Paper ID #6214

Combining In-class Design Problems and EFFECTs to Stimulate Critical

Thinking Skills
Dr. Nathan Huynh, University of South Carolina
Dr. Juan M Caicedo, University of South Carolina

Dr. Caicedo is currently an associate professor at the Department of Civil and Environmental Engineering
at the University of South Carolina. He obtained his doctorate degree from Washington University in St.
Louis in 2003. Dr. Caicedos research interests include engineering education, numerical and experimen-
tal research in the areas of structural dynamics, model updating, structural health monitoring, earthquake
engineering and structural control. Dr. Caicedo is member of the American Society of Civil Engineers,
the George E. Brown Jr. Network for Earthquake Engineering Simulation, the Society of Experimental
Mechanics and the American Society of Engineering Educators.

Dr. Charles E. Pierce, University of South Carolina

Jason W Gantt, University of South Carolina

American
c Society for Engineering Education, 2013
 Combining In-Class Design Problems and EFFECTs to
Stimulate Critical Thinking Skills
Abstract

In this study, we explore how to use the In-Class Design Problems (ICDP) to not only help
students learn the class material, but also improve their critical thinking skills. To accomplish
this, we use some key elements from the Environments for Fostering Effective Critical Thinking
(EFFECTs) pedagogy to get students to guesstimate an answer to a design problem and to reflect
on how the learned material helped them answer the design problem. In order to facilitate core
knowledge and critical thinking, a necessary and preliminary task is for the instructor to identify
students' misconceptions about a topic and use that information to guide students to the proper
solution. To this end, this paper proposes a novel approach for identifying students'
misconceptions, using word cloud and Bayes theorem. Preliminary results using one of the
developed ICDPs for an undergraduate Transportation Engineering course indicate that this is a
promising approach.

Introduction

Engineering judgment is generally regarded as a critical attribute of successful engineers. It is

generally developed through experience. Engineering judgment goes beyond the development of
a solution. It is a product of critical thinking regarding the appropriateness of the solution.
However, engineering judgment is not a tangible concept with clearly defined components or
procedures that can be easily taught.

Currently, most engineering courses provide students with core knowledge and technical skills.
Specifically, engineering students are generally instructed about the procedures for solving
problems. To this end, students tend to learn and memorize the procedures, without critically
thinking about the procedures themselves and final solutions. For this reason, engineering
instructors often strive to not only provide fundamental knowledge in a course but also good
engineering judgment1.

This study modifies the Environments for Fostering Effective Critical Thinking (EFFECT)
instructional approach to improve students' critical thinking skills and engineering judgment.
The EFFECT instructional approach is an active learning framework that is problem-based,
collaborative, and includes active learning components that are designed to facilitate critical
thinking in students2. Each EFFECT is centered on a driving question that is designed to elicit
critical thinking skills and enhance the transfer of core knowledge through active learning. An
outline of the EFFECT components is provided in Figure 1.
 Figure 1. EFFECTs pedagocial structure.

As shown in Figure 1, each EFFECT begins with a decision worksheet. The decision worksheet
provides a brief context about the problem and asks students to provide an answer to a design
problem. For example, the context to a problem concerning the design of a solar power system3
could be as follows. A complete decision worksheet is provided in Appendix 1; it is one of many
that have been developed as part of the NSF CCLI project and is publicly accessible.

"A highway rest-stop between Battle Mountain, NV and Winnemucca, NV is

being upgraded to include restroom facilities and vending machines. Because of
its remote location, electricity is not available from the grid, and solar power is
being explored as an alternative. "

The question that the students may be asked to answer is:

"How much will it cost to install a solar power system at the renovated rest-stop?"

The decision worksheet typically also includes a few supporting questions designed to help
students visualize the problem, think about the data they would need to collect, methods and
experiments they would need to conduct in order to answer the driving question. For the solar
power EFFECT, the support questions could be:

Draw a sketch depicting the system as it operates at the highway rest stop. Label the
system components.
List the factors that will determine the size of system components
What information would you need to gather in order to provide a reasonably good
estimate of the cost of the system?
How could you gather this information?
Identify some things that your partner thought of that you initially did not, and briefly
explain why they are important.
After completing an individual decision worksheet, students discuss their answers in small
groups and then with the entire class. In subsequent class periods, there are demonstrations and
hands-on exercises, followed by journal questions that are designed to bridge knowledge gaps to
help students answer the driving question. For the solar power EFFECT, the hands-on exercises
could involve:

Developing an estimate of the electric demands of the system

Calculating the efficiency of an inverter at varying fan speeds and heat settings
Estimating the necessary battery weight and capacity for the determined demand
Identifying factors that influence electrical demands of the solar panel, data sources that
are used to calculate the solar panel surface, and the solar panel area required to collect a
specific kilowatt per day.

The EFFECT culminates in a final student design report that utilizes all of the experimential
learning, core knowledge, and engineering judgment that have been cultivated during the
EFFECT process. The final design could ask students to: (1) estimate demand, (2) select an
inverter based on demand estimation, (3) identify batteries needed, and (4) determine total
surface area of Solar Panels needed. A complete final design requirement for this EFFECT is
provided in Appendix 2.

As illustrated in Figure 2, the outcome of the application of the EFFECT is that students have
received core knowledge, which leads to the development of fundamental technical skills that
can be applied to solve design problems. Through this process students develop good
Engineering Judgment the capacity to apply critical thinking to assess the rationality,
practicality, reasonableness, and correctness of possible solutions to problems3. In this way, the
EFFECTs pedagogy provides a method for instructors to effectively teach both core knowledge
and critical thinking skills.

Figure 2. Relationship between design and analysis process, critical thinking,

and engineering judgement.
In this study, we explore how to effectively use the In-Class Design Problems (ICDP) that is to
be done by students at the end of each major topic (i.e. chapters). Specifically, we would like to
use the ICDPs to not only help students learn the class material, but also improve their critical
thinking skills. To accomplish this, we use some key elements from the EFFECTs pedagogy to
get students to think and guesstimate an answer to a design problem and to reflect on how the
learned material helped them answer the design problem. The study is carried out in an
undergraduate Transportation Engineering course with 57 students. The remainder of this paper
discusses our modified EFFECT approach and a novel approach for assessing students' learning.
Preliminary results are presented and discussed.

Modified EFFECT Approach

The specific structure of the modified EFFECT approach is as follows. During the first lecture
of each of the seven topics, the instructor provides students with a simplified EFFECT decision
worksheet that simply asks students to make an educated guess of the answer to a design
problem and to list the parameters they considered in estimating their answers. For example, one
of the driving questions is estimate the theoretical stopping distance of a Honda civic with and
without antilock brakes from 60 mi/hr on good, dry, level pavement. The purpose of the
decision worksheet is to pique the students' interest and to let them know what it is they will be
learning to solve in the next few classes. The students are given time to answer the question on
their own and then additional time to discuss the answer with classmates, in groups of 4 to 5
students. Each student records his/her answer as well as the groups answer. This is followed by
lectures and homework in a series of subsequent classes. In the homework, students are asked
how the topics covered in class and which homework problems can help them answer the driving
question. The purpose of this requirement is to make students reflect on the learned material and
to remind them of the big picture (i.e. the design problem). Finally, students are given an ICDP
which has the same design problem as the one presented in the decision worksheet. In solving
the ICDP, students discuss the problem among themselves within the group of 4 to 5 students
and across groups. While they do this, the instructor provides assistance as needed. At the end
of the ICDP, students are asked to state how their final answers have changed from their initial
estimates and explain why. This is done to engage students in metacognition (i.e. assess their
own learning). Of particular interest to the instructor is the students' awareness of their own
level of knowledge and thought process.

The differences between the EFFECT approach used in this study and a typical EFFECT like the
solar power EFFECT discussed previously are: (1) decision worksheets do not include as many
supporting questions; (2) there is only one journal entry as opposed to several; (3) there are no
hands-on learning modules, just lectures with example problems; and (4) there is no final design
problem and report, the ICDP is used instead. This modified EFFECT approach takes fewer
classes (from decision worksheet to ICDP) and thus can be used in a transportation engineering
class that typically covers about seven topics.

Appendix 3 provides a summary of the topics and decision worksheets developed for an
undergraduate Transportation Engineering course with 57 students. They can also be found
online at: http://sdii.ce.sc.edu/effects. These driving questions were developed based on the
teaching material that accompanied the Mannering and Washburn textbook: Highway
Engineering and Traffic Analysis4.

Analysis of Decision Worksheet Response

In the developed decision worksheets, students were asked to not only estimate the solution but
also provide a list of parameters they considered in their estimation. By analyzing the students'
lists of factors, the instructor can identify misconceptions and incorrect use of engineering
terminologies. Equipped with this information, the instructor has the opportunity to use it in a
way that effectively guides the students to the proper solution approach. The ability to quickly
identify students' misconceptions is thus an important step in the process of facilitating core
knowledge and critical thinking. In this paper, we propose the use of word clouds to help
instructors quickly visualize the student responses. Furthermore, we propose the use of Bayes
inference to estimate the probability that certain words would be included in the response.

The frequency of keywords that the students specified for a particular answer provides only
partial evidence of their understanding. That is, some students may not understand the design
problem fully, but know of terminologies used in a similar context. On the other hand, some
students may understand the design problem, but used the wrong terminologies. To identify
students' misconceptions, Bayes inference could be used by considering both the students'
keywords and the instructors expectation of the keywords. We surmise that the word cloud
showing the posterior probability density function (after including the instructors expectation of
the student knowledge) is a better method than the word cloud showing word frequency in
identifying students' misconceptions. Formally, the probability of randomly selecting a
particular word is:

| |

Where is the prior Probability Density Function (PDF) and it indicates the belief of the
instructor on the probability of a particular word. This probability is based on previous
interaction between the instructor and the students. | is the
likelihood, | is the posterior probability that combines the
information provided by the instructor and the student responses. The parameter is a
normalization constant5.

Preliminary Results

Figure 3 shows the word cloud of the student responses to the driving question, "Suppose we
want to construct a road with the following characteristics and that the speed limit on this road is
70 mi/hr. What would be an appropriate radius to allow vehicles to travel on it safely?"
Additional information about the driving question can be found in the Appendix. The word
cloud was created using http://www.wordle.com. The more times a word is used, the bigger it
appears in the word cloud. Thus, the words Elevation and Speed were the two most used
words in the student responses. As expected, other engineering key terminologies such as
superelevation, grade and SSD (stopping sight distance) were used less often.
 Figure 3. Word cloud of student response from decision worksheet.

Figure 4 shows the word cloud of the posterior probability obtained using Bayes inference. It
can be seen that the word speed gained importance (bigger) while the word elevation
became less important (smaller) with respect to the instructor's prior probability. Figure 4 better
captures how well the students understand the topic within the context of the EFFECT. It can be
used to identify misconceptions from students. For example, it can be inferred that the word
elevation is misunderstood or misused in this particular context because it is prominent
(bigger) in the word cloud in Figure 3, but not Figure 4. That is because this parameter is not
expected to play a key role in the solution. Conversely, "friction" is less prominent (smaller) in
Figure 3, but is more prominent in Figure 4, which suggests that this term is not understood by
students and that the instructor should subsequently explain how friction plays a key role in the
solution.
 Figure 4. Word cloud of the posterior probability of each word.

Summary and Conclusion

This study investigates the effectiveness of using ICDPs that are combined with key elements
from the EFFECT pedagogical approach to stimulate and improve critical thinking skills of civil
engineering students. In order to facilitate core knowledge and critical thinking, a necessary and
preliminary task is for the instructor to identify students' misconceptions about a topic and use
that information to guide students to the proper solution. To this end, this paper proposes a
novel approach for identifying students' misconceptions, using word cloud and Bayes inference.
Preliminary results using one of the developed ICDPs indicate that this is a promising approach.

Acknowledgment

This work is supported by the National Science Foundation under Grant No. 1022971.

Bibliography

1. Van Mier, J. G. M., & van Vliet, M. R. A. (1999). Experimentation, numerical simulation and the role of
engineering judgement in the fracture mechanics of concrete and concrete structures. Construction and Building
Materials, 13, 3-14.

2. J. M. Caicedo, J. Flora, C. Pierce, A. Nichols, W. Graf, and B. Timmerman. Introducing the

environments for fostering effective critical thinking (effects). Proceedings of the International
Modal Analysis Conference, Orlando, FL, February 4-8, 2008, 2008.

3. I.W. Wait. Solar Power System Design to Promote Critical Thinking in Freshman Engineering Students.
Proceedings of the American Society of Engineering Education, San Antonio, TX, June 10-13, 2012.
4. F.L. Mannering and S.S. Washburn. Principles of Highway Engineering and Traffic Analysis, 5th Edition. John
Wiley & Sons, 2012.

5. A.H-S. Ang and W.H. Tang. Probability Concepts in Engineering Planning and Design, Vol. 1. Wiley, New
York, 1975.

Appendix 1: Solar Power EFFECT Decision Worksheet

A highway rest-stop between Battle Mountain, NV and Winnemucca, NV is being upgraded to

include restroom facilities and vending machines. Because of its remote location, electricity is
not available from the grid, and solar power is being explored as an alternative.

Driving Question: How much will it cost to install a solar power system at the renovated rest-
stop?

Supporting Questions
Draw a sketch depicting the system as it operates at the highway rest stop. Label the
system components.
List the factors that will determine the size of system components
What information would you need to gather in order to provide a reasonably good
estimate of the cost of the system?
How could you gather this information?
Identify some things that your partner thought of that you initially did not, and briefly
explain why they are important.
Appendix 2: Solar Power EFFECT Final Design

Fill in the blanks and submit calculations that address the following sections:

1. Demand Estimation
Based on the devices that you believe will be used in the rest stop, revise your Demand
Estimation calculations in order to take into account the principles that have been illustrated in
the active learning exercises, and the corrections that you previously identified as necessary.

Attach calculations and a summary of your demand estimation, and fill in the blanks below.
Estimated peak electrical requirements for the rest stop: _____________ Watts
Estimated power demand for the rest stop: _____________ kWh per day

2. Inverter

Based on the peak electrical requirements identified above, go to the following website and
select an inverter to be used at the rest stop. http://www.solarelectricsupply.com/inverters.html

Fill in the blanks below, attach a printout of the specifications of the inverter that you have
selected, and attach a printout of your numerical model of cost (explained below).
Inverter selected Size / Capacity: ___________________
Manufacturer: _______________________ Model #: ______________________
Efficiency: _________________

Estimate the cost of this inverter by creating a linear-regression model in Excel with the
following inverter price data:
Cost
Watts ($) Manufacturer
5000 4100 SMA
2000 1680 Magnum
8000 4676 Radian
2000 1277 Exceltech
3000 1188 Samlex
300 241 Morningstar

Estimated Cost of Inverter: ___________________________

3. Batteries

Based on the power demand identified in Part 1, the efficiency of the inverter, the batteries
available on this website (http://www.solar-electric.com/batteries.html) and any other factors that
you deem important (e.g., extra power storage for cloudy days), fill in the following:
Total battery capacity required (Amp-Hr @ 12 Volts): ________________________
Battery manufacturer & model selected: _______________________________________
Number of Batteries Required: ____________________________
Total Cost of Batteries: _________________________________
4. Solar Panels

Based on the estimated power demand for the rest stop (i.e., kWh required per day), the
assumption of 15% efficiency for solar panels, and the solar irradiance data provided below,
determine the total surface area of solar panels required for the highway rest stop.
Irradiance
Time (W/m2)
7:00 0.0
8:00 14.6
9:00 86.1
10:00 137.1
11:00 182.7
12:00 201.1
13:00 255.9
14:00 293.7
15:00 302.7
16:00 203.7
17:00 0.0

Total surface area of solar panels required (m2): ____________________
Assuming $375 per m2, total cost of solar panels: ___________________

Summary: How much will it cost to install a solar power system at this rest stop? ____________

Appendix 3: Transportation EFFECTs Decision Worksheets

Horizontal Curve

Suppose we want to construct a road with the following characteristics and that the speed limit
on this road is 70 mi/hr. What would be an appropriate radius to allow vehicles to travel on it
safely?

PT

e = 6% Elev. = 194'
G=0%
PC
STA = 5+00 R=?

Elev. = 74'
G=0% Plan (aerial) View

Traffic Flow Queuing

The entry plaza to a new sea port is being designed. It is expected that 230 semi-trucks (hauling
cargo containers) will arrive during the peak hour of demand. The mean service rate at each
entry booth is such that one truck can be processed every 40 seconds.

What is the number of booths that would need to be open to keep the average time waiting in the
queue less than 1 minute? Assume that both arrival and departure headways are exponentially
distributed.

Level of Service

A new segment of freeway is being built to connect two existing parallel freeway facilities. The
following traffic and roadway characteristics are expected:

Traffic Characteristics
AADT = 86,500 veh/day
K = 12%
D = 56%
PHF = 0.92
6% trucks and buses
2% RVs
Primarily commuters

Roadway Characteristics
Grade in peak direction: 1.5 miles, 2.25 percent
Total ramp density = 1.25 per mile
Lane widths = 12 ft
Shoulder widths = 4 ft

How many lanes are necessary to ensure that this new freeway segment will operate at no worse
than LOS D during the peak hour in the peak direction?

Traffic Signal Design

Consider the following intersection and characteristics.

 Assume the following:
Saturation flow rate per lane = 1800 veh/h (for protected movements)
Saturation flow rate per lane = 700 veh/h (for any lane with permitted movements)
Start-up lost time/phase: 2 sec; Clearance lost time/phase = 2 sec
Lane widths:
o Thru: 12 ft
o Left turn: 11 ft
Speeds:
o E/W: 45 mph
o N/S: 35 mph
Vehicle length: 20 ft
Level grade
Eff. crosswalk width = 8 ft
# of crossing pedestrians/phase = 15

What is the cycle length that would minimize the intersection delay?

Traffic Assignment User Equilibrium

There are three potential routes from a housing area to the proposed theater site. These routes
have the following speed and length characteristics:

Route Free-Flow Speed (mi/h) Length (mi)

1 45 5.5
2 40 3.5
3 35 3.75

It is known that the individual route travel times increase (in units of minutes) according to the
following functions (with x in units of 1000 vehicles per hour):
 Travel Time increase,
Route as a function of traffic volume
1 0.5x1
2 x2
3 0.25x3

If a total of 5500 vehicles go from the origin to destination, estimate how many vehicles will be
using Route 1, 2, and 3.