Active and Cooperative Learning in an Introductory Chemical Engineering Course
David DiBiasio
Department of Chemical Engineering
Worcester Polytechnic Institute
James E. Groccia
Center for Curricular Innovation and Educational Development
Worcester Polytechnic Institute
Abstract:
A sophomore level chemical engineering course was redesigned to emphasize
active and cooperative learning. The structure used was a peer-assisted
cooperative learning model developed at WPI. The experimental course was
compared to a control course taught by the passive lecture method. The
control and test courses were compared using student performance, attitudes,
evaluations of the course and instructor, and faculty time. We found that
student performance was better and content coverage was increased in the
test class. Faculty time was reduced by 24%using the peer assisted cooperative
learning model. Composite student evaluations of the course and instructor
increased slightly from the control to the test course. Student attitudes
about the profession increased during the test course, but were mixed regarding
working in teams.
Many engineering institutions are initiating new approaches to education,
with the hope of increasing teaching effectiveness and student learning.
We are asked to put more technical material into an already crowded curriculum,
increase faculty productivity, and increase student retention. We are
also concerned with increasing students' problem solving abilities, their
ability to work in teams, and their ability to integrate knowledge from
various courses.
Although the WPI undergraduate program is heavily project based at the
upper levels, the introductory courses, particularly in engineering are
typically taught by the passive, lecture format. It is generally thought
that active and cooperative learning are better than the traditional lecture
methods used successfully for so long in engineering education. WPI's
junior and senior years have been based upon this premise since 1970.
A significant portion of each of those year's courses is replaced by team-based,
extended project work.
Since 1993, a grant from the Davis Education Foundation has supported
an extensive curriculum revision project in first and second year courses[1].
The project evaluates the effectiveness of a novel cooperative learning
methodology based upon peer-assisted group monitoring. Outstanding upperclass
students are used as Peer Learning Assistants (PLA) whose role is to assist
in group processing, promote group efficiency, monitor group progress,
and mediate conflict resolution. These are tasks that can be easily accomplished
by a trained upperclass student, and are a great time sink for faculty.The
result is that students take more responsibility for their learning, they
cooperate rather than compete, they are actively involved, and faculty
contact time may be reduced.
Guided by the hypothesis that active and cooperative learning is better
than passive, competitive learning (2), we completely restructured a sophomore
level chemical engineering (CM) course using active and peer-assisted cooperative
strategies. We evaluated the impact on student attitudes and performance,
and on faculty productivity. The course was redesigned based upon the
WPI-Davis-PLA model. Attention was focused on addressing the major learning
styles of engineering students by varying the teaching style. Students
were actively engaged in every class session with one-minute papers, think-pair-share
activities, informal problem solving, demonstrations, and class discussions.
Specifically, our goals were to:
- Improve student problem solving skills in four of the six
cognitive domains described by Bloom, et al [3]: knowledge, comprehension,
application and analysis.
- Maintain or improve student performance relative to mastering
the technical material in the course.
- Improve student attitudes regarding the course, the instructor,
chemical engineering in general, confidence in problem solving,
and ability to work in a team.
- Increase faculty productivity, as measured by contact time
per course.
WPI courses are taught in a seven week term format. Students take three courses during one of these terms. Students normally meet for a
course 4-5 times per week, and the professor may be involved in more meetings
depending on the size of the class and the number of sections. Class length
is 50 minutes. There are four terms in an academic year. The course examined
in this study is ``Introduction to Chemical Engineering'' and it is fourth
in the initial CM sequence.
A control course (55 students) was taught in the spring
of 1994, and used for comparison against the experimental class. The control
was taught by the traditional lecture method with six class meetings per
week. Four of these were lectures and the others were problem solving
sessions (conferences). There were seven out-of-class homework assignments
and one laboratory exercise. Although the lab assignment was done in groups
it was not conducted in a formal cooperative learning mode. There was
no group work other than the lab. The course was graded using a norm-referenced
system, with the class average being a C grade.
The experimental course (71 students) was taught
in the spring of 1995. With the exception of the teaching method, variations
from the control class were minimized. The course was taught by the same
professor, at the same time of day, and in the same room as the control.
It met six times per week. Three of these meetings were ``lecture'' sessions,
two were problem solving sessions, and the sixth involved formal group
processing sessions. There were seven homework assignments, and three
group project assignments. The ``lecture'' sessions used a variety of
student involvement techniques (see above).
There were three formal cooperative learning assignments. The first
was a short (3 day) chemical process flowsheet evaluation. The second
was a moderate length (2.5 week) analysis of a chemical process accident.
Groups turned in a written report with detailed calculations for grading.
The third group project involved two stages, completed over 3-4 weeks.
The first deliverable was a pre-lab report outlining the background and
goals of the lab exercise. The second was a formal report containing the
lab results and analysis. The assignment involved an energy balance and
efficiency analysis of a pilot scale distillation column. Final reports
had a written analysis and detailed calculations, including the design
of a plant scale distillation column using the lab data. Projects were
graded on technical content and report format and quality. The course
was graded on a criteria-referenced system ignoring the class average.
The experimental course contained 14 groups of 5
students per group. To facilitate group processing, improve group efficiency,
and reduce faculty time involvment in group dynamics, the PLA structure
was used. The PLA's were each assigned 2-3 groups from the experimental
class. PLA's met at least once per week with each group and supervised
much of the laboratory activities. Graduate Teaching Assistants were not
used to supervise groups, but were used to grade homework and exams. The
same teaching assistants were used to grade both the control and the test
classes. The instructor graded all group project reports.
The assessment tools we used covered student performance,
student attitudes, student evaluation of the course and instructor, and
faculty time.
Student performance was measured by grade distribution, exam scores,
quality of lab reports, and attrition. Three, 1.5 hour, individual, inclass
exams were given to both classes. The same exams were used for the control
and the experimental course. Security was maintained from 1994 to 1995
by retaining all exams from the control group. Students were allowed to
view their own exam and discuss it with the professor but no exams left
the office. Anecdotal evidence suggested that this system was secure.
Each exam contained problems covering the first four levels of Bloom's
taxonomy: knowledge, comprehension, application and analysis. Performance
on each type was compared for each year.
The standard WPI Student Course Evaluation (SCE) form was used to compare
student assessment of the course from the control to the experimental version.
The form contains questions evaluating specific student perceptions of
the instructor, the course, and the textbook. A standard analysis of variance
was done on this data using the SPSS software package.
Student attitudes about chemical engineering, cooperative learning,
and learning methods were examined at the beginning and end of the experimental
course using a survey constructed by the Center for Curricular Innovation
and Educational Development (CCIED) at WPI. Faculty time was tracked by
the instructor for both course offerings.
Analysis of the final grade distribution for both
courses showed that the experimental group performed slightly better.
There was a small decrease in the percentage of students receiving C's
or lower, from the control to the test class. There was a small increase
in the percentage of students receiving A's and B's from the control to
the test class. Most of this was due to an increase in B grades for the
test group. Retention for the test group was better than the control.
Only 4%of the test group did not show for the third (final) exam, while
7%of the control did not show for that exam. All of the students that
did not show for the third exam in the test course, completed the final
project with their group. This was despite knowing they would receive
a failing grade in the course. None of the control dropouts did that.
It appears that these students in the test class felt some obligation
to their group that was not evident in control course. The overall exam
performance is shown in Table 1. These results show that the test group
scored as well as the control on the first two exams and better than the
control on the third (and hardest) exam. Further examination of the distribution
of scores on each exam showed that the test group had the same percentage
of scores above 90 as the control for the first exam. There was a 4%increase
in scores above 90 for the test group on the second exam. On the third
exam the control group had 9%score above 80 (only 2%, one student, scored
above 90) while the test group had 5 students (7%) score above 90 and
a total of 19%score above 80. This showed a substantial improvement
for the test group in handling the most difficult material encountered
near the end of the course. The analysis of the individual problems from
each exam (classed according to Bloom's first four levels) for both groups
showed that there was no trend indicating that the test class could handle
any one type better than the control group. The improvement for the third
exam came largely from better performance from the test group on knowledge
and application problems. The instructor's evaluation of the lab project
reports showed that they were better, overall, in the test class.
A pre- and post-course survey was
done on the test class to assess student attitudes about chemical engineering,
cooperative learning, and perceptions about the course. Space does not
allow inclusion of all the survey questions. However, questions can be
grouped in three broad categories: course/profession perceptions, attitudes
toward group work, and perceptions about problem solving and learning.
Questions concerning course and profession perceptions all showed increases
from the pre- to post-survey. For example, 85%of the class agreeed that
``CM is an exciting subject to study'' prior to the course, while 94%agreeed after the course. Attitudes toward group work were mixed. There
was a 10%decrease in students agreeing with the statement ``The best
chemical engineering work is done in teams'', yet there was a 5%increase
in those who said they liked group work. Regarding perceptions about learning,
there was an increase (pre- to post-survey) in those who preferred lecture,
and in those who preferred to work alone. This apparantly contradicted
the increase in those liking group work. It is our feeling that these
results are somewhat consistent with other findings that the introduction
of new cooperative teaching methodologies generally results in some student
discomfort [4,5].
The SCE form contained 24 questions. Fourteen
relate to the instructor and how s/he conducted the course, three relate
to laboratory courses, and seven cover other items such as the textbook,
the room, and a self-assessment of how much was learned. Responses are
in the form of ``strongly agree, agree, disagree, strongly disagree''.
Although this course had a laboratory project, it was not a lab course
and those three questions were considered irrelevant. We conducted a standard
analysis of variance on the remaining 21 responses. Space limitiations
prevent a detailed presentation of each question. The analysis showed
that there was no statistical difference between the test and control class
on most questions when we examined total positive responses compared to
total negative responses. A composite of the 14 questions regarding the
instructor resulted in a small but statistically significant (p<0.05)
increase in positive responses from the control to the test class.
Faculty time was evaluated on the basis of contact hours
with students, since preparation time and grading were essentially equivalent
for either type of course. Contact hours involved in-class time, office
hour contacts, and laboratory time.
Office hour contacts were unchanged between the test class and the control.
This meant that group dynamic problems, that may have resulted in an increase
in faculty time, were adequately handled by the PLA's. This was confirmed
by feedback from the PLA's.
In-class time was reduced by 17%. This was a result of the replacement
of a lecture period with a group processing meeting run by the PLA's.
This reduction did not result in any loss of content in course coverage.
The text material, homework, and exam converage was the same for each
year, In fact, the test group covered more material in greater depth than
the control class. This was a result of the additional project assignments.
The projects required students to understand topics not covered in the
control course: chemical process safety, flammability of gas mixtures,
flowsheet construction, distillation column diameter calculations, and
report writing.
The major time savings was in preparing groups for the lab project and
monitoring them during each of their lab periods. Laboratory work is not
normally a part of this course, and we have only one lab unit that is used
by only one group at a time. The instructor very strongly believes that
``hands-on'' work is essential to learning the course material. However,
in the past, the burden of running this lab exercise had been overwhelming.
Use of PLA's and formal cooperative learning groups in the test class
resulted in an 80%reduction in the professor's time for this part of
the course. A big advantage of using PLA's here is that they have recently
completed the course and have good knowledge of the distillation experiment.
Overall, the course format allowed a 24%reduction in contact hours.
Most of this came from reduction of repetitive lab oriented hours, previously
handled by the professor.
Based upon the assessment of the experimental course, we have reached
the following conclusions relative to our goals:
- The was no change in problem solving skills of students at
the first four levels of Bloom's taxonomy, as measured by exam
performance.
- Student performance was better in the expeerimental course.
This was determined by exam performance, final grades, attrition,
project quality and content coverage.
- Student course evaluations showed a small improvement for
the test course, on a composite basis. Attitude changes during
the test course were mixed.
- A 24%increase in faculty productivity, measured by contact
hours, was realized.
- Groccia, J.E. Increasing Educational Quality and Faculty Productivity Through Cooperative and Peer Assisted Learning, ASEE Conference
Proc., Anaheim, CA., 1520-1524, 1995.
- Ercolano, V., Learning Through Cooperation, ASEE Prism, 26-29, Nov., 1994.
- Bloom, B.S., et al. Taxonomy of Educational Objectives: The Classification of Educational Objectives. Handbook I: Cognitive Domain, D. McKay,
New York, 1956.
- Miller, J.E., J. Wilkes, R.D. Cheetham, and L. Goodwin, Tradeoffs in Student Satisfaction: Is the Perfect Course an Illusion?, J.
Excell. College Teaching, 4, 27-47, 1993.
- Felder, R.M., We Never Said it Would be Easy, Chemical Engr.
Educ., 1, 32-33, 1995.