Proceeding of International Conference On Research, Implementation And Education

Of Mathematics And Sciences 2014, Yogyakarta State University, 18-20 May 2014

PE - 14
THE DEVELOPMENT OF PHYSICS ESSAY TEST
FOR HIGHER ORDER THINKING SKILLS IN JUNIOR HIGH SCHOOL
Edi Istiyono
Physics Education Department, Mathematics dan Natural Science Faculty,
Yogyakarta State University
email: edi_istiyono @uny.ac.id; [email protected]
Abstract
This research has been done to develop an instrument for measuring junior high school students’ physics
higher order thinking skills (PhysETHOTSS) and to obtain the characteristics of the PhysETHOTS. The
instrument blue print has been developed based on the aspects and sub-aspects of higher order thinking
skills, then it was used to develop the items. The instrument consisting of 24 items were validated by
physics educational measurement experts. The validated instrument was tried out on students of junior
high school (SMPN 1 Sewon). The polytomous data were analyzed according to the partial credit model
(PCM). The results show that the 24 items of PhysETHOTS were fit to the PCM, the reliability of the test
was 0.75, the items’ difficulty indexes ranged from -1.22 to 0.34. Therefore, the PhysETHOTS is
qualified for the measurement of junior high school students’ physics higher order thinking skills.
Keywords: instrument development, physics essay test of higher order thinking skills, polytomous, and
PCM

INTRODUCTION
Today the world is in an era of globalization that needs quite tight competition. In this
era the competition is quite tight, the competition of human resources (HR). The quality of the
nation's human resources is determined by the education level of the nation. Improving the
quality of education can begin from improving the learning quality. Improving the learning
quality can begin by setting appropriate learning objectives.
One of the aims Science learning in junior high school so that learners have the ability
to develop reasoning skills in the analysis of inductive and deductive thinking using concepts
and principles of physics to explain the events of nature and solving problems both qualitatively
and quantitatively (BSNP, 2006: 160). Thus, through the study of physics students are expected
to develop themselves in thinking. Learners are required not only have the ability of lower order
thinking, but the higher order thinking skills (HOTS). With regard to the higher order thinking
skills, the fact remains that the Indonesia physics achievement as measured on the reasoning
aspect is ranked 40th of 42 countries (TIMSS & PIRLS International Study Center, 2012:48).
Thus, the physics achievement of Indonesian high school students that requires of HOTS in the
international level is low. The low physics achievement can be caused by an improper learning
process or assessment model. In this case, only assessment will be discussed, because proper
assessment can encourage students to learn higher order thinking skills.
Based on Piaget's development theory, the formal operational stage is a stage of children
beginning from eleven years old. At this stage the children have begun to develop the ability to
manipulate abstract concepts through the use of propositions and hypotheses (Piaget, 2005: 122
and Reedal, 2010:7). The junior high school students are between 12 to 15 years, so that higher
order thinking skills of junior high school students have been established.
The revised Bloom’s taxonomy divide the cognitive aspect into lower order thinking
skills (LOTS) and higher order thinking skills (HOTS). LOTS include the ability to remember,
understand, and apply, while the HOTS include the ability to analyze, evaluate, and create

PE-121
 Edi Istiyono / The Development of … ISBN 978 – 979 – 99314 – 8 – 1

(Anderson and Krathwohl, 2001:30). Bloom’s taxonomy has been applied in the education.
Bloom's taxonomy is still used in many curricula and teaching materials (Brookhart, 2010: 39;
Schraw and Robinson, 2011: 158-159). Thus the HOTS in physics includes physics abilities, i.e.
analyzing, evaluating, and creating.
According to Brookhart (2010:5) higher order thinking skills (HOTS): (1) high-order
thinking is at the top of Bloom's cognitive taxonomy, (2) The purpose behind teaching cognitive
taxonomy to equip learners to transfer knowledge, (3) able to think meaning that learners are
able to apply the knowledge and skills they developed while studying in a new context. In this
case the term "new" is a concept application that has not been thought of before by learners, this
means that the universal is not necessarily something new. Higher-lorder thinking means the
ability of learners to connect learning with other things that have never been taught.
To monitor the process, progress, and improvement of students’ learning outcomes on
an ongoing basis, the necessary assessment. Educational assessment is the process of
information collecting and processing to determine the achievement of students’ learning
outcomes (Regulation of the Minister of National Education, No. 20, 2007). Assessment can be
done orally or in writing. Written assessment is conducted by a written test. There are two forms
of the written tests, namely selecting and supplying the answers. Written test by selecting
answers include: multiple choice, two- choice (true - false, yes - no), matching, and cause and
effect.
Keep in mind that the evaluation model also effects the thinking skills of students.
According to van den Berg (2008:15) that the curriculum has a rich potential for developing
higher-order thinking skills of learners. Teachers have to plan well and engage learners in
activities that encourage and develop the higher order thinking. Assessment can be implemented
to bring students in improving their higher order thinking skills. This is supported by the other
opinions, higher-order thinking questions that encourage students to think deeply about the
subject matter (Barnett & Francis, 2012 : 209). Based on this argument means that the
assessment, test of higher-order thinking skills, provide stimulation of students to develop high
order thinking skills as well.
Nitko and Brookhart (2011:223) describe that the basic provisions of the assessment is
the ability to use higher order thinking tasks that require the use of knowledge and skills in new
situations. Must use new materials to assess the higher order thinking skills. One way done use
sets of items that depend on the context .
There are disadvantages of multiple choice test, namely: (1) students chances to guess
the answer is still quite large and (2) the students’ thinking process cannot be seen exactly
(Sujana, 1990:49 ). Therefore, essay test is good alternative test.
Assessment are based on the stages can be completed examinees. Although only just
completed the initial stage, the examinees had to get the value. The highest value of course
obtained when the examinee has completed all phases of the exam in point. The assessment
procedure is the same with how individuals respond to the items in the psychological scale. For
example, an item that provides four categories of response of 'never', 'rarely', 'often', and
'always ' analogous to the completion stage. Just about to finish the first stage is analogous to
the category of 'never', while when it comes to the final stage, analogous to the category of
'always'. This assumption was later developed into a partial credit model (PCM). When it is
assumed that a partial credit item then follow the pattern of higher ability individuals are
expected to have higher scores than individuals who have a low ability (Widhiarsa, 2010: 6).
According to Wright & Masters, PCM is also appropriate to analyze the response to the
measurement of critical thinking and conceptual understanding in science (Linden &
Hambleton, 1997: 101-102)

PE-122
 Proceeding of International Conference On Research, Implementation And Education
Of Mathematics And Sciences 2014, Yogyakarta State University, 18-20 May 2014

Based on the above, of various types of written test, essay test is a good one to explore
the physics higher order thinking skills in junior high school. To measure higher-order thinking
skills is used test that called Physics Essay Test for Higher Order Thinking (PhysETHOTS). So
that we need to develop of physics essay test for higher order thinking skills (PhysETHOTS).
Based on the description in the future, the goal is : (1) to develop an instrument for measuring
junior high school students’ physics higher order thinking skills (PhysETHOTS); and (2) to
obtain the characteristics of the PhysETHOTS

RESEARCH METHOD
This research is the development research with quantitative approach. This instrument
development research was done with the modified of the Wilson Model and Antonio Oriondo
Model.
The test instrument development used a modified form of the Wilson and Antonio
Oriondo model, consisting of: (1) the design of the test and (2) the test tryout. The test design
phase included: (1) the determination of objective tests, (2) the determination of competency to
be tested, (3) the determination of the tested material, (4) the preparation of test blue print, (5)
the writting of items based on the principles of HOT test development, (6) the preparation the
scoring guidelines, (7) test validation and (8) the repairing the items and assembling the test,
The stages of the development of the test are presented in Figure 1. The try out included: (1) the
establishment of try out subjects and (2) the implementation of the tryout.
The determination of objective tests

The determination of competency to be tested

The determination of the tested material,

The preparation of test blue print

The writting of items based on the principles of HOTS test development

The preparation the scoring guidelines

Test validation

The repairing the items and assembling the test

The establishment of try out subjects

The implementation of the tryout.

Analyzing of Tryout Data

Test Assembling

Figure 1. Steps of The Instrumen Development

PE-123
 Edi Istiyono / The Development of … ISBN 978 – 979 – 99314 – 8 – 1

Related to the sample number, according to some measurement experts IRT analysis
requires 200 to 1000 people (Seon, 2009: 3). Reckase (2000) concluded that the minimum
sample size for estimating the three parameters, which include discrimination, the difficulty
index, and pseudoguessing, is 300 (Haladyna, 2004: 206). So with the PCM model of 1PL, the
students for the tryout subjects as many as 100 are more than enough.

RESULT AND DISCUSSION

Result of The Test developed
The PhyTHOTS instrument consisted of 24 items, The test included sub physics matter:
force, Newton’s law, work and energy, simple machines, pressure, vibrations and waves, sound,
light, and optical instruments and sub-aspect of HOTS: analyze, evaluate, and create. The items
distribution is presented on Table 1. The PhyETHOTS was validated by experts judgment.
Description of the Physics instruments higher order thinking skills (PhysETHOTS) in JHS
validation has been done then do the next step is try out. Tests conducted on 100 students in
grade VII of SMP N 1 Sewon. The response of students then assessed and given a score on
respondents (examinee). Score of the students are coded in note pad for analysis preparation.

Table 1. Distribution of Item PhyETHOTS in Grade VII of Junior High School

Kompetency Standard and Physics Matter
Dimension 5. Understanding the role of 6. Understanding the
Kognitive Kognitive work, force, and energy in concept and
Category Process daily life application of
vibration, waves and
optics in daily
technology products

instrument
and waves
Vibrations
Work and

machines
Newton's

Pressure

Optical
Simple
energy

Sound
Force

Light
law

s
Differenting 1 15 18
Analyze Organizing 4 12 23
Attributing 7 10 21
Checking 2 8 19
Evaluate Critiquing 5 13 16
Generating 9 11 22
Create
Planning 3 6 17

Producing 14 20 24

Goodness of fit of Instrument

Testing for goodness of fit for the overall test and each item is carried out. Testing
goodness of fit the fit for the overall test developed Adam Khoo (1996:30) based on the mean
value of INFIT Mean Square (Mean INFITMNSQ) and its standard deviation or average values
INFIT t (Mean INFIT t) and its standard deviation. If the average INFIT MNSQ approximately
1 and 0.0 standard deviation or mean INFIT t close to 0 and standard deviation 1.0, then the

PE-124
 Proceeding of International Conference On Research, Implementation And Education
Of Mathematics And Sciences 2014, Yogyakarta State University, 18-20 May 2014

overall fit test with PCM. The INFITMNSQ is 1.01 (about 1) and a standard deviation is 0.13
(approximately 0.0), therefore the overall test fits with 1 PL PCM model.
Testing for goodness of fit Item and testee is determined that an item or testee is fit by
models with boundary MNSQ INFIT range of 0.77 to 1.30. The MNSQ INFIT values of items
between 0.78 to 1.27. Thus, 24 items are fit with PCM model.
Reliability
Based on the analysis, the reliability of the instrument (test) is estimated at 0.75.
Reliability value is qualified as good instrument.
Item Characteristic Curva (ICC)
The characteristics of the item indicated by the curves characteristic of the item (ICC)
and the index of difficulty. Based on the analysis, there were obtained item characteristic curves
(ICCs) as many as 24 pieces. Figure 2 presented the characteristic curve item 1, that means: (1)
score of 1 is largely for very low ability students (θ = -3), (2) score 2 mostly to moderate ability
students (θ = 0), (3) score 3 mostly for high ability students (θ = 1), (4) a score of 4 and 5
mostly for very high ability students (θ = 3). The items’ difficulty index from the small to the
large ones sequential categories 1, 2, 3, and 4.

Figure 2. Item Characteristic Curve 11

The Difficulty Index

The items’ difficulty index were from -1.22 to 0.34 with an average of 0 and a standard
deviation of 0.35. So that based on difficulty (-2.0 < b < 2.0), all of 24 items were good. For
more details, please see diagram distribution of items according to index difficulty and subaspek
aspects of the instrument in Figure 3. Based on Figure 3, the order of item difficulty index of
each aspect is analyzing, evaluating, and creating.

PE-125
 Edi Istiyono / The Development of … ISBN 978 – 979 – 99314 – 8 – 1

Figure 3. Item’s difficulty Index of Each Aspect and Subaspect Instruments

Information Function and SEM

Based on the data analysis, it was obtained information and standard error of
measurment (SEM). Based Functions information and SEM presented Figure 4, the test is
suitable for the students that whose ability (θ) is high, ie 1 ≤ θ ≤ 2.8. This is consistent with the
purpose of the developed instrument to measure Physics higher-order thinking skills.

Figure 4. Information Function and SEM

Discussion
PhysETHOTS reliability is 0.75 that mean the test is good. It is said to be good, if the
test has a reliability coefficient of more than 0.65 (Mchrens & Lehman (1991: 263). In addition,
the information function is relatively for high ability between 1.0 to 28. This means that this
instrument has high strength and reliable because it is composed of items that have high
information function (Hambleton and Swaminathan, 1985: 94). Based on the reliability
coefficient, the test information functions, and parameter estimation, this means PhysETHOTS

PE-126
 Proceeding of International Conference On Research, Implementation And Education
Of Mathematics And Sciences 2014, Yogyakarta State University, 18-20 May 2014

realiabel and has high stability.

Content validity of the test has been proven by expert judgment. Empirically verified
the validity of the goodness of fit of the partial credit model (PCM). Based on Table 2, the
average value and the standard deviation INFIT MNSQ 1.01 each (about 1) and 0.13 (about 0) ,
then the fit test with 1 PL PCM. This means that the test empirically valid. This is supported by
all the items have a value between 0.78 INFIT MNSQ up with 1.27 which lies between the
limits of receipt of the item using INFIT MNSQ or fit according to the model (between 0.77 to
1.30) means that all items fit many as 24 items of all. This is caused by several things, among
others: (1) the items were developed according to the procedure correct item development
instruments, (2) the items were developed from indicators derived from aspects of high order
thinking skills and materials physics, (3) test which consists of 24 items that has tested the
content validity by expert judgment, and (4) the respondents (students) were tested in earnest
in doing because it involves supervisors of their physics teacher.
According to Hambleton & Swaminathan (1985:36), the item’s difficulty index are good
if they varied between 2.00 to 2.00. Items whose difficulty index of -2.00 indicates this is very
easy, while the difficulty index of 2.00 means that the item is very difficult. Thus, based on the
item’s difficulty index the instruments (from -1.22 to 0.34 ) are good.

CONCLUSION AND SUGESSTION

Conclusion
Based on the analysis, the conclusions are as follows:
1. PhysETHOTS instrument was developed on junior high school students’ abilities to
analyze, evaluate, and create and on sub physics matter: force, Newton’s law, work and
energy, simple machines, pressure, vibrations and waves, sound, light, and optical
instruments. The PhysETHOTS instrument is essay test that consisted of 24 items.
2. Characteristic of PhysETHOTS are:
a. PhysETHOTS has content validity provided by expert judgment and empirical
evidence has been getting fit with Partial Credit Model (PCM) based on polytomous
data five categories.
b. All items of PhysETHOTS on the criteria well as the difficulty index is in the range
between 2.00 to 2.00.
c. PhysETHOTS reliability is qualified
d. Based on the information function and SEM, PhysETHOTS is very appropriately used
to measure students’ higher order thinking skills physics of 1.0 to 2.8.

Sugesstion
Based on the analysis, it are recommended:
1. Teachers can implement physics tests of high order thinking skills in junior high school.
2. Training for the development of physics test of higher order thinking skills is required for
teachers.
3. Further research can be done using the data analysis by generalized partial polytomus
credit model (GPCM 3PL).

REFERENCES
Anderson, L.W., & Krathwohl, D.R. (2001). A Taxonomy of Learning, Teaching, and
Assessing: A Revision of Bloom’s Taxonomy of Educational Objectives. New York:
Longman.

PE-127
 Edi Istiyono / The Development of … ISBN 978 – 979 – 99314 – 8 – 1

Barnett, J. E & Francis, A.L. (2012). Using higher order thinking questionsto foster critical
thinking: a classroom study. Educational Psychology: An International Journal of
Experimental Educational Psychology. http://www.tandfonline.com/loi/cedp20. Diakses
tanggal 10 Desember 2012
Bloom, B.S., et al. (1979). Taxonomy of Educational Objectives: Handbook I Cognitive
Domain. London:Longmans Group Ltd.
Brookhart, S. M. (2010). How to Assess Higher Order Thinking Skills in Your Classroom.
Alexandria: ASCD
BSNP. (2006). Standar Kompetensi dan Kompetensi Dasar Mata Pelajaran Fisika Untuk SMA
dan MA. Jakarta: BSNP-Depdiknas
Depdiknas. (2003). Undang-undang No. 20 tahun 2003 tentang Sistem Pendidikan Nasional
---------------. (2007). Peraturan Menteri Pendidikan Nasional No 20 Tahun 2007 tentang
Standar Penilaian
Haladyna, T. M. (1997). Writing Test Item to Evaluate Higher Order Thinking. Boston: Allyn
and Bacon
Haladyna, T. M. (2004). Devoping and Validating Multiple Choise Test Items. New Jersey:
Lawrence Erlbaum Associates, Inc.
Hambleton & Swaminathan (1991): Fundamentals of Item Response Theory. Los Angeles:
SAGE Publications, Inc
Mchrens, W.A & Lehman, I. J. (1991). Measuement and Evaluation in Education and
Psycology. Belmont: Wadsworth
Sudjana, N. (1990). Penilaian Hasil Blajar Mengajar. Bandung: PT Remaja Rosdakarya
Oriondo, L.L.and Dallo-Antonio, E.M. (1998). Evaluation Educational Outcomes. Manila: Rex
Printing Compagny, inc
Piaget, J. (2005). The psychology of intellegence [Versi elektronik]. Taylor & Francis e-Library.
Reedal, K.E. (2010). Jean Piaget’s Cognitive Development Theory in Mathematics Education.
Department of Mathematics and Computer Science – Ripon College. Summation, May
2010, pp. 16-20 http://ripon.edu/macs/summation.
Schraw , G, & Robinson, D.H. (2011). Assessment of Higher Order Thinking Skills. New York:
Information Age Publishing, Inc
TIMSS & PIRLS International Study Center. (2012). TIMSS 2011 international results in
science. Boston: The TIMSS & PIRLS International Study Center, Boston College.
Diambil tanggal 5 Januari 2013, dari http: timss.bc.edu/timss2011/release.html
Van der Linden, Wim J & Hambleton, Ronald K. (1997). Handbook of Modern Item Response
Theory . New York: Springer-Verlag New York, Inc
Widhiarso, W. (2010). Model politomi dalam teori respons butir. Yogyakarta: Psikologi UGM
Wright, B.D. & Masters, G.N. (1982). Rating scale analysis. Chicago: Mesa Press.

PE-128