2003). Learning in complex domains: When and why do multiple
representations help?. Learning and instruction, 13(2), 239-244.

Goswami, U. (2004). Neuroscience and education. British Journal of Educational

Psychology, 74(1), 1-14.

Gravemeijer, K., & Doorman, M. (1999). Context problems in realistic mathematics

education: A calculus course as an example. Educational studies in mathematics, 39(1-3),
111-129.

Grayson, D. J. (1996). Concept substitution: A strategy for promoting conceptual change.

Improving teaching and learning in science and mathematics, 152-161.

Gries, D., Marion, B., Henderson, P., & Schwartz, D. (2001, February). How mathematical
thinking enchances computer science problem solving. In ACM SIGCSE Bulletin (Vol. 33,
No. 1, pp. 390-391). ACM.

Griffard, P. B., & Wandersee, J. H. (2001). The two-tier instrument on photosynthesis: What
does it diagnose?. International Journal of Science Education, 23(10), 1039-1052.

Griffith, W. T. (1985). Factors affecting performance in introductory physics courses.

American Journal of Physics, 53(9), 839-842.

Haja, S., & Clarke, D. (2011). Middle school students’ responses to two-tier tasks.
Mathematics Education Research Journal, 23(1), 67-76.

Hammer, D. (2000). Student resources for learning introductory physics. American Journal
of Physics, 68(S1), S52-S59.

Hannula, M. S. (2006). Affect in mathematical thinking and learning: Towards integration

of emotion, motivation, and cognition. In New mathematics education research and practice
(pp. 209-232). Brill Sense.

Harter, W. G. (1971). Velocity Amlification in Collision Experiments Involving Superballs.

American Journal of Physics. 39: 656 – 663.

Hasan, S., Bagayoko, D., & Kelley, E. L. (1999). Misconceptions and the certainty of
response index (CRI). Physics education, 34(5), 294.

Haslam, F., & Treagust, D. F. (1987). Diagnosing secondary students' misconceptions of

photosynthesis and respiration in plants using a two-tier multiple choice instrument. Journal
of biological education, 21(3), 203-211.

407
Haylock, D. (1997). Recognising mathematical creativity in schoolchildren. ZDM, 29(3), 68-
74.

Hein, T. L. (1999). Using writing to confront student misconceptions in physics. European

Journal of Physics, 20(3), 137.

Helm, H. (1980). Misconceptions in physics amongst South African students. Physics

Education, 15(2), 92.

Henderson, P. B., Baldwin, D., Dasigi, V., Dupras, M., Fritz, J., Ginat, D., ... & Marion Jr,
B. (2001). Striving for mathematical thinking. ACM SIGCSE Bulletin, 33(4), 114-124.

Hennessey, M. G. (1999). Probing the Dimensions of Metacognition: Implications for

Conceptual Change Teaching-Learning. Paper presented at the Annual Meeting of the
National Association for Research in Science Teaching (Boston, MA, 1999).

Hermita, N., Suhandi, A., Syaodih, E., Samsudin, A., & Rosa, F. (2017b). Assessing Pre-
Service Elementary School Teachers’ Alternative Conceptions through a Four-Tier
Diagnostic Test on Magnetism Concepts. Advanced Science Letters, 23(11), 10910-10912.

Hermita, N., Suhandi, A., Syaodih, E., Samsudin, A., Johan, H., Rosa, F., ... & Safitri, D.
(2017a, September). Constructing and implementing a four tier test about static electricity to
diagnose pre-service elementary school teacher’s misconceptions. In Journal of Physics:
Conference Series (Vol. 895, No. 1, p. 012167). IOP Publishing.

Hersh, R. (1997). What is mathematics, really?. Oxford University Press.

Hestenes, D. and Wells, M. (1992). A mechanics baseline test. The Physics Teacher 30(3):
159 – 166.

Hestenes, D., Wells, M., and Swackhamer, G. (1992). Force concept inventory. The Physics
Teacher 30(3): 141 – 158.

Hewson, P. W. (1981). A conceptual change approach to learning science. European journal

of science education, 3(4), 383-396.

Hewson, P. W., & Hewson, M. G. B. (1984). The role of conceptual conflict in conceptual
change and the design of science instruction. Instructional Science, 13(1), 1-13.

Hill, M. J. (2015). Scientific representational fluency: defining, diagnosing, and developing

(Doctoral dissertation, University of Sydney).

Hill, M., & Sharma, M. D. (2015). Students’ representational fluency at university: A cross-
sectional measure of how multiple representations are used by physics students using the

408
representational fluency survey. Eurasia Journal of Mathematics, Science and Technology
Education, 11(6), 1633-1655.

Hoe, K. Y., & Subramaniam, R. (2016). On the prevalence of alternative conceptions on

acid–base chemistry among secondary students: insights from cognitive and confidence
measures. Chemistry Education Research and Practice, 17(2), 263-282.

Holmegaard, H. T., Ulriksen, L., & Madsen, L. M. (2010). Why students choose (not) to
study engineering. In Proc. of the Joint International IGIP-SEFI Annual Conference.

Holyoak, K. J., & Morrison, R. G. (Eds.). (2005). The Cambridge handbook of thinking and
reasoning (Vol. 137). Cambridge: Cambridge University Press.

Hong, S. C., & Hong, S. I. (2013). Moments of inertia of disks and spheres without
integration. The Physics Teacher, 51, 139-140.

Howison, S. (2005). Practical applied mathematics: modelling, analysis, approximation

(No. 38). Cambridge university press.

Hrepic, Z. (2004). Development of a real-time assessment of students' mental models of

sound propagation (Doctoral dissertation, Kansas State University).

Hudson, H. T., & Liberman, D. (1982). The combined effect of mathematics skills and formal
operational reasoning on student performance in the general physics course. American
journal of Physics, 50(12), 1117-1119.

Hudson, H. T., & McIntire, W. R. (1977). Correlation between mathematical skills and
success in physics. American Journal of Physics, 45(5), 470-471.

Hudson, H. T., & Rottmann, R. M. (1981). Correlation between performance in physics and
prior mathematics knowledge. Journal of Research in Science Teaching, 18(4), 291-294.

Hwang, G. J. (2003). A conceptual map model for developing intelligent tutoring systems.
Computers & Education, 40(3), 217-235.

Hwang, G. J., Tseng, J. C., & Hwang, G. H. (2008). Diagnosing student learning problems
based on historical assessment records. Innovations in Education and Teaching
International, 45(1), 77-89.

Ingec, S. K. (2008). Use of Concept Cartoons as an Assessment Tool in Physics Education.

Online Submission, 5(11), 47-54.

409
IOP Physics World (2014). “Walter Lewin demonstrates moment of inertia.” Retrieved from
http://physicsworld.com/cws/article/multimedia/2014/mar/18/walter-lewin-demonstrates-
moment-of-inertia.

Isoda, M. (2006). Developing Mathematical Thinking in Classroom. Paper presented at the

meeting of the APEC, Tsukuba International Conference, Japan.

Jacobi, A., Martin, J., Mitchell, J., & Newell, T. (2004, October). Work in progress: A
concept inventory for heat transfer. In 34th Annual Frontiers in Education, 2004. FIE 2004.
(pp. T3F-3). IEEE.

Jacobs, V. R., Lamb, L. L., & Philipp, R. A. (2010). Professional noticing of children's
mathematical thinking. Journal for Research in Mathematics Education, 169-202.

JOHNSTONE, A. H., & Selepeng, D. (2001). A language problem revisited. Chemistry

Education Research and Practice, 2(1), 19-29.

Kadir, M. S., Lim, J., Foong, S. K., Rajendran, P., and Lee, P. C. K. (2009). PbI1@ School:
On Singapore secondary one students’ perception and understanding of work done and
moment of force. The 3rd Redesigning Pegagogy International Conference, Singapore.
Available online at http://hdl.handle.net/10497/6516.

Kaiser, M. K., Jonides, J., and Alexander, J. (1986). Intuitive reasoning about abstract and
familiar physics problems. Memory & Cognition 14(4): 308 – 312.

Kaltakçi, D., & Didiş, N. (2007, April). Identification of Pre‐Service Physics Teachers’
Misconceptions on Gravity Concept: A Study with a 3‐Tier Misconception Test. In AIP
Conference Proceedings (Vol. 899, No. 1, pp. 499-500). AIP.

Kaltakci-Gurel, D., Eryilmaz, A., & McDermott, L. C. (2016). Identifying pre-service

physics teachers’ misconceptions and conceptual difficulties about geometrical optics.
European Journal of Physics, 37(4), 045705.

Kaltakci-Gurel, D., Eryilmaz, A., & McDermott, L. C. (2017). Development and application
of a four-tier test to assess pre-service physics teachers’ misconceptions about geometrical
optics. Research in Science & Technological Education, 35(2), 238-260.

Kane, T. R., & Levinson, D. A. (1985). Dynamics, theory and applications. McGraw Hill.

Kant, I. (2004). Kant: Metaphysical foundations of natural science. Cambridge University

Press.

Karadag, Z. (2009). Analyzing students' mathematical thinking in technology-supported

environments (Doctoral dissertation, University of Toronto).

410
Kazemi, E., & Franke, M. L. (2004). Teacher learning in mathematics: Using student work
to promote collective inquiry. Journal of Mathematics Teacher Education, 7(3), 203-235.

Kerwin, J. D. (1972). Velocity, Momentum, and Energy Transmissions in Chain Collisions.

American Journal of Physics. 40: 1152 – 1157.

Khalijah, M. S., Deraman, M., Omar, R., & Othman, M. Y. H. (1995). Perception of the
Malaysian secondary-school science-students toward physics. Journal of Instructional
Psychology.

Kim, T. (2006). Impact of inquiry-based teaching on student mathematics achievement and

attitude (Doctoral dissertation, University of Cincinnati).

Kim, Y., Bao, L., & Acar, O. (2006, February). Students’ cognitive conflict and conceptual
change in a physics by inquiry class. In AIP Conference Proceedings (Vol. 818, No. 1, pp.
117-120). AIP.

Kitcher, P. (1984). The nature of mathematical knowledge. Oxford University Press on

Demand.

Kızılcık, H. S., & Güneş, B. (2011). Developing three-tier misconception test about regular
circular motion. Hacettepe University Journal of Education, 41, 278-292.

Kline, P. (2000). The handbook of psychological testing. Psychology Press.

Klüver, C., & Klüver, J. (2019, June). Using Boolean-and Self-Enforcing-Networks for
Mathematical E-Tutorial Systems. In International Work-Conference on Artificial Neural
Networks (pp. 845-856). Springer, Cham.

Kogan, M., & Laursen, S. L. (2014). Assessing long-term effects of inquiry-based learning:
A case study from college mathematics. Innovative higher education, 39(3), 183-199.

Konicek-Moran, R., & Keeley, P. (2015). Teaching for conceptual understanding in science.
NSTA Press, National Science Teachers Association.

Krockover, G., Adams, P., Eichinger, D., Nakhleh, M., & Shepardson, D. (2001). Action-
based research teams: Collaborating to improve science instruction. Journal of College
Science Teaching, 30(5), 313.

Kuo, E., Hull, M. M., Gupta, A., & Elby, A. (2013). How students blend conceptual and
formal mathematical reasoning in solving physics problems. Science Education, 97(1), 32-
57.

411
Kutluay, Y. (2005). Diagnosis of eleventh grade students’ misconceptions about geometric
optic by a three-tier test (Unpublished master’s thesis, Middle East Technical University,
Ankara).

Kwen, B. H., & Cheng, A. K. (2005). Using two-tier reflective multiple choice questions to
cater to creative thinking. In International Education Research Conference, Singapore,
AARE.

Langley, D., Ronen, M., & Eylon, B. S. (1997). Light propagation and visual patterns:
Preinstruction learners' conceptions. Journal of Research in Science Teaching, 34(4), 399-
424.

Leikin, R. (2007, February). Habits of mind associated with advanced mathematical thinking
and solution spaces of mathematical tasks. In Proceedings of the Fifth Congress of the
European Society for Research in Mathematics Education (pp. 2330-2339). Nicosia,,
Cyprus: Department of Education, University of Cyprus.

Lemos, N. A. (2008). Failure of intuition in elementary rigid body dynamics. European

Journal of Physics, 29(2), N1. http://dx.doi.org/10.1088/0143-0807/29/2/n01

Lewis, A., & Smith, D. (1993). Defining higher order thinking. Theory into practice, 32(3),
131-137.

Liberman, D., & Hudson, H. T. (1979). Correlation between logical abilities and success in
physics. American Journal of Physics, 47(9), 784-786.

Lin, S. W. (2004). Development and application of a two-tier diagnostic test for high school
students’ understanding of flowering plant growth and development. International Journal
of Science and Mathematics Education, 2(2), 175-199.

Liu, P. H., & Liu, S. Y. (2011). A cross-subject investigation of college students’

epistemological beliefs of physics and mathematics. The Asia-Pacific Education Researcher,
20(2), 336-351.

Loh, A. S. L., Subramaniam, R., & Tan, K. C. D. (2014). Exploring students’ understanding
of electrochemical cells using an enhanced two-tier diagnostic instrument. Research in
Science & Technological Education, 32(3), 229-250.

Long, C. Y., & Jiar, Y. K. (2014). Mathematical thinking and physics achievement of
secondary school students. Sains Humanika, 2(4).

Lundeberg, M. A., Fox, P. W., Brown, A. C., & Elbedour, S. (2000). Cultural influences on
confidence: Country and gender. Journal of Educational Psychology, 92(1), 152.

412
Lupart, J. L. (1995). Exceptional learners and teaching for transfer. Teaching for transfer:
Fostering generalization in learning, 215-228.

Lutfiyya, L. A. (1998). Mathematical thinking of high school students in Nebraska.

International Journal of Mathematical Education in Science and Technology, 29(1), 55-64.

Lyons, T. (2006). The puzzle of falling enrolments in physics and chemistry courses: Putting
some pieces together. Research in science education, 36(3), 285-311.

Ma, Y., Lai, G., Prejean, L., Ford, M. J., & Williams, D. (2007, March). Acquisition of
Physics Content Knowledge and Scientific Inquiry Skills in a Robotics Summer Camp.
In Society for Information Technology & Teacher Education International Conference (Vol.
2007, No. 1, pp. 3437-3444).

Ma’Moon, M. M. (2005). Mathematical Thinking and Mathematics Achievement of Students

in The Year 11 Scientific Stream in Jordan (Doctoral dissertation, University of Newcastle).

Mach, E. (1960). The Science of Mechanics-A Critical and Historical Account of its
Development. Open Court. La Salle.

Maloney, D. P., O’Kuma, T. L., Hieggelke, C. K., and Van Heuvelen, A. (2001). Surveying
students’ conceptual knowledge of electricity and magnetism. American Journal of Physics
69(7): S12 – S23.

Martínez, I. A., Roldán, É., Dinis, L., Petrov, D., Parrondo, J. M., & Rica, R. A. (2016).
Brownian carnot engine. Nature physics, 12(1), 67.

Mashood, K. K., & Singh, V. A. (2015). Rotational kinematics of a rigid body about a fixed
axis: development and analysis of an inventory. European Journal of Physics, 36(4), 045020.

Mason, L., & Boscolo, P. (2000). Writing and conceptual change. What changes?.
Instructional Science, 28(3), 199-226.

Mathworks. (2015). MATLAB, Natick, MA.

Mazens, K., & Lautrey, J. (2003). Conceptual change in physics: children’s naive
representations of sound. Cognitive Development, 18(2), 159-176.

McClary, L. M., & Bretz, S. L. (2012). Development and assessment of a diagnostic tool to
identify organic chemistry students’ alternative conceptions related to acid strength.
International Journal of Science Education, 34(15), 2317-2341.

413
McDermott, L. C. (1995). Physics by Inquiry, Volume 1. Physics by Inquiry, Volume 1, by
Lillian C. McDermott, Physics Education Group, Univ. of Washington, pp. 400. ISBN 0-471-
14440-1. Wiley-VCH, August 1995., 400.

McDermott, L. C. (2001). Oersted medal lecture 2001:“Physics Education Research—the

key to student learning”. American Journal of Physics, 69(11), 1127-1137.

McDonnell, F. (2005). Why so few choose physics: An alternative explanation for the leaky
pipeline. American Journal of Physics, 73(7), 583-586.

McDonough, J. M. (2009). Lectures in elementary fluid dynamics: Physics. Mathematics and

Applications.

McKagan, S. B., Perkins, K. K., & Wieman, C. E. (2010). Design and validation of the
quantum mechanics conceptual survey. Physical Review Special Topics-Physics Education
Research, 6(2), 020121.

McKeachie, W. J. (1986). Teaching and learning in the college classroom: A review of the
research literature (Vol. 86). University of Michigan Press.

Mehrens, W. A., & Lehmann, I. J. (1978). Measurement and evaluation in education and
psychology (pp. 594-629). New York: Holt, Rinehart and Winston.

Mellen, W. R. (1968) Superball Rebound Projectiles. American Journal of Physics 36: 865.

Meltzer, D. E. (2002). The relationship between mathematics preparation and conceptual

learning gains in physics: A possible “hidden variable” in diagnostic pretest scores. American
journal of physics, 70(12), 1259-1268.

Mestre, J. (2002). Transfer of learning: Issues and research agenda. Arlington, VA: National
Science Foundation.

Midkiff, K. C., Litzinger, T. A., & Evans, D. L. (2001). Development of engineering

thermodynamics concept inventory instruments. In 31st Annual Frontiers in Education
Conference. Impact on Engineering and Science Education. Conference Proceedings (Cat.
No. 01CH37193) (Vol. 2, pp. F2A-F23). IEEE.

Muller, D. A., Sharma, M. D., Eklund, J., & Reimann, P. (2007). Conceptual change through
vicarious learning in an authentic physics setting. Instructional Science, 35(6), 519-533.

Muller, P. and Pöschel, T. (2011). Two-ball problem revisited: Limitations of event-driven

modeling. Phys. Rev. E. 83: 041304.

414
National Research Council. (2007). Taking science to school: Learning and teaching science
in grades K-8. National Academies Press.

National Research Council. (2012). A framework for K-12 science education: Practices,
crosscutting concepts, and core ideas. National Academies Press.

Newton, I. (1846). Newton’s principia: The mathematical principles of natural philosophy

(A. Motte, Trans.). New York: Daniel Adee. (Original work published 1687).

Notaros, B. M. (2002). Concept inventory assessment instruments for electromagnetics

education. In IEEE Antennas and Propagation Society International Symposium (IEEE Cat.
No. 02CH37313) (Vol. 1, pp. 684-687). IEEE.

Nunnally, J. C. (1978). Psychometric Theory: 2d Ed. McGraw-Hill.

Olympiou, G., & Zacharia, Z. C. (2012). Blending physical and virtual manipulatives: An
effort to improve students’ conceptual understanding through science laboratory
experimentation. Science Education, 96(1), 21-47.

Oon, P. T., & Subramaniam, R. (2010). Views of physics teachers on how to address the
declining enrolment in physics at the university level. Research in Science & Technological
Education, 28(3), 277-289.

Oon, P. T., & Subramaniam, R. (2011). On the declining interest in physics among students—
from the perspective of teachers. International Journal of Science Education, 33(5), 727-746.

Ornek, F., Robinson, W. R., & Haugan, M. P. (2008). What Makes Physics Difficult?.
International Journal of Environmental and Science Education, 3(1), 30-34.

Palacios, F. J. P., Cazorla, F. N., & Madrid, A. C. (1989). Misconceptions on geometric optics
and their association with relevant educational variables. International Journal of Science
Education, 11(3), 273-286.

Pei, C., Weintrop, D., & Wilensky, U. (2018). Cultivating Computational Thinking Practices
and Mathematical Habits of Mind in Lattice Land. Mathematical Thinking and Learning,
20(1), 75-89.

Peirce, B. (1881). Linear associative algebra. American Journal of Mathematics, 4(1), 97-
229.

Peltzer, A. (1988). The intellectual factors believed by physicists to be most important to

physics students. Journal of Research in Science Teaching, 25(9), 721-731.

415
Peşman, H., & Eryılmaz, A. (2010). Development of a three-tier test to assess misconceptions
about simple electric circuits. The Journal of Educational Research, 103(3), 208-222.

Pfundt, H., & Duit, R. (1991). Bibliography: Students’ alternative frameworks and science.
Kiel, Germany: IPN.

Pintrich, P. R., & Sinatra, G. M. (2003). Future Directions for Theory and Research on
Intentional Conceptual Change: Paul R. Pintrich and Gale M. Sinatra. In Intentional
conceptual change (pp. 423-434). Routledge.

Pintrich, P. R., Marx, R. W., & Boyle, R. A. (1993). Beyond cold conceptual change: The
role of motivational beliefs and classroom contextual factors in the process of conceptual
change. Review of Educational research, 63(2), 167-199.

Podolsky, B. (1966). Conservation of Angular Momentum. American Journal of Physics,

34(1), 42-45. http://dx.doi.org/10.1119/1.1972743

Politis, Y., Killeavy, M., & Mitchell, P. I. (2007). Factors influencing the take-up of physics
within second-level education in Ireland—The teachers’ perspective. Irish Educational
Studies, 26(1), 39-55.

Poschel, T. and Brilliantov, N. V. (2001). Extremal collision sequences of particles on a line:

Optimal transmission of kinetic energy. Physical Review E 63: 021505.

Posner, G. J., Strike, K. A., Hewson, P. W., & Gertzog, W. A. (1982). Accommodation of a
scientific conception: Toward a theory of conceptual change. Science Education, 66(2), 211-
227.

Potgieter, M., Malatje, E., Gaigher, E., & Venter, E. (2010). Confidence versus performance
as an indicator of the presence of alternative conceptions and inadequate problem‐solving
skills in mechanics. International Journal of Science Education, 32(11), 1407-1429.

Prain, V., & Tytler, R. (2012). Learning through constructing representations in science: A
framework of representational construction affordances. International Journal of Science
Education, 34(17), 2751-2773.

Prince, M., Vigeant, M., & Nottis, K. (2012). Development of the heat and energy concept
inventory: Preliminary results on the prevalence and persistence of engineering students’
misconceptions. Journal of Engineering Education, 101(3), 412-438.

Ramírez, R., Pöschel, T., Brilliantov, N. V., & Schwager, T. (1999). Coefficient of restitution
of colliding viscoelastic spheres. Physical review E, 60(4), 4465.

416
Reay, N. W., Li, P., & Bao, L. (2008). Testing a new voting machine question methodology.
American Journal of Physics, 76(2), 171-178.

Rebello, N. S., Zollman, D. A., Allbaugh, A. R., Engelhardt, P. V., Gray, K. E., Hrepic, Z.,
& Itza-Ortiz, S. F. (2004). Dynamic transfer: A perspective from physics education research.
Transfer of Learning: Research and Perspectives. Greenwich: Information Age Publishing.

Redish, E. F. (2003). Teaching physics with the physics suite. Hoboken, NJ: Johns Wiley &
Sons.

Reece, G. J. (2005). Critical thinking and cognitive transfer: Implications for the
development of online information literacy tutorials. Research Strategies, 20(4), 482-493.

Reid, N., & Skryabina, E. A. (2002). Attitudes towards physics. Research in Science &
Technological Education, 20(1), 67-81.

Reid, N., & Skryabina, E. A. (2003). Gender and physics. International Journal of Science
Education, 25(4), 509-536.

Reif, F., & Allen, S. (1992). Cognition for interpreting scientific concepts: A study of
acceleration. Cognition and Instruction, 9(1), 1-44.

Renner, C. H., & Renner, M. J. (2001). But I thought I knew that: Using confidence
estimation as a debiasing technique to improve classroom performance. Applied Cognitive
Psychology, 15(1), 23-32.

Resnick, D. H., & Kenneth, S. K. (2001). Physics Volume I (5th ed., pp. 356-357). New York:
Wiley.

Resnick, L. B., & Ford, W. W. (2012). Psychology of mathematics for instruction. Routledge.

Ricardo, B. (2015). Using scaling to compute moments of inertia of symmetric objects.

European Journal of Physics, 36(5), 055003. http://dx.doi.org/10.1088/0143-
0807/36/5/055003

Ricardo, B. (2016). Comprehensive Analysis of the Failure of Intuition in Elementary Rigid

Body Dynamics. Applied Physics Research, 8(1), 125.
http://dx.doi.org/10.5539/apr.v8n1p125

Ricardo, B., & Ho, L. X. T. (2019). Distinct Feature of Inverse Square Force and Harmonic
Force: A Simplistic Proof of Bertrand’s Theorem. The Physics Educator, 1(04), 1950012.

417
Ricardo, B., & Lee, P. (2015). Maximizing kinetic energy transfer in one-dimensional many-
body collisions. European Journal of Physics, 36(2), 025013.
http://dx.doi.org/10.1088/0143-0807/36/2/025013

Richardson, G. M., & Liang, L. L. (2008). The use of inquiry in the development of
preservice teacher efficacy in mathematics and science. Journal of Elementary Science
Education, 20(1), 1-16.

Rimoldini, L. G., & Singh, C. (2005). Student understanding of rotational and rolling motion
concepts. Physical Review Special Topics-Physics Education Research, 1(1), 010102.

Rips, L. J., & Conrad, F. G. (1989). Folk psychology of mental activities. Psychological
Review, 96(2), 187.

Rizcallah, J. A. (2015). Moment of Inertia by Differentiation. The Physics Teacher, 53(8),

482.

Roberts, A. L., Sharma, M. D., Britton, S., & New, P. B. (2007). An index to measure the
ability of first year science students to transfer mathematics. International Journal of
Mathematical Education in Science and Technology, 38(4), 429-448.

Roedel, R. J., El-Ghazaly, S., Rhoads, T. R., & El-Sharawy, E. (1998, November). The wave
concepts inventory-an assessment tool for courses in electromagnetic engineering. In FIE'98.
28th Annual Frontiers in Education Conference. Moving from ‘Teacher-Centered’ to
‘Learner-Centered’ Education. Conference Proceedings (Cat. No. 98CH36214) (Vol. 2, pp.
647-653). IEEE.

Romberg, T. A. (1992). The scholarly basis of the school mathematics reform movement in
the United States. International Journal of Educational Research, 17(5), 419-437.

Ronen, M., & Eylon, B. S. (1993). To see or not to see: the eye in geometrical optics-when
and how?. Physics Education, 28(1), 52.

Rust, J., & Golombok, S. (2014). Modern psychometrics: The science of psychological
assessment. Routledge.

Saat, R. M., Fadzil, H. M., Aziz, A., Azlina, N., Haron, K., Rashid, K. A., & Shamsuar, N.
R. (2016). Development of an Online Three-Tier Diagnostic Test to Assess Pre-University
Students’ Understanding of Cellular Respiration. Journal of Baltic Science Education, 15(4).

Sadler, D. R. (1989). Formative assessment and the design of instructional systems.

Instructional science, 18(2), 119-144.

418
Sadler, P. M., & Tai, R. H. (2001). Success in introductory college physics: The role of high
school preparation. Science Education, 85(2), 111-136.

Sahin, M. (2010). The impact of problem-based learning on engineering students’ beliefs

about physics and conceptual understanding of energy and momentum. European Journal of
Engineering Education, 35(5), 519-537.

Salas-Brito, A. L., Núnez-Yépez, H. N., & Màrtinez-y-Romero, R. P. (1997).

Superintegrability in classical mechanics: A contemporary approach to Bertrand's theorem.
International Journal of Modern Physics A, 12(01), 271-276.

Sally, J. D. (2007). Roots to research: a vertical development of mathematical problems.

American Mathematical Soc..

Sam, L. C., & Yong, H. T. (2006, December). Promoting mathematical thinking in the
Malaysian classroom: Issues and challenges. In Meeting of the APEC-Tsukuba International
Conference. meeting of the APEC-Tsukuba International Conference, Japan.

Samsudin, A., Liliawati, W., Sutrisno, A. D., Suhendi, E., & Kaniawati, I. (2015, January).
The use of computer simulation in cooperative learning to minimize students’
misconceptions of momentum and impulse. In 2014 International Conference on Advances
in Education Technology (ICAET-14). Atlantis Press.

Sanger, M. J., & Greenbowe, T. J. (2000). Addressing student misconceptions concerning

electron flow in aqueous solutions with instruction including computer animations and
conceptual change strategies. International Journal of Science Education, 22(5), 521-537.

Santos, F. C., Soares, V., & Tort, A. C. (2009). Determination of the apsidal angles and
Bertrand’s theorem. Physical Review E, 79(3), 036605.

Santos, J., B.P.W.d. Oliveira, and O.R. Nelson. (2012). Impedance of rigid bodies in one-
dimensional elastic collisions. Revista Brasileira de Ensino de Fisica. 34(1): p. 1 – 4.

Santos-Trigo, M., & Camacho-Machín, M. (2009). Towards the construction of a framework

to deal with routine problems to foster mathematical inquiry. Primus, 19(3), 260-279.

Saravanan, N. S., & Ricardo, B. (2020). Moment of Inertia Computation via Analysis of
Differential Elements. The Physics Educator, 2(01), 2050002. DOI:
10.1142/S266133952050002X

Sautoy, M. (2012, December 6). A Brief History of Mathematics: 1. Newton and Leibniz.
Retrieved from https://www.bbc.co.uk/programmes/b00sr3fm.

419
Savinainen, A., & Scott, P. (2002). Using the Force Concept Inventory to monitor student
learning and to plan teaching. Physics Education, 37(1), 53.

Sawyer, A. (2014). Professional Development across Borders: The Promise of US-Mexico

Binational Teacher Education Programs. Teacher Education Quarterly, 41(4), 3-27.

Schoenfeld, A. (1991). On mathematics as sense-making: An informal attack on the

unfortunate divorce of formal and informal mathematics. In J. Voss, D. Perkins & J. Segal
(Eds.), Informal reasoning and education. Hillsdale, NJ: Lawrence Erlbaum, 311-343.

Schoenfeld, A. (2009). Learning to think mathematically: Problem solving, metacognition,

and sense-making in mathematics. Colección Digital Eudoxus, (7).

Schoenfeld, A. H. (1992). Learning to think mathematically: Problem solving,

metacognition, and sense making in mathematics. Handbook of Research on Mathematics
Teaching and Learning, 334370.

Schriesheim, C. A., & Hill, K. D. (1981). Controlling acquiescence response bias by item
reversals: The effect on questionnaire validity. Educational and Psychological Measurement,
41(4), 1101-1114.

Şekercıoğlu, A. G., & Kocakülah, M. S. (2008). Grade 10 Students' Misconceptions about

Impulse and Momentum. Journal of Turkish Science Education (TUSED), 5(2).

Seoh, K. H. (2017). Exploring grade 11-12 students’ understanding of evolution and

effecting conceptual change of selected alternative conceptions through conceptual change
strategies (Doctoral dissertation, National Institute of Education).

Sholihat, F. N., Samsudin, A., & Nugraha, M. G. (2017). Identifikasi Miskonsepsi dan
Penyebab Miskonsepsi Siswa Menggunakan Four-Tier Diagnostic Test Pada Sub-Materi
Fluida Dinamik: Azas Kontinuitas. Jurnal Penelitian & Pengembangan Pendidikan Fisika,
3(2), 175-180.

Shure, M. B. (2001). How to think, not what to think: A problem-solving approach to

prevention of early high-risk behaviors. In A. C. Bohart & D. J. Stipek (Eds.), Constructive
& destructive behavior: Implications for family, school, & society (p. 271–290). American
Psychological Association. https://doi.org/10.1037/10433-013

Singh, C., & Rosengrant, D. (2003). Multiple-choice test of energy and momentum concepts.
American Journal of Physics, 71(6), 607-617.

Smith, E. L., Blakeslee, T. D., & Anderson, C. W. (1993). Teaching strategies associated
with conceptual change learning in science. Journal of Research in Science Teaching, 30(2),
111-126.

420
Smithers, A., & Robinson, P. (2008). What can be done to increase the take-up of A-level
physics?. School Science Review, 89(328), 49.

Sohn, M. H., Goode, A., Koedinger, K. R., Stenger, V. A., Fissell, K., Carter, C. S., &
Anderson, J. R. (2004). Behavioral equivalence, but not neural equivalence—neural evidence
of alternative strategies in mathematical thinking. Nature Neuroscience, 7(11), 1193.

Sreenivasulu, B., & Subramaniam, R. (2013). University students’ understanding of chemical

thermodynamics. International Journal of Science Education, 35(4), 601-635.

Sreenivasulu, B., & Subramaniam, R. (2014). Exploring undergraduates’ understanding of

transition metals chemistry with the use of cognitive and confidence measures. Research in
Science Education, 44(6), 801-828.

Stacey, K. (2006). What is mathematical thinking and why is it ımportant. In APECTsukuba

International Conference.

Steinberg, R. N. (2012). An inquiry into science education, where the rubber meets the road.
Springer Science & Business Media.

Strauss, A., & Corbin, J. (1998). Basics of qualitative research techniques. Thousand Oaks,
CA: Sage publications.

Streib, J. F. (1989). A theorem on moments of inertia. American Journal of Physics, 57(2),

181-181.

Streveler, R. A., Miller, R. L., Santiago-Román, A. I., Nelson, M. A., Geist, M. R., & Olds,
B. M. (2011). Rigorous Methodology for Concept Inventory Development: Using the
Assessment Triangle to Develop and Test the Thermal and Transport Science Concept
Inventory (TTCI). The International Journal of Engineering Education, 27(5), 968-984.

Tall, D. (2002). The psychology of advanced mathematical thinking. In Advanced

Mathematical Thinking (pp. 3-21). Springer, Dordrecht.

Tamir, P. (1989). Some issues related to the use of justifications to multiple-choice answers.
Journal of Biological Education, 23(4), 285-292.

Tan, K. C. D., Goh, N. K., Chia, L. S., & Treagust, D. F. (2002). Development and application
of a two‐tier multiple choice diagnostic instrument to assess high school students'
understanding of inorganic chemistry qualitative analysis. Journal of Research in Science
Teaching, 39(4), 283-301.

421
Tao, P. K., & Gunstone, R. F. (1999). The process of conceptual change in force and motion
during computer‐supported physics instruction. Journal of Research in Science Teaching,
36(7), 859-882.

Taslidere, E. (2016). Development and use of a three-tier diagnostic test to assess high school
students’ misconceptions about the photoelectric effect. Research in Science &
Technological Education, 34(2), 164-186.

Team, R. C. (2015). R: A language and environment for statistical computing. R Foundation

for Statistical Computing. 2014.

Teichert, M. A., & Stacy, A. M. (2002). Promoting understanding of chemical bonding and
spontaneity through student explanation and integration of ideas. Journal of Research in
Science Teaching, 39(6), 464-496.

Tekkaya, C. (2003). Remediating high school students' misconceptions concerning diffusion

and osmosis through concept mapping and conceptual change text. Research in Science &
Technological Education, 21(1), 5-16.

Tenenbaum, R. A. (2006). Fundamentals of applied dynamics. Springer Science & Business

Media.

Thoads, T. R., & Roedel, R. J. (1999, November). The wave concept inventory-a cognitive
instrument based on Bloom's taxonomy. In FIE’99 Frontiers in Education. 29th Annual
Frontiers in Education Conference. Designing the Future of Science and Engineering
Education. Conference Proceedings (IEEE Cat. No. 99CH37011 (Vol. 3, pp. 13C1-14).
IEEE.

Thornton, C. (1997). Coefficient of restitution for collinear collisions of elastic-perfectly

plastic spheres. Journal of Applied Mechanics, 64(2), 383-386.
http://dx.doi.org/10.1115/1.2787319

Thornton, R. K., & Sokoloff, D. R. (1998). Assessing student learning of Newton’s laws:
The force and motion conceptual evaluation and the evaluation of active learning laboratory
and lecture curricula. American Journal of Physics, 66(4), 338-352.

Tikochinsky, Y. (1988). A simplified proof of Bertrand’s theorem. American Journal of

Physics, 56(12), 1073-1075.

Tobin, K. (1999). The value to science education of teachers researching their own praxis.
Research in Science Education, 29(2), 159-169.

Tomlinson, C. A. (2001). How to differentiate instruction in mixed-ability classrooms.

ASCD.

422
Treagust, D. F. (1988). Development and use of diagnostic tests to evaluate students’
misconceptions in science. International Journal of Science Education, 10(2), 159-169.

Tsai, C. C., & Chou, C. (2002). Diagnosing students' alternative conceptions in science.
Journal of Computer Assisted Learning, 18(2), 157-165.

Tseng, J. C., Chu, H. C., Hwang, G. J., & Tsai, C. C. (2008). Development of an adaptive
learning system with two sources of personalization information. Computers & Education,
51(2), 776-786.

Tsui, C. Y., & Treagust, D. (2010). Evaluating secondary students’ scientific reasoning in
genetics using a two‐tier diagnostic instrument. International Journal of Science Education,
32(8), 1073-1098.

Tuminaro, J. (2004). A cognitive framework for analyzing and describing introductory

students’ use and understanding of mathematics in physics (Doctoral dissertation, University
of Maryland).

Türker, F. (2005). Developing a three-tier test to assess high school students’ misconceptions
concerning force and motion (Unpublished master’s thesis, Middle East Technical
University).

Tyson, L. M., Venville, G. J., Harrison, A. G., & Treagust, D. F. (1997). A multidimensional
framework for interpreting conceptual change events in the classroom. Science Education,
81(4), 387-404.

Ueno, N. (1993). Reconsidering p-prims theory from the viewpoint of situated cognition.
Cognition and Instruction, 10(2-3), 239-248.

Uhden, O., Karam, R., Pietrocola, M., & Pospiech, G. (2012). Modelling mathematical
reasoning in physics education. Science & Education, 21(4), 485-506.

Ültay, E. (2012). Implementing REACT strategy in a context-based physics class: Impulse

and momentum example. Energy Education Science and Technology Part B: Social and
Educational Studies, 4(1), 233-240.

Uyangör, S. M. (2019). Investigation of the Mathematical Thinking Processes of Students in

Mathematics Education Supported with Graph Theory. Universal Journal of Educational
Research, 7(1), 1-9.

Venville, G. J., & Dawson, V. M. (2010). The impact of a classroom intervention on grade
10 students' argumentation skills, informal reasoning, and conceptual understanding of
science. Journal of Research in Science Teaching, 47(8), 952-977.

423
Version, M. A. T. L. A. B. (2015). 8.5. 0.197613 (R2015a).

Voska, K. W., & Heikkinen, H. W. (2000). Identification and analysis of student conceptions
used to solve chemical equilibrium problems. Journal of Research in Science Teaching,
37(2), 160-176.

Vosniadou, S. (1994). Capturing and modeling the process of conceptual change. Learning
and Instruction, 4(1), 45-69.

Vosniadou, S. (1999). Conceptual change research: State of the art and future directions. New
Perspectives on Conceptual Change, 3-13.

Vosniadou, S. (2002). On the nature of naive physics. In Reconsidering Conceptual Change:

Issues in Theory and Practice (pp. 61-76). Springer, Dordrecht.

Vosniadou, S. (2003). Exploring the relationships between conceptual change and intentional
learning. Intentional Conceptual Change, 377-406.

Vosniadou, S. (2007). Conceptual change and education. Human Development, 50(1), 47-54.

Wagh, S. M., & Deshpande, D. A. (2012). Essentials of physics (Vol. 1). PHI Learning Pvt.
Ltd..

Wandersee, J. H., Mintzes, J. J., & Novak, J. D. (1994). Research on alternative conceptions
in science. Handbook of Research on Science Teaching and Learning, 177, 210.

Wang, J., & Ricardo, B. (2019). Squashing Method for Moment of Inertia Calculations. The
Physics Teacher, 57(8), 551-554. https://doi.org/10.1119/1.5131123

Way, J. (2008). Using questioning to stimulate mathematical thinking. Australian Primary

Mathematics Classroom, 13(3), 22.

Wells, M., Hestenes, D., & Swackhamer, G. (1995). A modeling method for high school
physics instruction. American Journal of Physics, 63(7), 606-619.

Wenning, C. J. (2008). Dealing more effectively with alternative conceptions in science.

Journal of Physics Teacher Education Online, 5(1), 11-19.

Wergin, J. F. (1988). Basic Issues and Principles in Classroom Assessment. New Directions
for Teaching and Learning.

West, L. H., & Pines, A. L. (1985). Cognitive structure and conceptual change. Academic
Pr.

424
White, R. T., & Gunstone, R. F. (1989). Metalearning and conceptual change. International
Journal of Science Education, 11(5), 577-586.

Whitehead, A. N. (1942). Mathematics and the Good. Journal of Symbolic Logic, 7 (2):101-
101.

Williams, C., Stanisstreet, M., Spall, K., Boyes, E., & Dickson, D. (2003). Why aren't
secondary students interested in physics?. Physics Education, 38(4), 324.

Williamson, K., Willoughby, S., & Prather, E. (2013). Development of the Newtonian
gravity concept inventory. Astronomy Education Review, 12(1), 010107.

Wing, J. M. (2008). Computational thinking and thinking about computing. Philosophical

Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences,
366(1881), 3717-3725.

Wiser, M., & Amin, T. (2001). “Is heat hot?” Inducing conceptual change by integrating
everyday and scientific perspectives on thermal phenomena. Learning and Instruction, 11(4-
5), 331-355.

Witt, C., & Ulmer, J. (2010). The impact of inquiry-based learning on the academic
achievement of middle school students. In Proceeding of the 29th Annual Western Region
AAAE Research Conference (Vol. 269, p. 282).

Wittmann, M. C. (1998). Making sense of how students come to an understanding of physics:

An example from mechanical waves (Doctoral dissertation, University of Maryland).

Wollman, W., & Lawrenz, F. (1984). Identifying potential “dropouts” from college physics
classes. Journal of Research in Science Teaching, 21(4), 385-390.

Woolnough, B. E. (1993). Teachers' perception of reasons students choose for, or against,

science and engineering. School Science Review, 75, 112-112.

Woolnough, B. E. (1994). Why students choose physics, or reject it. Physics Education,
29(6), 368.

Yan, Y. K., & Subramaniam, R. (2018). Using a multi-tier diagnostic test to explore the
nature of students’ alternative conceptions on reaction kinetics. Chemistry Education
Research and Practice, 19(1), 213-226.

Yang, D. C., & Lin, Y. C. (2015). Assessing 10-to 11-year-old children’s performance and
misconceptions in number sense using a four-tier diagnostic test. Educational Research,
57(4), 368-388.

425
Yeo, S., & Zadnik, M. (2001). Introductory thermal concept evaluation: Assessing students’
understanding. The Physics Teacher, 39(8), 496-504.

Young, H. D., Freedman, R. A., and A.L. Ford (2012) University Physics with Modern
Physics – 13th ed. Addison-Wesley. USA.

Yürük, N. (2007). A Case Study of One Student's Metaconceptual Processes and the Changes
in Her Alternative Conceptions of Force and Motion. Eurasia Journal of Mathematics,
Science & Technology Education, 3(4).

Zacharia, Z. C. (2007). Comparing and combining real and virtual experimentation: an effort
to enhance students' conceptual understanding of electric circuits. Journal of Computer
Assisted Learning, 23(2), 120-132.

Zhu, G., & Singh, C. (2012). Surveying students’ understanding of quantum mechanics in
one spatial dimension. American Journal of Physics, 80(3), 252-259.

Ziegler, A., Heller, K. A., & Broome, P. (1996). Motivational preconditions for girls gifted
and highly gifted in physics. High Ability Studies, 7(2), 129-143.

426
APPENDIX 1: VALIDATION OF CHAIN-COLLISION ASSESSMENT

Validation Checklist and Comments from Validator 1

427
428
429
430
431
432
433
434
435
Validation Checklist and Comments from Validator 2

436
437
438
439
440
441
442
443
444
445
446
447
448
449
Validation Checklist and Comments from Validator 3

450
451
452
453
454
455
456
457
458
459
460
461
APPENDIX 2: VALIDATION OF RIGID BODY COLLISION ASSESSMENT

Validation Checklist and Comments from Validator 1

462
463
464
465
466
467
468
Validation Checklist and Comments from Validator 2

469
470
471
472
473
474
475
476
477
478
479
480
Validation Checklist and Comments from Validator 3

481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
 APPENDIX 3: CHAIN-COLLISION ASSESSMENT

Name :

Gender : Male / Female

Class :

Date :

ASSESSMENT

on

MOMENTUM AND COLLISION

INSTRUCTIONS:

• Choose only one correct answer and one correct reason for each question.

• Answer every single question and circle the confidence level for each
answer and reason. If you don’t agree with any of the options in the
reason tier, provide your own reason under “other reason”.

• Provide mathematical workings in relation to your responses for each

question in the space provided, and use texts for
clarifications/explanations.

This assessment consists of 33 pages and 15 questions.

499
CONSIDER THE FOLLOWING SCENARIO FOR
QUESTIONS (1)-(4)
Consider the following one-dimensional collision. A particle of mass 𝑚1 moves
to the right with velocity 𝑣1. It then collides elastically with another particle of
mass 𝑚2 that moves to the right with velocity 𝑣2 . After the collision, the
velocity of the first particle is 𝑣1′ and the velocity of the second particle is 𝑣2′ .

Before collision:

𝑣1 𝑣2
𝑚1 𝑚2

After collision:

𝑣1′ 𝑣2′
𝑚1 𝑚2

Figure 1

500
Question 1 (True/False)
The total momentum of the system of two particles is conserved.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision is elastic.
(B) The two particles are constrained to move in one dimension.
(C) No external force is acting on the system.
(D) Both particles do not experience any acceleration during the collision.
(E) Momentum is always conserved in any case.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
As the system of two particles is isolated, no external force is acting on the system, and the
system’s momentum is conserved.

Mathematically,
⃗⃗
∆𝐩
∑ 𝐅⃗ext =
∆𝑡
⃗⃗, 𝐩
As ∑ 𝐅⃗ext = 𝟎 ⃗⃗ = constant.

501
Question 2 (True/False)
The total kinetic energy of the system of two particles is conserved.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision is elastic.
(B) The two particles are constrained to move in one dimension.
(C) No external force is acting on the system.
(D) Both particles do not experience any acceleration during the collision.
(E) Kinetic energy is always conserved in any case.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
Elastic collision is defined as a collision that preserves the kinetic energy of the system.
1 1 1 2 1 2
𝑚1 𝑣1 2 + 𝑚2 𝑣2 2 = 𝑚1 𝑣1′ + 𝑚2 𝑣2′
2 2 2 2

502
Question 3 (True/False)
𝑣1 > 𝑣2 and 𝑣1′ < 𝑣2′ .

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The particle with larger mass has larger momentum, and vice versa.
(B) The two particles are constrained to move in one dimension.
(C) The kinetic energy of the system of two particles decreases as a result of collision.
(D) There is a collision, and the collision is elastic.
(E) The total momentum of the system of two particles is zero.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
For a collision to happen, 𝑣1 must be bigger than 𝑣2 . And the condition after the collision has
to be 𝑣1′ < 𝑣2′ since for an elastic collision, the relative velocity of approach equals to the
relative velocity of separation.

Mathematically, for an elastic collision,

𝑣1 − 𝑣2 = 𝑣2′ − 𝑣1′

503
Question 4 (True/False)
As we move together with the centre of mass of two particles (i.e. being in the frame of their
centre of mass), we will observe that the velocities of two particles are in opposite directions
to each other before the collision, and the particle with larger mass has larger momentum
than the particle with smaller mass.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The particle with larger mass has larger momentum, and vice versa.
(B) The two particles are constrained to move in one dimension.
(C) The kinetic energy of the system of two particles decreases as a result of collision.
(D) There is a collision, and the collision is elastic.
(E) The total momentum of the system of two particles is zero.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
The centre of mass frame of a system is defined as the frame where the total momentum is
zero. For a system with only two masses, their momentum vectors must add up to zero, i.e.
their velocity vectors must point to the opposite direction.

Mathematically,
⃗⃗1 + 𝐩
𝐩 ⃗⃗
⃗⃗2 = 𝟎
and the two momenta have equal magnitude.

504
CONSIDER THE FOLLOWING SCENARIO FOR
QUESTIONS (5)-(6)
Consider the following one-dimensional collision. Block A, of mass 6.0 kg,
moves with velocity 3.0 m/s to the right. It hits Block B, of mass 2.0 kg, which
is initially at rest. After the collision, Block A moves with velocity 2.0 m/s to
the right. The friction between the blocks and the floor can be ignored.

Before collision:

6.0 kg 2.0 kg
3.0 m/s
A B

After collision:

6.0 kg 2.0 kg
2.0 m/s 𝑣𝐵′
A B

Figure 2

505
Question 5
Which one of the following statements about the system’s momentum is correct?

Answer:
(A) During the collision, the momentum of Block A is partially transferred to Block B
and is also partially lost to the surroundings.
(B) During the collision, the momentum of Block A is partially transferred to Block B
while the total momentum of the system remains conserved.
(C) During the collision, there is a momentum transfer from Block A to Block B while
the total momentum of the system remains conserved, and the velocity gain of Block
B is lower than the velocity reduction of Block A.
(D) During the collision, it is possible to transfer all of the momentum of Block A to
Block B if Block A stops moving after the collision and no momentum is lost to the
surroundings.
(E) The collision described in Figure 2 is impossible.

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision in Figure 2 is elastic.
(B) The inelastic collision in Figure 2 is possible, and both blocks are in motion after the
collision.
(C) Both blocks are constrained to move in one dimension.
(D) Momentum is always conserved in any case.
(E) Momentum cannot be transferred from one block to another block.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

506
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
As the system of Blocks A and B is isolated, its total momentum is conserved.
(6.0)(3.0) + (2.0)(0.0) = (6.0)(2.0) + (2.0)𝑣𝐵′ → 𝑣𝐵′ = 3.0 m/s to the right
Hence, the velocity gain of Block B is higher than the velocity reduction of Block A.

The case is possible as the total energy of the system after the collision is lower than it is
before the collision, i.e.
1 1 1 1
(6.0)(2.0)2 + (2.0)(3.0)2 = 21 J < (6.0)(3.0)2 + (2.0)(0.0)2 = 27 J
2 2 2 2
or equivalently, the relative speed of separation is less than the relative speed of approach,
i.e.
3.0 − 2.0 = 1.0 m/s < 3.0 − 0.0 = 3.0 m/s
which implies that the system is losing energy during the collision as we see it in the centre
of mass frame.

If Block A transfers all of its momentum to block B,

(6.0)(3.0) + (2.0)(0.0) = (6.0)(0.0) + (2.0)𝑣𝐵′ → 𝑣𝐵′ = 9.0 m/s to the right
The case is impossible as the total energy of the system after the collision is higher than it is
before the collision, i.e.
1 1 1 1
(6.0)(0.0)2 + (2.0)(9.0)2 = 81 J > (6.0)(3.0)2 + (2.0)(0.0)2 = 27 J
2 2 2 2
or equivalently, the relative speed of separation is less than the relative speed of approach,
i.e.
9.0 − 0.0 = 9.0 m/s > 3.0 − 0.0 = 3.0 m/s
which implies that the system is gaining energy during the collision as we see it in the centre
of mass frame.

507
Question 6
Which one of the following statements about the system’s kinetic energy is correct?

Answer:
(A) During the collision, the kinetic energy of Block A is partially transferred to Block B
and is also partially lost to the surroundings.
(B) During the collision, the kinetic energy of Block A is partially transferred to Block B
while the total kinetic energy of the system remains conserved.
(C) During the collision, there is a kinetic energy transfer from Block A to Block B while
the total kinetic energy of the system remains conserved, and the velocity gain of
Block B is lower than the velocity reduction of Block A.
(D) During the collision, it is possible to transfer all of the kinetic energy of Block A to
Block B if Block A stops moving after the collision and no kinetic energy is lost to
the surroundings.
(E) The collision described in Figure 2 is impossible.

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision in Figure 2 is elastic.
(B) The inelastic collision in Figure 2 is possible, and both blocks are in motion after the
collision.
(C) Both blocks are constrained to move in one dimension.
(D) Kinetic energy is always conserved in any case.
(E) Kinetic energy cannot be transferred from one block to another block.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

508
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
From question 5, 𝑣𝐵′ = 3.0 m/s to the right.

The case is possible to happen as the total energy of the system after the collision is lower
than it is before the collision, i.e.
1 1 1 1
(6.0)(2.0)2 + (2.0)(3.0)2 = 21 J < (6.0)(3.0)2 + (2.0)(0.0)2 = 27 J
2 2 2 2
Hence, the energy is partially transferred from Block A to Block B and is also partially lost
to the surrounding. This is also implied by the fact that the relative speed of separation is less
than the relative speed of approach, i.e.
3.0 − 2.0 = 1.0 m/s < 3.0 − 0.0 = 3.0 m/s

If there is no energy loss, i.e. elastic collision, it is impossible for Block A to stop moving,
unless both blocks have identical mass.

509
CONSIDER THE FOLLOWING SCENARIO FOR
QUESTIONS (7)-(8)
Consider the following one-dimensional collision. Block C, of mass 6.0 kg,
moves with velocity 3.0 m/s to the right. It hits Block D, of mass 2.0 kg, which
is initially at rest. After the collision, Block C moves with velocity 1.4 m/s to
the right. The friction between the blocks and the floor can be ignored.

Before collision:

6.0 kg 2.0 kg
3.0 m/s
C D

After collision:

6.0 kg 2.0 kg
1.4 m/s 𝑣𝐷′
C D

Figure 3

510
Question 7
Which one of the following statements about the system’s momentum is correct?

Answer:
(A) During the collision, the momentum of Block C is partially transferred to Block D
and is also partially lost to the surroundings.
(B) During the collision, the momentum of Block C is partially transferred to Block D
while the total momentum of the system remains conserved.
(C) During the collision, there is a momentum transfer from Block C to Block D while
the total momentum of the system remains conserved, and the velocity gain of Block
D is higher than the velocity reduction of Block C.
(D) During the collision, it is possible to transfer all of the momentum of Block C to
Block D if Block C stops moving after the collision and no momentum is lost to the
surroundings.
(E) The collision described in Figure 3 is impossible.

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision in Figure 3 is elastic.
(B) The inelastic collision in Figure 3 is possible, and both blocks are in motion after the
collision.
(C) Both blocks are constrained to move in one dimension.
(D) Momentum is always conserved in any case.
(E) Momentum cannot be transferred from one block to another block.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

511
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
As the system of Blocks C and D is isolated, its total momentum should be conserved.
(6.0)(3.0) + (2.0)(0.0) = (6.0)(1.4) + (2.0)𝑣𝐷′ → 𝑣𝐷′ = 4.8 m/s to the right

However, the case is impossible as the total energy of the system after the collision is higher
than it is before the collision, i.e.
1 1 1 1
(6.0)(1.4)2 + (2.0)(4.8)2 = 29 J > (6.0)(3.0)2 + (2.0)(0.0)2 = 27 J
2 2 2 2
Equivalently, the relative speed of separation is higher than the relative speed of approach,
i.e.
4.8 − 1.4 = 3.4 m/s > 3.0 − 0.0 = 3.0 m/s
which implies that the system gains energy during the collision as we see it in the centre of
mass frame.

If Block C transfers all of its momentum to Block D,

(6.0)(3.0) + (2.0)(0.0) = (6.0)(0.0) + (2.0)𝑣𝐷′ → 𝑣𝐷′ = 9.0 m/s to the right
The case is impossible as the total energy of the system after the collision is higher than it is
before the collision, i.e.
1 1 1 1
(6.0)(0.0)2 + (2.0)(9.0)2 = 81 J > (6.0)(3.0)2 + (2.0)(0.0)2 = 27 J
2 2 2 2
Equivalently, the relative speed of separation is higher than the relative speed of approach,
i.e.
9.0 − 0.0 = 9.0 m/s > 3.0 − 0.0 = 3.0 m/s

512
Question 8
Which one of the following statements about the system’s kinetic energy is correct?

Answer:
(A) During the collision, the kinetic energy of Block C is partially transferred to Block D
and is also partially lost to the surroundings.
(B) During the collision, the kinetic energy of Block C is partially transferred to Block D
while the total kinetic energy of the system remains conserved.
(C) During the collision, there is a kinetic energy transfer from Block C to Block D while
the total kinetic energy of the system remains conserved, and the velocity gain of
Block D is higher than the velocity reduction of Block C.
(D) During the collision, it is possible to transfer all of the kinetic energy of Block C to
Block D if Block C stops moving after the collision and no kinetic energy is lost to
the surroundings.
(E) The collision described in Figure 3 is impossible.

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision in Figure 3 is elastic.
(B) The inelastic collision in Figure 3 is possible, and both blocks are in motion after the
collision.
(C) Both blocks are constrained to move in one dimension.
(D) Kinetic energy is always conserved in any case.
(E) Kinetic energy cannot be transferred from one block to another block.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

513
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
From question 7, 𝑣𝐷′ = 4.8 m/s to the right.

However, the case is impossible as the total energy of the system after the collision is higher
than it is before the collision, i.e.
1 1 1 1
(6.0)(1.4)2 + (2.0)(4.8)2 = 29 J > (6.0)(3.0)2 + (2.0)(0.0)2 = 27 J
2 2 2 2
Equivalently, the relative speed of separation is higher than the relative speed of approach,
i.e.
4.8 − 1.4 = 3.4 m/s > 3.0 − 0.0 = 3.0 m/s
which implies that the system gains energy during the collision as we see it in the centre of
mass frame.

If there is no energy loss, i.e. elastic collision, it is impossible for Block C to stop moving,
unless both blocks have identical mass.

514
CONSIDER THE FOLLOWING SCENARIO FOR
QUESTIONS (9)-(10)
Consider the following one-dimensional collision. Block E, of mass 6.0 kg,
moves with velocity 3.0 m/s to the right. It hits Block F, of mass 9.0 kg, which
is initially at rest. After the collision, Block E stops moving. The friction
between the blocks and the floor can be ignored.

Before collision:

6.0 kg 9.0 kg
3.0 m/s
E F

After collision:

6.0 kg 9.0 kg
𝑣𝐹′
E F

Figure 4

515
Question 9
Which one of the following statements about the system’s momentum is correct?

Answer:
(A) During the collision, the momentum of Block E is partially transferred to Block F and
is also partially lost to the surroundings.
(B) During the collision, the momentum of Block E is partially transferred to Block F
while the total momentum of the system remains conserved.
(C) During the collision, there is a momentum transfer from Block E to Block F while the
total momentum of the system remains conserved, and the velocity gain of Block F
is higher than the velocity reduction of Block E.
(D) During the collision, all of the momentum of Block E is transferred to Block F.
(E) The collision described in Figure 4 is impossible.

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision in Figure 4 is elastic.
(B) The inelastic collision in Figure 4 is possible, and block E comes to rest after the
collision.
(C) Both blocks are constrained to move in one dimension.
(D) Momentum is always conserved in any case.
(E) Momentum cannot be transferred from one block to another block.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

516
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
As the system of Blocks E and F is isolated, its total momentum should be conserved.
(6.0)(3.0) + (9.0)(0.0) = (6.0)(0.0) + (9.0)𝑣𝐹′ → 𝑣𝐹′ = 2.0 m/s to the right
Hence, the velocity gain of Block F is lower than the velocity reduction of Block E.

Since Block E stops moving after the collision, 100% of its momentum is transferred to Block
F.

The case is possible as the total energy of the system after the collision is lower than it is
before the collision, i.e.
1 1 1 1
(6.0)(0.0)2 + (9.0)(2.0)2 = 18 J < (6.0)(3.0)2 + (9.0)(0.0)2 = 27 J
2 2 2 2
This is also implied by the fact that the relative speed of separation is less than the relative
speed of approach, i.e.
2.0 − 0.0 = 2.0 m/s < 3.0 − 0.0 = 3.0 m/s
which implies that the system loses energy during the collision as we see it from the centre
of mass frame.

517
Question 10
Which one of the following statements about the system’s kinetic energy is correct?

Answer:
(A) During the collision, the kinetic energy of Block E is partially transferred to Block F
and is also partially lost to the surroundings.
(B) During the collision, the kinetic energy of Block E is partially transferred to Block F
while the total kinetic energy of the system remains conserved.
(C) During the collision, there is a kinetic energy transfer from Block E to Block F while
the total kinetic energy of the system remains conserved, and the velocity gain of
Block F is higher than the velocity reduction of Block E.
(D) During the collision, all of the kinetic energy of Block E is transferred to Block F.
(E) The collision described in Figure 4 is impossible.

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision in Figure 4 is elastic.
(B) The inelastic collision in Figure 4 is possible, and block E comes to rest after the
collision.
(C) Both blocks are constrained to move in one dimension.
(D) Kinetic energy is always conserved in any case.
(E) Kinetic energy cannot be transferred from one block to another block.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

518
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
From question 9, 𝑣𝐹′ = 2.0 m/s to the right.

The case is possible as the total energy of the system after the collision is lower than it is
before the collision, i.e.
1 1 1 1
(6.0)(0.0)2 + (9.0)(2.0)2 = 18 J < (6.0)(3.0)2 + (9.0)(0.0)2 = 27 J
2 2 2 2
Hence, the energy is partially transferred from Block E to Block F and is also partially lost
to the surrounding as the final energy of the system is less than the initial energy of the
system. This is also implied by the fact that the relative speed of separation is less than the
relative speed of approach, i.e.
2.0 − 0.0 = 2.0 m/s < 3.0 − 0.0 = 3.0 m/s
which implies that the system loses energy during the collision as we see it from the centre
of mass frame.

519
CONSIDER THE FOLLOWING SCENARIO FOR
QUESTIONS (11)-(12)
Consider the following one-dimensional collision. Block G, of mass 4.0 kg,
moves with velocity 3.0 m/s to the right. It hits Block H, of mass 𝑚𝐻 , which is
initially at rest. After the collision, Block G moves with velocity 1.0 m/s to the
right, and Block H moves with velocity 4.0 m/s to the right. The friction
between the blocks and the floor can be ignored.

Before collision:

4.0 kg 𝑚𝐻
3.0 m/s
G H

After collision:

4.0 kg 𝑚𝐻
1.0 m/s 4.0 m/s
G H

Figure 5

520
Question 11
Which one of the following statements about the system’s momentum is correct?

Answer:
(A) During the collision, the momentum of Block G is partially transferred to Block H
and is also partially lost to the surroundings.
(B) During the collision, the momentum of Block G is partially transferred to Block H
while the total momentum of the system remains conserved.
(C) During the collision, it is possible to transfer all of the momentum of Block G to
Block H if no energy is lost to the surroundings.
(D) The momentum transfer described by the collision in Figure 5 is possible regardless
of the mass of Block H.
(E) The collision described in Figure 5 is impossible.

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision in Figure 5 is inelastic.
(B) The collision in Figure 5 is elastic, and both blocks are in motion after the collision.
(C) Both blocks are constrained to move in one dimension.
(D) Momentum is always conserved in any case.
(E) Momentum cannot be transferred from one block to another block.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

521
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
As the system of Blocks G and H is isolated, its total momentum is conserved.
(4.0)(3.0) + 0.0 = (4.0)(1.0) + 𝑚𝐻 (4.0) → 𝑚𝐻 = 2.0 kg

The case is possible as the total energy of the system after the collision is equal to the total
energy of the system before the collision, defining an elastic collision.
1 1 1 1
(4.0)(1.0)2 + (2.0)(4.0)2 = (4.0)(3.0)2 + (2.0)(0.0)2 = 18 J
2 2 2 2
Equivalently, the relative speed of separation is equal to the relative speed of approach, i.e.
4.0 − 1.0 = 3.0 − 0.0 = 3.0 m/s
which implies that no energy is lost during the collision as we see it from the centre of mass
frame.

522
Question 12
Which one of the following statements about the system’s kinetic energy is correct?

Answer:
(A) During the collision, the kinetic energy of Block G is partially transferred to Block H
and is also partially lost to the surroundings.
(B) During the collision, the kinetic energy of Block G is partially transferred to Block H
while the total kinetic energy of the system remains conserved.
(C) During the collision, it is possible to transfer all of the kinetic energy of Block G to
Block H if no energy is lost to the surroundings.
(D) The kinetic energy transfer described by the collision in Figure 5 is possible
regardless of the mass of Block H.
(E) The collision described in Figure 5 is impossible.

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision in Figure 5 is inelastic.
(B) The collision in Figure 5 is elastic, and both blocks are in motion after the collision.
(C) Both blocks are constrained to move in one dimension.
(D) Kinetic energy is always conserved in any case.
(E) Kinetic energy cannot be transferred from one block to another block.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

523
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
From question 11, 𝑚𝐻 = 2.0 kg.

The case is possible as the total energy of the system after the collision is equal to the total
energy of the system before the collision, defining an elastic collision.
1 1 1 1
(4.0)(1.0)2 + (2.0)(4.0)2 = (4.0)(3.0)2 + (2.0)(0.0)2 = 18 J
2 2 2 2
Equivalently, the relative speed of separation is equal to the relative speed of approach, i.e.
4.0 − 1.0 = 3.0 − 0.0 = 3.0 m/s

In this elastic collision, the energy is conserved and is not lost to the surroundings.

524
CONSIDER THE FOLLOWING SCENARIO FOR
QUESTIONS (13)-(15)
Consider the following one-dimensional collision. Block I, of mass 4.0 kg,
moves with velocity 9.0 m/s to the right. It hits Block J, of mass 9.0 kg, which
is initially at rest. It is desired to transfer the initial kinetic energy of Block I as
much as possible to Block J. The friction between the blocks and the floor can
be ignored.

Before collision:

4.0 kg 9.0 kg
9.0 m/s J
I

After collision:

4.0 kg 9.0 kg
𝑣𝐼′ 𝑣𝐽′
I J

Figure 6

525
Question 13
Which one of the following statements about the system’s momentum is correct?

Answer:
(A) During the collision, the momentum of Block I is partially transferred to Block J and
is also partially lost to the surroundings.
(B) During the collision, the momentum of Block I is partially transferred to Block J while
the total momentum of the system remains conserved.
(C) During the collision, all of the momentum of Block I is transferred to Block J while
the total momentum of the system remains conserved.
(D) During the collision, there is a momentum transfer from Block I to Block J while the
total momentum of the system remains conserved. Block I and Block J move together
after the collision.
(E) The collision described in Figure 6 is impossible.

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision in Figure 6 is elastic.
(B) The inelastic collision in Figure 6 is possible, and block I comes to rest after the
collision.
(C) Both blocks are constrained to move in one dimension.
(D) Momentum is always conserved in any case.
(E) Momentum cannot be transferred from one block to another block.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

526
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
As the system of Blocks I and J is isolated, its total momentum is conserved.
(4.0)(9.0) + (9.0)(0.0) = (4.0)𝑣𝐼′ + (9.0)𝑣𝐽′ → 4.0𝑣𝐼′ + 9.0𝑣𝐽′ = 36

The word “as much as possible” implies conservation of energy (elastic collision):
1 1 1 1
(4.0)(9.0)2 + (9.0)(0.0)2 = (4.0)𝑣𝐼′ 2 + (9.0)𝑣𝐽′ 2
2 2 2 2
or equivalently, the relative speed is preserved,
𝑣𝐽′ − 𝑣𝐼′ = 9.0
Solving,
45
𝑣𝐼′ = − m/s = −3.5 m/s
13
72
𝑣𝐽′ = m/s = 5.5 m/s
13

527
Question 14
Which one of the following statements about the system’s kinetic energy is correct?

Answer:
(A) During the collision, the kinetic energy of Block I is partially transferred to Block J
and is also partially lost to the surrounding.
(B) During the collision, the kinetic energy of Block I is partially transferred to Block J
while the total energy of the system remains conserved.
(C) During the collision, there is a kinetic energy transfer from Block I to Block J while
the total kinetic energy of the system remains conserved. Block I and Block J move
together after the collision.
(D) During the collision, all of the kinetic energy of Block I is transferred to Block J.
(E) The collision described in Figure 6 is impossible.

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision in Figure 6 is elastic.
(B) The inelastic collision in Figure 6 is possible, and block I comes to rest after the
collision.
(C) Both blocks are constrained to move in one dimension.
(D) Kinetic energy is always conserved in any case.
(E) Kinetic energy cannot be transferred from one block to another block.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

528
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
The word “as much as possible” implies conservation of energy (elastic collision). As Block
I still moves after the collision, energy of Block I is only transferred partially.

Mathematically,
(4.0)(9.0) + (9.0)(0.0) = (4.0)𝑣𝐼′ + (9.0)𝑣𝐽′ → 4.0𝑣𝐼′ + 9.0𝑣𝐽′ = 36
1 1 1 1
(4.0)(9.0)2 + (9.0)(0.0)2 = (4.0)𝑣𝐼′ 2 + (9.0)𝑣𝐽′ 2
2 2 2 2
or equivalently,
𝑣𝐽′ − 𝑣𝐼′ = 9.0
Solving,
45 72
𝑣𝐼′ = − m/s = −3.5 m/s, 𝑣𝐽′ = m/s = 5.5 m/s
13 13

529
Question 15 (True/False)
Another block (Block K) of mass 𝑚𝐾 , which is also at rest, is inserted between Blocks I and
J. So in this scenario, Block I will hit Block K transferring as much energy as possible, and
subsequently Block K will hit Block J transferring as much energy as possible. The friction
between the blocks and the floor can be ignored. Refer to Figure 7.

The amount of kinetic energy transferred from Block I to Block J may increase.

4.0 kg 𝑚𝐾 9.0 kg
9.0 m/s
I K J

Figure 7

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The amount of energy transferred cannot exceed 100%.
(B) Involving more masses implies more energy is lost to the surroundings.
(C) The amount of energy transferred depends on the mass of Block K.
(D) Kinetic energy is always conserved in any case.
(E) Kinetic energy cannot be transferred from one block to another block.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

530
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
Using conservation of momentum for the collision between Blocks I and K,
(4.0)(9.0) + 𝑚𝐾 (0.0) = (4.0)𝑣𝐼′ + 𝑚𝐾 𝑣𝐾′
4.0𝑣𝐼′ + 𝑚𝐾 𝑣𝐾′ = 36

Using conservation of energy for the collision between Blocks I and K,

𝑣𝐾′ − 𝑣𝐼′ = 9.0

Solving,
72
𝑣𝐾′ =
𝑚𝐾 + 4.0

Using conservation of momentum for the collision between Blocks K and J,

72
𝑚𝐾 + (9.0)(0.0) = 𝑚𝐾 𝑣𝐾′′ + 9.0𝑣𝐽′
𝑚𝐾 + 4.0
72𝑚𝐾
𝑚𝐾 𝑣𝐾′′ + 9.0𝑣𝐽′ =
𝑚𝐾 + 4.0

Using conservation of energy for the collision between Blocks K and J,

72
𝑣𝐽′ − 𝑣𝐾′′ =
𝑚𝐾 + 4.0

Solving,
144𝑚𝐾
𝑣𝐽′ =
(𝑚𝐾 + 4.0)(𝑚𝐾 + 9.0)

Plugging in any value for 𝑚𝐾 within the range of 4.0 < 𝑚𝐾 < 9.0 yields a value for 𝑣𝐽′ that
is bigger than 5.5 m/s. Hence, it is true.

531
 APPENDIX 4: RIGID BODY COLLISION ASSESSMENT

Name :

Gender : Male / Female

Class :

Date :

ASSESSMENT

on

ROTATIONAL MECHANICS

INSTRUCTIONS:

• Choose only one correct answer and one correct reason for each question.

• Answer every single question and circle the confidence level for each
answer and reason. If you don’t agree with any of the options in the
reason tier, provide your own reason under “other reason”.

• Provide mathematical workings in relation to your responses for each

question in the space provided, and use texts for
clarifications/explanations.

This assessment consists of 28 pages and 16 questions.

532
CONSIDER THE FOLLOWING SCENARIO FOR
QUESTIONS (1)-(3)
Consider a uniform rigid rod of mass 𝑀 and length ℓ that lies on a smooth
horizontal table. Initially, the rod is at rest. A constant force 𝐹 is applied
perpendicularly at one end of the rod along the surface of the table. The moment
1
of inertia of the rod about its centre is 𝑀ℓ2 .
12

Top view:

Figure 1

533
Question 1 (True/False)
Subsequently, the centre of the rod will move and the rod will also rotate about its centre.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) Force 𝐹 acts as an external force and external torque on the rod about its centre.
(B) The rod is a rigid body.
(C) The rod exerts a reaction force in the opposite direction of force 𝐹.
(D) The centre of the rod will remain stationary as the force does not act at the rod’s
centre.
(E) The force is not sufficient to move the rod’s centre.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
The force acts as an external force on the rod, hence the centre of mass of the rod will
accelerate. At the same time, the force also exerts a torque as it acts perpendicularly at one
end of the rod, hence the rod will rotate about its centre.

Mathematically,
𝐹
𝐹 = 𝑚𝑎 → 𝑎 =
𝑚
ℓ 1 6𝐹
𝐹 = 𝑚ℓ2 𝛼 → 𝛼 =
2 12 𝑚ℓ

534
Question 2 (True/False)
If the force acts perpendicularly at some other point other than the rod’s end, it is more
difficult to rotate the rod.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) How difficult it is to rotate a rod depends on the length of the rod.
(B) How difficult it is to rotate a rod depends on the length and the mass of the rod.
(C) The centre of the rod will remain stationary as the force does not act at the rod’s
centre.
(D) The force produces a smaller amount of torque.
(E) Angular momentum is conserved.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
The magnitude of the torque is obtained by multiplying the force and the perpendicular
distance from the centre of rotation, which is the centre of the rod in this case. If the force
acts at some other point other than the rod’s end, the amount of torque will be smaller.

Mathematically,
𝐹
𝐹 = 𝑚𝑎 → 𝑎 =
𝑚
1 2
12𝐹𝑥
𝐹𝑥 = 𝑚ℓ 𝛼 → 𝛼 =
12 𝑚ℓ2
Smaller 𝑥 implies smaller 𝛼.

535
Question 3 (True/False)
The length of the rod is now doubled while its mass is kept constant. The force 𝐹 still acts
perpendicularly at the rod’s end. The rod will now rotate with a larger angular acceleration.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) There is now a larger amount of torque acting on the rod.
(B) Rotating a rod that is twice as long is more difficult while the torque remains
unchanged.
(C) The angular acceleration is independent of the length of the rod.
(D) Rotating a rod that is twice as long is more difficult though the amount of torque
increases.
(E) The centre of the rod accelerates slower as the rod is twice as long.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
Applying Newton’s law for rotation for the first rod,
1 12𝐹
𝐹ℓ = 𝑚ℓ2 𝛼 → 𝛼 =
12 𝑚ℓ
and for the second rod,
1 6𝐹
𝐹2ℓ = 𝑚(2ℓ)2 𝛼 → 𝛼 =
12 𝑚ℓ

536
CONSIDER THE FOLLOWING SCENARIO FOR
QUESTIONS (4)-(7)
Consider the following collision. A particle of mass 𝑚 moves to the right with
velocity 𝑢. It then collides elastically with a uniform rigid rod of mass 𝑀 and
length ℓ that lies on a smooth horizontal table. The particle hits the rod
perpendicularly at a distance 𝑥 from its centre.

Top view:

centre of rod

Figure 2

After the collision, the particle moves with velocity 𝑣 horizontally, the rod’s
centre moves with velocity 𝑉 horizontally, and the rod also rotates about its
centre with angular velocity 𝜔. The moment of inertia of the rod about its centre
1
is 𝑀ℓ2.
12

537
Question 4 (True/False)
The total linear momentum of the system of particle and rod is conserved.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision is elastic.
(B) The system is isolated.
(C) The external force acting on the system does not produce a torque.
(D) The rod rotates after the collision, so the linear momentum is transferred/transformed
to angular momentum.
(E) Linear momentum is always conserved in any case.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
As the system of particle and rod is isolated, no external force is acting on the system, and
the system’s linear momentum is conserved.
𝑚𝑢 = 𝑚𝑣 + 𝑀𝑉

538
Question 5 (True/False)
The total angular momentum of the system of particle and rod is conserved about the rod’s
centre.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision is elastic.
(B) The system is isolated.
(C) The external force acting on the system does not produce a torque.
(D) The rod rotates after the collision, so the linear momentum is transferred/transformed
to angular momentum.
(E) Angular momentum is always conserved in any case.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
As the system of particle and rod is isolated, no external torque is acting on the system, and
the system’s angular momentum is conserved.
1
𝑚𝑢𝑥 = 𝑚𝑣𝑥 + 𝑀ℓ2 𝜔
12

539
Question 6 (True/False)
The angular momentum of the rod only (excluding the particle) is conserved about the point
of collision, i.e. the point where the particle hits the rod.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision is elastic.
(B) The system is isolated.
(C) The external force acting on the system does not produce a torque.
(D) The rod rotates after the collision, so the linear momentum is transferred/transformed
to angular momentum.
(E) Angular momentum is always conserved in any case.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
The rod experiences an external force from the particle. However, this force does not create
any torque about the point of collision.
1
𝑀𝑉𝑥 − 𝑀ℓ2 𝜔 = 0
12
12𝑉𝑥
𝜔= 2
ℓ

540
Question 7 (True/False)
The total kinetic energy of the system of particle and rod is conserved.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision is elastic.
(B) The system is isolated.
(C) The external force acting on the system does not produce a torque.
(D) The rod rotates after the collision, so the kinetic energy is partially
transferred/transformed.
(E) Kinetic energy is always conserved in any case.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
Elastic collision is defined as a collision that preserves the kinetic energy of the system.
1 1 1 1 1
𝑚𝑢2 = 𝑚𝑣 2 + 𝑀𝑉 2 + ( 𝑀ℓ2 ) 𝜔2
2 2 2 2 12
1
𝑚𝑢2 = 𝑚𝑣 2 + 𝑀𝑉 2 + 𝑀ℓ2 𝜔2
12

541
CONSIDER THE FOLLOWING SCENARIO FOR
QUESTIONS (8)-(11)
Refer to the previous scenario in Figure 2.

In addition, it is desired to create a condition such that the rod rotates the
fastest after being hit by the particle, that is, 𝝎 is maximum. It is also
known that 𝑴 < 𝟐𝒎.

542
Question 8 (True/False)
After the collision, the particle comes to rest while the centre of the rod moves.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) All of the linear momentum of the particle is transferred/transformed into angular
momentum of the rod.
(B) All of the kinetic energy of the particle is transferred/transformed into rotational
kinetic energy of the rod.
(C) Conservation of momentum is violated.
(D) In the centre of mass frame of particle and rod, there is still translational motion after
the collision.
(E) The amount of torque produced is maximum.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

543
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
The described condition does not result in the maximum rotational kinetic energy of the rod.
In the centre of mass frame of particle and rod, both of them approach each other with equal
momentum before the collision, and move away from each other with equal momentum after
the collision. However, they don’t need to preserve their speeds in this frame, as some of the
energy is transferred/transformed into the rotational kinetic energy of the rod. Maximum
angular velocity of the rod occurs as the particle and the rod’s centre are at rest in this frame,
i.e. they move with the same velocity in the original frame.
𝑚
𝑣=𝑉= 𝑢
𝑀+𝑚

544
Question 9 (True/False)
After the collision, the rod’s centre stays at rest while the particle still moves.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) All of the linear momentum of the particle is transferred/transformed into angular
momentum of the rod.
(B) All of the kinetic energy of the particle is transferred/transformed into rotational
kinetic energy of the rod.
(C) Conservation of momentum is violated.
(D) In the centre of mass frame of particle and rod, there is still translational motion after
the collision.
(E) The amount of torque produced is maximum.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

545
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
The described condition does not result in the maximum rotational kinetic energy of the rod.
In the centre of mass frame of particle and rod, both of them approach each other with equal
momentum before the collision, and move away from each other with equal momentum after
the collision. However, they don’t need to preserve their speeds in this frame, as some of the
energy is transferred/transformed into the rotational kinetic energy of the rod. Maximum
angular velocity of the rod occurs as the particle and the rod’s centre are at rest in this frame,
i.e. they move with the same velocity in the original frame.
𝑚
𝑣=𝑉= 𝑢
𝑀+𝑚

546
Question 10 (True/False)
After the collision, both the particle and rod’s centre are at rest.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) All of the linear momentum of the particle is transferred/transformed into angular
momentum of the rod.
(B) All of the kinetic energy of the particle is transferred/transformed into rotational
kinetic energy of the rod.
(C) Conservation of momentum is violated.
(D) In the centre of mass frame of particle and rod, there is still translational motion after
the collision.
(E) The amount of torque produced is maximum.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

547
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
The described situation is definitely false as it violates the conservation of momentum. From
question 4,
𝑚𝑢 = 𝑚𝑣 + 𝑀𝑉

548
Question 11 (True/False)
ℓ
The value of 𝑥 equals to 2.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) All of the linear momentum of the particle is transferred/transformed into angular
momentum of the rod.
(B) All of the kinetic energy of the particle is transferred/transformed into rotational
kinetic energy of the rod.
(C) Conservation of momentum is violated.
(D) In the centre of mass frame of particle and rod, there is still translational motion after
the collision.
(E) The amount of torque produced is maximum.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

549
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
ℓ
If 𝑥 = 2,
𝑚𝑢 = 𝑚𝑣 + 𝑀𝑉
12𝑉𝑥 6𝑉
𝜔= 2 =
ℓ ℓ
1
𝑚𝑢 = 𝑚𝑣 + 𝑀𝑉 + 𝑀ℓ 𝜔 = 𝑚𝑣 2 + 4𝑀𝑉 2
2 2 2 2 2
12
Solving the equations yields
4𝑚 − 𝑀
𝑣= 𝑢
4𝑚 + 𝑀
2𝑚
𝑉= 𝑢
4𝑚 + 𝑀
which implies 𝑣 ≠ 𝑉. This means, in the centre of mass frame, the particle and the rod’s
centre are not at rest, but moving away from each other, taking some of the kinetic energy.
To maximise the angular speed of the rod, in the centre of mass frame of the particle and the
rod, they should come to rest after the collision. This implies that they move with the same
velocity in the original frame.
From the conservation of momentum,
𝑚𝑢 = 𝑚𝑣 + 𝑀𝑉 = (𝑀 + 𝑚)𝑉
𝑚
𝑉= 𝑢
𝑀+𝑚
From the conservation of angular momentum about the point of collision,
1
𝑀𝑉𝑥 − 𝑀ℓ2 𝜔 = 0
12
12𝑉𝑥
𝜔= 2
ℓ
From the conservation of energy,
2 2 2
1 2 2 2 2
1 2
12𝑉𝑥 2
𝑚𝑢 = 𝑚𝑣 + 𝑀𝑉 + 𝑀ℓ 𝜔 = 𝑚𝑣 + 𝑀𝑉 + 𝑀ℓ ( 2 )
12 12 ℓ
2 2
12𝑥 𝑚𝑢
𝑚𝑢2 = [𝑀 (1 + 2 ) + 𝑚] ( )
ℓ 𝑀+𝑚
12𝑥 2
𝑀2 + 2𝑀𝑚 + 𝑚2 = 𝑀𝑚 (1 + 2 ) + 𝑚2
ℓ
2
12𝑥
𝑀+𝑚 = 2 𝑚
ℓ
ℓ 𝑀
𝑥= √1 +
2√3 𝑚
Since it is known that 𝑀 < 2𝑚,
ℓ 𝑀 ℓ
𝑥= √1 + <
2√3 𝑚 2

550
CONSIDER THE FOLLOWING SCENARIO FOR
QUESTIONS (12)-(14)
Consider the following collision. A particle of mass 𝑚 moves to the right with
velocity 𝑢. It then collides elastically with a uniform rigid rod of mass 𝑀 and
length ℓ that lies on a smooth horizontal table. The particle hits the rod
perpendicularly at a distance 𝑥 from its centre.

Now, the rod is pivoted at its centre.

Top view:

pivot at the centre

of rod

Figure 3

After the collision, the particle moves with velocity 𝑣 horizontally and the rod
rotates about its centre with angular velocity 𝜔. The moment of inertia of the
1
rod about its centre is 𝑀ℓ2 .
12

551
Question 12 (True/False)
The total linear momentum of the system of particle and rod is conserved.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision is elastic.
(B) The system is isolated.
(C) The external force acting on the system does not produce a torque.
(D) The rod rotates after the collision, so the linear momentum is transferred/transformed
to angular momentum.
(E) Linear momentum is always conserved in any case.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
The pivot provides an external force on the system. Hence, the system is not isolated and the
momentum is not conserved.

552
Question 13 (True/False)
The total angular momentum of the system of particle and rod is conserved about the rod’s
centre.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision is elastic.
(B) The system is isolated.
(C) The external force acting on the system does not produce a torque.
(D) The rod rotates after the collision, so the linear momentum is transferred/transformed
to angular momentum.
(E) Angular momentum is always conserved in any case.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
The system of particle and rod is not isolated as there is an external force from the pivot.
However, this force creates no external torque on the system about the centre of the rod.
Hence, the angular momentum of the system about the centre of the rod is conserved.
1
𝑚𝑢𝑥 = 𝑚𝑣𝑥 + 𝑀ℓ2 𝜔
12

553
Question 14 (True/False)
The total kinetic energy of the system of particle and rod is conserved.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) The collision is elastic.
(B) The system is isolated.
(C) The external force acting on the system does not produce a torque.
(D) The rod rotates after the collision, so the kinetic energy is partially
transferred/transformed.
(E) Kinetic energy is always conserved in any case.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
Elastic collision is defined as a collision that preserves the kinetic energy of the system.
1 1 1 1
𝑚𝑢2 = 𝑚𝑣 2 + ( 𝑀ℓ2 ) 𝜔2
2 2 2 12
1
𝑚𝑢2 = 𝑚𝑣 2 + 𝑀ℓ2 𝜔2
12

554
CONSIDER THE FOLLOWING SCENARIO FOR
QUESTIONS (15)-(16)
Refer to the previous scenario in Figure 3.

In addition, it is desired to create a condition such that the rod rotates the
fastest after being hit by the particle, that is, 𝝎 is maximum. It is also
known that 𝑴 < 𝟑𝒎.

555
Question 15 (True/False)
After the collision, the particle comes to rest.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) All of the linear momentum of the particle is transferred/transformed into angular
momentum of the rod.
(B) All of the kinetic energy of the particle is transferred/transformed into rotational
kinetic energy of the rod.
(C) Conservation of momentum is violated.
(D) In the centre of mass frame of particle and rod, there is still translational motion after
the collision.
(E) The amount of torque produced is maximum.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

556
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
As the rod is constrained to rotate and not translate, to maximise its angular velocity, the
particle is required to stop moving after the collision and transfer all its kinetic energy to the
rotational kinetic energy of the rod. Momentum does not need to be conserved as the system
is not isolated due to the presence of the pivot. From question 14,
1
𝑚𝑢2 = 𝑀ℓ2 𝜔2
12
3𝑚 𝑢
𝜔 = 2√
𝑀 ℓ

557
Question 16 (True/False)
ℓ
The value of 𝑥 equals to 2.

Answer:
(A) True (B) False

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Answer unconfident confident confident

Reason:
(A) All of the linear momentum of the particle is transferred/transformed into angular
momentum of the rod.
(B) All of the kinetic energy of the particle is transferred/transformed into rotational
kinetic energy of the rod.
(C) Conservation of momentum is violated.
(D) There is still translational motion after the collision.
(E) The amount of torque produced is maximum.
(F) Other reason: ________________________________________________________
___________________________________________________________________

Confidence 1 2 3 4 5 6
Rating for Very Very Absolutely
Just guessing Unconfident Confident
Reason unconfident confident confident

558
Express your answer and/or reason mathematically, and use texts for
clarifications/explanations:
From question 13,
1
𝑚𝑢𝑥 = 𝑀ℓ2 𝜔
12
and from question 14,
1
𝑚𝑢2 + 𝑚𝑣 2 = 𝑀ℓ2 𝜔2
12
ℓ
If 𝑥 = 2,
6𝑚𝑢
𝜔=
𝑀ℓ
3𝑚
𝑣2 = ( − 1) 𝑢2
𝑀
Since it is known that 𝑀 < 3𝑚, this results in nonzero value of 𝑣, which means some of the
kinetic energy is taken by the particle. And this should not be the condition for the maximum
angular speed of the rod.

From questions 13 and 15,

1 3𝑚 𝑢
𝑚𝑢𝑥 = 𝑀ℓ2 (2√ )
12 𝑀 ℓ

ℓ 𝑀
𝑥= √
2 3𝑚

Since it is known that 𝑀 < 3𝑚,

ℓ 𝑀 ℓ
𝑥= √ <
2 3𝑚 2

559
 APPENDIX 5: INTERVIEW PROTOCOLS FOR STUDY ON

MOMENTUM/COLLISION

1. First of all, thank you for participating in the study by taking the diagnostic test(s).
And thank you for taking time today to come for the interview. This interview is very
crucial for our study. We want to better understand your mathematical thinking by
looking at how you approached certain question(s) in the test.

2. In your response to Question 9/10/13/14/15, you explained the mathematical

formalism in this manner (to show test paper answer). Can you elaborate on this?

3. Are there any other factors you should consider?

4. (For students who obtained full marks for the fifth tier, ask a further question,
as described below.)

For Q9, is there any possibility that some momentum is actually lost?
Expected answer: no, as long as no net external force, momentum is conserved.
What if I change the values of the masses?
Expected answer: there is no difference.

For Q10, is there any possibility that all the energy of the first mass is transferred to
the second mass?
Expected answer: no, because they have different mass.
What if their masses are equal?
Expected answer: then it is the only exception.

For Q13, is there any possibility that the maximum energy transfer occurs when the
first mass gives out all its momentum?
Expected answer: when both masses are equal.
What is the masses are not equal?
Expected answer: it’s not possible as either it’s not maximum transfer or the total
kinetic energy of the system would increase.

For Q14, is there any possibility that the maximum energy transfer occurs when the
first mass gives out all its energy?
Expected answer: when both masses are equal.
What is the masses are not equal?
Expected answer: it’s not possible as it would violate conservation of momentum.

For Q15, would it change your answer if the collision is perfectly inelastic?
Expected answer: yes, the energy transferred would have decreased.

560
 (For students who obtained partial marks for the fifth tier, lead them to obtain
the correct answer, and ask them further questions as described above.)
(For students who obtained zero marks for the fifth tier, no need to ask them
further questions.)

5. Thank you for taking time to come for the interview.

561
 APPENDIX 6: INTERVIEW PROTOCOLS FOR STUDY ON ROTATIONAL

MECHANICS

1. First of all, thank you for participating in the study by taking the diagnostic test(s).
And thank you for taking time today to come for the interview. This interview is very
crucial for our study. We want to better understand your mathematical thinking by
looking at how you approached certain question(s) in the test.

2. In your response to Question 1/3/5/6/16, you explained the mathematical formalism

in this manner (to show test paper answer). Can you elaborate on this?

3. Are there any other factors you should consider?

4. (For students who obtained full marks for the fifth tier, ask a further question,
as described below.)

For Q1, what will happen if the force is applied at the centre of the rod?
Expected answer: then there would only be translation and not rotation.

For Q3, so how much force is needed if we want to produce the same amount of
angular acceleration?
Expected answer: twice larger than the initial force.

For Q5, what if I change the location of the point of collision, will the angular
momentum of the system still be conserved?
Expected answer: yes, as the net external torque is still zero.

For Q6, what if I change the reference point, will the angular momentum of the rod
still be conserved?
Expected answer: no, as now there is a nonzero external torque.

ℓ
For Q16, would it be possible that 𝑥 = 2 gives you the maximum angular speed for
the rod?
Expected answer: yes, as long as their masses fulfill a certain ratio.

(For students who obtained partial marks for the fifth tier, lead them to obtain
the correct answer, and ask them further questions as described above.)
(For students who obtained zero marks for the fifth tier, no need to ask them
further questions.)

5. Thank you for taking time to come for the interview.

562
APPENDIX 7: SLIDES USED FOR TEACHING INTERVENTION ON

MOMENTUM/COLLISION

563
564
565
566
567
568
569
570
APPENDIX 8: SLIDES USED FOR TEACHING INTERVENTION ON

ROTATIONAL MECHANICS

571
572
573
574
575
576
577
578
APPENDIX 9: VALIDATION OF SURVEY INSTRUMENT

Comments from Validator 1

579
580
Comments from Validator 2

581
582
 APPENDIX 10: SURVEY INSTRUMENT ON TEACHING INTERVENTION ON

MOMENTUM/COLLISION

On the use of mathematical thinking in Physics

Topic: Momentum / Collision

Section 1

There are five responses for each statement:

SA – strongly agree
A – agree
N – neutral/not sure
D – disagree
SD – strongly disagree

For each statement, circle the response that most represents your view.

A On the nature of concepts on momentum / collision in mechanics

I can understand the topic of momentum / collision better if
1 SA A N D SD
mathematics is involved.
Mathematical analysis is crucial in the study of momentum /
2 SA A N D SD
collision in mechanics.
Mathematical analysis helps me to understand a number of
3 SA A N D SD
concepts on momentum / collision better than using intuition.
I like to explore more problems in momentum / collision with
4 SA A N D SD
the involvement of mathematics.
I prefer to learn the topic of momentum / collision qualitatively
5 SA A N D SD
without any use of mathematics.

B On the intervention lesson conducted by teacher

I find the intervention lesson on momentum / collision to be
6 SA A N D SD
useful.
I can now better appreciate why mathematics is needed for
7 SA A N D SD
understanding concepts on momentum / collision.
The practice problems gave me valuable experience in using
8 SA A N D SD
mathematical thinking to understand momentum / collision.
The discussions led by the teacher have helped me to apply
9 mathematical thinking in analysing problems in momentum / SA A N D SD
collision.
The discussions with peers on the practice problems in
10 momentum / collision have helped me to apply mathematical SA A N D SD
thinking in analysing these problems.
After this session, I am more confident of using mathematical
11 SA A N D SD
thinking in analysing problems in momentum / collision.

583
Section 2

1. What is it about the topic of momentum / collision that makes it difficult to understand?

2. What do you like or dislike about the use of mathematical thinking on the topic of
momentum / collision?

Student ID: _____________

584
 APPENDIX 11: SURVEY INSTRUMENT ON TEACHING INTERVENTION ON

ROTATIONAL MECHANICS

On the use of mathematical thinking in Physics

Topic: Rotational Mechanics

Section 1

There are five responses for each statement:

SA – strongly agree
A – agree
N – neutral/not sure
D – disagree
SD – strongly disagree

For each statement, circle the response that most represents your view.

A On the nature of concepts on rotational mechanics

I can understand the topic of rotational mechanics better if
1 SA A N D SD
mathematics is involved.
Mathematical analysis is crucial in the study of rotational
2 SA A N D SD
mechanics in mechanics.
Mathematical analysis helps me to understand a number of
3 SA A N D SD
concepts on rotational mechanics better than using intuition.
I like to explore more problems in rotational mechanics with
4 SA A N D SD
the involvement of mathematics.
I prefer to learn the topic of rotational mechanics qualitatively
5 SA A N D SD
without any use of mathematics.

B On the intervention lesson conducted by teacher

I find the intervention lesson on rotational mechanics to be
6 SA A N D SD
useful.
I can now better appreciate why mathematics is needed for
7 SA A N D SD
understanding concepts on rotational mechanics.
The practice problems gave me valuable experience in using
8 SA A N D SD
mathematical thinking to understand rotational mechanics.
The discussions led by the teacher have helped me to apply
9 mathematical thinking in analysing problems in rotational SA A N D SD
mechanics.
The discussions with peers on the practice problems in
10 rotational mechanics have helped me to apply mathematical SA A N D SD
thinking in analysing these problems.
After this session, I am more confident of using mathematical
11 SA A N D SD
thinking in analysing problems in rotational mechanics.

585
Section 2

1. What is it about the topic of rotational mechanics that makes it difficult to understand?

2. What do you like or dislike about the use of mathematical thinking on the topic of rotational
mechanics?

Student ID: _____________

586
APPENDIX 12: STUDENTS’ RESPONSES IN SURVEY ON LESSON

INTERVENTION ON THE TOPIC OF MOMENTUM/COLLISION

587
APPENDIX 13: STUDENTS’ RESPONSES IN SURVEY ON LESSON

INTERVENTION ON THE TOPIC OF ROTATIONAL MECHANICS

588
 APPENDIX 14: STUDENTS’ REFLECTIONS ON LESSON INTERVENTION

Student 27

The lesson on momentum was very meaningful and refreshing. Since my Year 1 days, I never
gave much heed to the mathematics behind physics and I strongly felt that it was unimportant
when studying physics as I thought that the concepts were ‘common sense’. However, this
lesson gave me a fantastic opportunity to understand the mathematical thinking and analysis
of various physical phenomena that occur all around me and that things may sometimes defy
‘common sense’. In addition, I never fully grasped the importance of the mathematical
concept of the centre of mass and how it could play an important role when solving problems.
I always believed that the centre of mass formula was unimportant but upon attending the
lesson, I realised the magnificence of a very simple mathematical concept that I actually had
learnt before: in the centre of mass frame, the total momentum is always zero when there is
no net force. In addition, the group discussions were very insightful as we would come
together and discuss all the concepts that were covered through the practice questions. This
enabled us to weed out any misconceptions that any of us would have and was a fantastic
platform to voice our thoughts and concerns regarding various concepts to our peers. Hence,
I feel that the group discussions were very productive and beneficial for my learning. Lastly,
attempting the practice questions allowed me to reflect and solve the problems using
mathematics. The practice questions were quite tricky but very ‘innovative’ and there were
times when the answers to some problems would just blow my mind as mathematical analysis
gives solutions that one will most-likely never obtain with normal ‘common sense’. There
were many times when the mathematics really showcased the solution to the problem in a
much clearer and elegant manner. Thus, in conclusion, I am very happy to have been part of
this research as it allowed me to brush up on my fundamentals. I truly appreciate the efforts
of this research and I hope to employ more mathematical analysis into the physics problems
that I will encounter in the years to come.

Student 28

The lesson was rather enjoyable. I had a fun time interacting with my peers and fixing our
misconceptions about how collisions worked in reality and where our intuitions failed
together. I found that it was far more concrete and less error-prone to deal with problems in
terms of mathematical quantities to see how a system’s final state was determined, as well as
to reduce possible errors due to false intuitions. Expressing momentum, kinetic energy and
classifying collisions using mathematics helped discretise the different possible systems and
made it easy to distinguish between elastic and inelastic collisions and the like simply by
observing the changes in kinetic energy. Performing mathematical analysis in the centre of
mass frames seemed to oversimplify problems until they were laughable – since the total
momentum would be zero in said frames, two bodies could only be moving toward each other
away from one another before and after collisions, which made the system very easy to solve.
This provided me with great insight on how mathematical thinking really helps in analysing
physical situations – something I never considered seriously prior to this lesson even after

589
learning both the concept and the mathematics. The fact that Newton’s Laws can so easily
predict collisions and be expressed in a few symbols is very useful and solving momentum
problems came down to solving equations which govern the system. Moreover, it is hard to
make mistakes when the equations are written on paper such that we can test any state we
think of by simply plugging in numbers and seeing which laws they would violate and hence
why those lines of reasoning would be erroneous. Now when I approach momentum
problems, or simply, physics problems in general, I find it easier to rely on equations that
cannot go wrong, compared to human intuition, which can be wrong at times.

Student 29

Through this lesson, I gained a deeper understanding of the topic of discussion. Even though
momentum is a topic I had learnt before and I knew the formula very well prior to this, I
often relied on my intuition instead of proper mathematical approach whenever solving
physics problems. As far as I agree that intuition plays a very important role in Physics, but
without calculation, intuition may lead us in the wrong direction sometimes. For example, I
thought the best case scenario in a collision would happen when a body stops after the
collision, but this turns out to be incorrect after performing thinking of it mathematically.
The mathematical method is also a good way to guide us thinking in a more logical way and
even develop our intuition to foresee the state of a physical system.

Additionally, we also had a meaningful discussion about the practice questions. During the
discussion, some of my groupmates still tended to use their intuition to solve the problem
and was hardly convinced. Nevertheless, after using a mathematical approach to analyse the
question carefully and solve it step by step instead of jumping to a conclusion based on our
intuition, we all agreed on the correct answer.

When we express mathematical thinking to solve a physics problem, there may be innovative
ways to address the problem, even sometimes faster. At the start, some questions require a
lot of calculation and we found it hard to get the result, but by reconsidering the scenario our
group finally managed to simplify the calculation. For example, after revisiting the lesson,
our group realised that looking at objects in the centre of mass frame was really an effective
method to approach the questions mathematically.

Generally, this is a really meaningful lesson and good revision on momentum and collision.
In fact, what we learnt is far beyond the topic itself, we learnt the importance of treating every
problem with rigorous mathematical approach instead of experiences and intuition.

Student 30

Physics is the science that lets us understand the various physical phenomena around us,
allows us to find out the inner relationship among variables, and trains our thinking patterns
and logical reasonings. The lesson conducted by Mr Ricardo was very helpful and I am
thankful to be part of it. With his clear explanations and rigorous mathematical thinking
processes, many situations revolving around the concept of momentum were made clearer to

590
us. Like peeling the onion, layer by layer, to the core. I corrected a lot of misconceptions I
used to have as I tended to ignore the mathematics in approaching conceptual questions.

The main thing I learnt from his lesson is the importance of a rigorous reasoning process,
especially with the use of mathematical tools. Math is the most powerful and fundamental
tool that we can fall back on when solving science problems. After Mr Ricardo’s lesson, I
realised that our intuition was not always correct. As the situation becomes more and more
complicated, we can not rely on our intuition any more. The application of mathematical
thinking is a more rigorous way of analysing physical concepts. Mathematical tools such as
calculus, manipulation of polynomials, limits, etc. are very useful tools in physics. So, I think
the most important lesson I learnt is to form a close connection between physical concepts
and mathematical thinking.

The highlight of the lesson is to use mathematics to analyse situations that are beyond our
imaginative abilities and a small group discussion. I realised that imagination and language
description could only meet the requirements of superficial understanding. As the problem
gets more complex, the best way is to introduce mathematical models to describe the
situation. While the imaginative description is more important in introducing of physics to a
larger audience, but when it comes to solving deeper physical problems, a rigorous logical
reasoning and mathematical derivation are necessary. Sometimes we get stuck in our thinking
when relying only on imagination, but mathematical thinking makes the problems easily
solved. The small group discussion involving whiteboarding was a highlight because it
allowed us to find out how our peers approached a problem, what their misconceptions were,
and to grow together. Indeed, some misunderstanding was still surfaced out even after Mr
Ricardo’s lesson, and the small group discussion helped us eliminate our doubts.

After the lessons, I gained the habit of applying mathematical thinking when solving physics
problems, even in my core physics module. It is very hard to arrive at the solution only by
qualitative reasoning. Mathematical calculations, derivations and proofs are irreplaceable in
getting the reliable solution.

Student 31

Personally, I felt that it was a rather enriching lesson, clearing up misconceptions that I had
by emphasising on the importance of mathematical analyses when analysing physical
phenomena. The practice questions were thought-provoking, and really tested the extents of
my understanding about momentum and collisions. The discussions phase was rather useful
for both my classmates and I, for after trying the practice questions, only when discussing
did we realise that what we thought to be correct might in fact not be the case, and it required
us to consider other ways of thinking. It was intriguing to see this happen, though the concept
of momentum and collision was not something new to us; we had learnt it rigorously before
though without the emphasis on mathematical thinking. For example, before the discussions
phase I used to think that the most efficient method for momentum/energy transfer would be
to use a really heavy object with a really light object, given situations like the “superball”,
where a light ball is observed to jump significantly higher when dropped a small height above

591
a dropping massive ball. Upon discussion with other students, however, I realised that such
intuition was, in fact, wrong, and suggestions by my groupmates about considering other
frames enlightened me, especially after doing the mathematics. After the lecture and
discussions phase, I realised the failure of intuition in a lot of cases and understood how to
better make use different reference frame. Subsequently, after analysing the scenarios
involved in the practice questions mathematically, I verified that my intuition was indeed
wrong. There was much satisfaction in eventually solving for the equations embodying the
collisions and obtaining a mathematical rigorous answer, and it was certainly intriguing that
intuition could be so wrong. I found analysing using the centre of mass frame of reference
rather neat, and it has certainly enabled me to better understand collisions and momentum
transfer, from a mathematical standpoint.

Student 32

Mr Ricardo is an excellent teacher and was able to get the content and points across well. He
used a multitude of visual aids and equations and every single one of them represented the
line of thinking he was trying to get across. The lesson enabled us to discover misconceptions
we had previously, though we had learnt the topic of rotational dynamics and all the equations
prior to this. Some diagrams included illustrations of the collisions between point masses and
rigid bodies, but the main highlight is the use of mathematical thinking when approaching a
physical situation. In fact, he often highlighted that intuition may not always be the best way
of deducing results.

We learnt to formulate our ideas on thoughts by utilising more mathematical thinking. Even
things that may seem trivial might not be easily explainable. Writing down equations gave a
better in-depth understanding compared to visualising the scenario in our head. In addition,
mathematical thinking gave us an edge on rationalising certain questions that might seem
challenging at first glance.

After attempting the practice questions individually, we discussed the questions amongst
ourselves. While most of us used similar methods to obtain the correct answer, some
groupmates were still unable to achieve the correct answer. After emphasising the
mathematical thinking behind the problem, we were then able to answer our doubts.

After attending Mr Ricardo’s revision on the conservation of momentum and conservation

of angular momentum, it was able to shine a light on the topics and erased most of the doubts
I had. Mathematical treatment in considering a situation in a different frame of reference
certainly reduces a lot of situations to a simpler problem. Mathematical thinking definitely
gives a more concrete explanation than intuition as intuition may not always be right
especially in non-trivial cases.

Student 33

The lesson was fun and eye-opening. Although I knew the concepts taught during lessons but
looking at the concept and reviewing my knowledge using a more mathematical approach is

592
different from what I am used to. This made the lesson refreshing as it had me constantly
thinking and processing about what was being mentioned. Mr Ricardo also made the concepts
he was trying to bring across extremely clear, making it easy to follow him throughout the
lesson.

Using a more mathematical approach to learn did clear up some previous misunderstanding
that I had as one can easily tell if something is true or false. It is also clearer to conclude the
dynamics of the system by looking at the math as opposed to just thinking about it intuitively.
The holes in my understanding of rotational dynamics have also been effectively patched up
using a mathematical approach. The biggest source of confusion I noticed during the group
discussion was in how the system would be rotating and about which points. It was easy to
get the conclusion wrong by purely relying on one’s intuition in certain setups and even after
discussing it with other group members, it remained quite challenging to settle on which was
right. After the discussion using a more mathematical approach, it was much easier to
conclude how the system would behave. The mathematical thinking shared by Mr Ricardo
during the lesson regarding tackling such dynamic systems were enlightening as they were
not taught in the ordinary physics lessons I took previously. After the sharing session, my
groupmates and I were able to quickly and correctly determine, through mathematical means,
how different systems would behave. After the sharing, I am now more aware that I should
try to use math more than my own intuition when approaching a physics problem. I now also
try to use a more mathematical approach when it comes to solving any physics problem and
to go through the math before concluding. I realised that by doing so I am more effective and
accurate when it comes to solving the especially tough and complicated physical systems.

Student 34

The rotational mechanics lesson was very reinvigorating as we went through the important
but commonly forgotten fundamentals that play a critical role in shaping our understanding
of rotational mechanics in general. When I first learnt about angular momentum in Year 3,
my understanding of it was very shaky. I did not have a clear idea of the physical significance
of angular momentum and as a result, I developed many misconceptions about it. However,
the lesson helped clear a lot of my misconceptions. For instance, for many years, I have
always thought that linear momentum could be ‘converted into’ angular momentum and vice
versa, even after learning the mathematics. Upon attending the lesson and actually relooking
the mathematics behind it all, I realised how ridiculous it actually is to say that they can be
converted into one another. The lesson was very beneficial and did a great job reinforcing
the fundamentals mathematically which one would commonly miss out when learning those
concepts for the first time. I enjoyed the group discussions a lot as we were able to interact
with and correct the misconceptions of many of our peers and in fact, my friends also helped
me correct my misconceptions too. The discussion provided an excellent platform for us to
debate and process the content which we had just covered in the lesson and it was very
insightful. Next, approaching physics problems from a mathematical analysis based
standpoint was a very new idea to me as I would commonly think of physics as ‘common
sense’ but upon attempting the practice questions, I realised that sometimes, ‘common sense’
may be awfully incorrect. I gained a greater sense of appreciation for the mathematical

593
analysis and I am very happy to have been part of this research as I have gained tremendously
out of participating in it. I now understand the importance of mathematical analysis and in
the future, I will approach more problems this way instead of using the traditional ‘common
sense’.

Student 35

The lesson serves as a great revision of essential concepts such as linear and angular
momentum, emphasising the concepts in terms of not just its physics but also the underlying
mathematics.

I enjoyed the lesson as well as the discussion. I had a fruitful discussion with peers using
similar yet not-quite-same mathematical methods. It turned out that the presentation of
physical concept in simultaneous equations was very fun indeed. We had disagreement and
discussion on how to impose certain condition by varying the variable in the simultaneous
equations. And in the end, I got a much concise solution than the one I worked out before.
Through this discussion, manipulation skills have been horned and most importantly it added
to my confidence in problem-solving in many ways.

Increasing confidence also made me love the concepts and this field of physics to a greater
extent and prompted me to discover more myself after the lesson.

If I am to name something I’ve learnt from the discussion, our group discovered the idea that
in a two-body linear collision, the velocities after the collision, i.e. 𝑣1′ − 𝑣2′ , suffices to
suggest the elastic, inelastic and perfectly inelastic cases. Before the lesson, I would only be
able to appreciate the separation of discussion on the three cases in the physical sense, even
though I had learnt all the formulas before. But the lesson made me feel thrilled about the
mathematical reason that the three cases of collision arise from.

Similar mathematical intuition has been forged in me by the intriguing questions posed by
the instructor on energy transfer. Deriving the result mathematically and engaging the
discussion on this result with my peers also made me rethink the nature of momentum and
make an analogy in my mind to chemical reaction. So all in all, the lesson motivated me to
realise that there is a great philosophy behind the mathematics describing a physical situation.

Student 36

The lesson was rather enjoyable. I had a fun time interacting with my peers and fixing our
misconceptions about how rigid body collisions worked in reality and where our intuitions
failed together. We had this running joke about converting momentum into angular
momentum when I realised that some of my peers (and myself included) might have been
using this erroneous line of reasoning to solve problems (albeit incorrectly). I found that it
was far more concrete and less error-prone to deal with problems in terms of mathematical
quantities to see how a system’s final state was determined, as well as to reduce possible
errors due to false intuitions. It was using the mathematical thinking that I managed to

594
observe momentum could not be converted into angular momentum since they were
independent quantities describing different parts of a system and ended up clearing that
misconception. I had to also recall that angular momentum was defined based on where you
took a moment about and hence it would be conserved about certain points and not others. It
is by abusing these special points that we can more easily find the final state of systems.
These lines of mathematical thinking are really elegant, though we all had learnt the concept
and mathematics behind it all prior to this lesson. Expressing angular momentum, linear
momentum and kinetic energy and using mathematics helped reach some non-intuitive
conclusions about what would result in certain situations. The mathematical analysis in the
centre of mass frames helped simplify the system and helped me observe how I could
maximise physical quantities in accordance with the conservation laws. Overall, it is hard to
make mistakes when the equations are written on paper such that we can test any state we
think of by simply plugging in numbers and seeing which laws they would violate and hence
why those lines of reasoning would be erroneous. Now when I approach collision-based
problems, or simply, physics problems in general, I find it easier to rely on equations that
cannot go wrong, compared to human intuition.

Student 37

The lesson was a good revision of rotational mechanics especially angular momentum, the
content of which I had learnt since last year. The instructor recapped the basic concepts and
formulae and went through the standard calculations that we used to do during exams. The
practice questions he gave us were seemingly revision questions, and hence at first I felt lazy
to do the calculations and performed analysis based on intuition. However, the instructor
proved me wrong and made me realise that intuition and common sense often fail us when it
comes to physics. Moreover, he provided us with alternative ways of thinking, for example,
changing the reference frame, which made the thinking process as well as the calculations
much easier. In some cases, it was actually faster than using intuitions. Such revelation has
incurred a revolutionary paradigm shift which prompts me to embrace more mathematical
thinking in the context of physics.

Later during the discussion session, I tried to utilise formulae and calculations to help
determine the answer, and as expected, the correctness and accuracy of my answers improved
tremendously. Meanwhile, some of my teammates still made the same mistakes I used to
make during the discussion, so I encouraged them to use mathematics to aid them. In fact, in
the process of our discussion, one of them actually proposed an alternative solution to the
problem that made the maths much easier.

The lesson was truly beneficial to me. At first, using mathematics may seem to slow me down
in answering the questions, but it turns out to improve the correctness and accuracy and
makes me much more confident in my answer. In fact, as I approached the later parts of the
practice set, my speed of problem-solving using mathematical thinking had drastically
improved. Therefore, I will definitely exploit this strategy in my future learning of physics.

595