Pedagogies of Engagement:
Classroom-Based Practices
KARL A. SMITH
Department of Civil Engineering
University of Minnesota
SHERI D. SHEPPARD
Department of Mechanical Engineering
Stanford University
DAVID W. JOHNSON
Department of Educational Psychology
ROGER T. JOHNSON
Department of Curriculum and Instruction
University of Minnesota
ABSTRACT
Educators, researchers, and policy makers have advocated student
involvement for some time as an essential aspect of meaningful
learning. In the past twenty years engineering educators have
implemented several means of better engaging their undergraduate
students, including active and cooperative learning, learning
communities, service learning, cooperative education, inquiry and
problem-based learning, and team projects. This paper focuses on
classroom-based pedagogies of engagement, particularly
cooperative and problem-based learning. It includes a brief
history, theoretical roots, research support, summary of practices,
and suggestions for redesigning engineering classes and programs
to include more student engagement. The paper also lays out the
research ahead for advancing pedagogies aimed at more fully
enhancing students’ involvement in their learning.
Keywords: cooperative learning, problem-based learning, student
engagement
I. INTRODUCTION TO THE PEDAGOGIES
OF
ENGAGEMENT
Russ Edgerton introduced the term “pedagogies of engage-
ment” in his 2001 Education White Paper [1], in which he
reflected on the projects on higher education funded by the Pew
Charitable Trusts. He wrote:
“Throughout the whole enterprise, the core issue, in my view, is the mode
of teaching and learning that is practiced. Learning ‘about’ things does
not enable students to acquire the abilities and understanding they will
need for the twenty-first century. We need new pedagogies of engagement
that will turn out the kinds of resourceful, engaged workers and citizens
that America now requires.”
Prior to Edgerton’s paper, the widely distributed and influential
publication called The Seven Principles for Good Practice in Undergraduate
Education
[2] stressed pedagogies of engagement in concept. Three
of the principles speak directly to pedagogies of engagement,
namely, that good practice encourages student-faculty contact, co-
operation among students, and active learning.
More recently, the project titled The National Survey of Stu-
dent Engagement (NSSE) [3] deepens our understanding of how
students perceive classroom-based learning, in all its forms, as an el-
ement in the bigger issue of student engagement in their college ed-
ucation. The NSSE project conceives that student engagement is
not just a single course in a student’s academic career, but rather a
pattern of his or her involvement in a variety of activities. As such,
NSSE findings are a valuable assessment tool for colleges and uni-
versities to track how successful their academic practices are in en-
gaging their student bodies. The NSSE project is grounded in the
proposition that student engagement, the frequency with which
students participate in activities that represent effective educational
practice, is a meaningful proxy for collegiate quality and, therefore,
by extension, quality of education. For example, the annual survey
of freshmen and seniors asks students how often they have partici-
pated in, for example, projects that required integrating ideas or in-
formation from various sources, used e-mail to communicate with
an instructor, asked questions in class or contributed to class discus-
sions, received prompt feedback from faculty on their academic
performance, participated in community-based projects, or tutored
or taught other students. Student responses are organized around
five benchmarks:
1. Level of academic challenge: Schools encourage achievement
by setting high expectations and emphasizing importance of
student effort.
2. Active and collaborative learning: Students learn more when
intensely involved in educational process and are encouraged
to apply their knowledge in many situations.
3. Student-faculty interaction: Students able to learn from ex-
perts and faculty serve as role models and mentors.
4. Enriching educational experiences: Learning opportunities
inside and outside classroom (diversity, technology, collabo-
ration, internships, community service, capstones) enhance
learning.
5. Supportive campus environment: Students are motivated and
satisfied at schools that actively promote learning and stimu-
late social interaction.
Astin’s [4] large-scale correlational study of what matters in col-
lege (involving 27,064 students at 309 baccalaureate-granting insti-
tutions) found that two environmental factors were by far the most
predictive of positive change in college students’ academic develop-
ment, personal development, and satisfaction. These two factors—
interaction among students and interaction between faculty and
students—carried by far the largest weights and affected more gen-
eral education outcomes than any other environmental variables
January 2005 Journal of Engineering Education 1
studied, including the curriculum content factors. This result indi-
cates that how students approach their general education and how the
faculty actually deliver the curriculum is more important than the
formal curriculum, that is, the content, collection, and sequence of
courses.
The assessment study by Light [5, 6] of Harvard students
strongly suggests that one of the crucial factors in the educational
development of the undergraduate is the degree to which the stu-
dent is actively engaged or involved in the undergraduate experience;
this is consistent with Astin’s work [4]. Astin and Light’s research
studies suggest that curricular planning efforts will reap much
greater payoffs in terms of student outcomes if more emphasis is
placed on pedagogy and other features of the delivery system, as well as
on the broader interpersonal and institutional context in which
learning takes place.
Pascarella and Terenzini’s summary of twenty years of research
on the impact college has on student development further supports
the importance of student engagement:
“Perhaps the strongest conclusion that can be made is the least surprising.
Simply put, the greater the student’s involvement or engagement in acad-
emic work or in the academic experience of college, the greater his or her
level of knowledge acquisition and general cognitive development… If
the level of involvement were totally determined by individual student
motivation, interest, and ability, the above conclusion would be uninter-
esting as well as unsurprising. However, a substantial amount of evi-
dence indicates that there are instructional and programmatic interven-
tions that not only increase a student’s active engagement in learning and
academic work but also enhance knowledge acquisition and some dimen-
sions of both cognitive and psychosocial change” [7]
.
Macgregor, Cooper, Smith, and Robinson [8] provided a syn-
thesis of interviews conducted with forty-eight individuals teaching
undergraduate classes across the United States who are infusing
their large classes with small-group activities or are working explic-
itly to create student communities within large classes. The faculty
who were interviewed are working with classes of more than 100
students, and some are teaching substantially larger classes, in the
350 to 600 student range. The faculty practicing small-group learn-
ing in large classes provided extensive empirical and theoretical ra-
tionale for their practices. Their reasons clustered in the following
categories:
1. promoting cognitive elaboration;
2. enhancing critical thinking;
3. providing feedback;
4. promoting social and emotional development;
5. appreciating diversity; and
6. reducing student attrition.
Edgerton, in the aforementioned white paper, goes on to cite four
strands of pedagogical reform that are moving in the same broad
direction: problem-based learning, collaborative learning, service
learning, and undergraduate research. This paper looks at a class of
pedagogies of engagement, namely, those that are classroom-based.
We focus particularly on cooperative learning and on problem-based
learning.
In the next section we present definitions of the classroom-based
pedagogies of engagement that are used in engineering undergrad-
uate classrooms followed by a brief summary of their history (sec-
tion III). Next we provide the theoretical foundations and research
evidence for effectiveness (section IV), and offer model practices for
implementation (section V). The paper concludes by presenting
some unanswered questions about classroom-based pedagogies of
engagement for engineering in particular and pedagogies in general.
II. AN OVERVIEW
“To teach is to engage students in learning.” This quote, from
Education for Judgment by Christensen et al. [9], captures the essence
of the state of the art and practice of pedagogies of engagement.
The thesis of this book, and this paper, is that engaging students in
learning is principally the responsibility of the teacher, who be-
comes less an imparter of knowledge and more a designer and facili-
tator of learning experiences and opportunities. In other words, the
real challenge in college teaching is not covering the material for the
students; it’s uncovering the material with the students.
Consider the most common model of the classroom-based
teaching and learning process used in engineering education in the
past fifty years (and maybe currently?). This model, illustrated in
Figure 1(a), is a presentational model where, as one pundit quipped,
“the information passes from the notes of the professor to the notes
of the students without passing through the mind of either one.”
An alternative to the “pour it in” model is the “keep it flowing
around” model. This is shown in Figure 1(b) and illustrates that
the information passes not only from teacher to student, but also
from students to teacher and among the students. The model of
teaching and learning represented in Figure 1(b) emphasizes that
the simultaneous presence of interdependence and accountability
are essential to learning, and their presence is at the heart of a
student-engaged instructional approach.
The model of the teaching-learning process in Figure 1(b) is
predicated on cooperation—working together to accomplish shared
goals. Within cooperative activities individuals seek outcomes that
are beneficial to themselves and beneficial to all other group mem-
bers. Cooperative learning is the instructional use of small groups so
that students work together to maximize their own and each others’
learning [10, 11]. Carefully structured cooperative learning involves
people working in teams to accomplish a common goal, under con-
ditions that involve both
positive interdependence (all members must
cooperate to complete the task) and individual and group accountability
(each member individually as well as all members collectively ac-
countable for the work of the group). Astin [12] reported that 14
percent of engineering faculty and 27 percent of all faculty said they
used cooperative learning in most or all of their classes.
A common question is, “What is the difference between cooper-
ative and collaborative learning?” Both pedagogies are aimed at
“marshalling peer group influence to focus on intellectual and sub-
stantive concerns” [13]. Their primary difference is that cooperative
learning requires carefully structured individual accountability,
while collaborative does not. Numerous authors, such as Barkley,
Cross, and Major [14], use the term collaborative learning to refer
to predominantly cooperative learning research and practice. To try
to minimize confusion, we will use the term cooperative learning
throughout the current paper.
Problem-based learning (PBL) “is the learning that results from
the process of working toward the understanding or resolution of a
problem. The problem is encountered first in the learning process”
[15]. Barrows [16] identified six core features of PBL:
2 Journal of Engineering Education January 2005
Figure 1. Two models of the classroom-based teaching learning process, as drawn by Lila Smith in about 1975. (a) “Pour it in” model;
(b) “Keep it flowing” model.
Figure 2. Problem-based learning contrasted with Subject
based learning.
Learning is student-centered.
Learning occurs in small student groups.
Teachers are facilitators or guides.
Problems are the organizing focus and stimulus for
learning.
Problems are the vehicle for the development of clinical
problem-solving skills.
New information is acquired through self-directed learning.
The process of problem-based learning was illustrated by
Woods [17], who contrasted it with subject-based learning (Figure 2).
Problem-based learning is suitable for introductory sciences and en-
gineering classes (as it is for medicine, where it is currently used) be-
cause it helps students develop skills and confidence for formulating
problems they have never seen before. This is an important skill,
since few science, mathematics, or engineering graduates are paid to
formulate and solve problems that follow from the material presented
in the chapter or have a single “right” answer that one can find at the
end of a book. An example of a PBL problem, adapted from
Adams’ [18] “dangling from a wire problem,” is to “estimate the di-
ameter of the smallest steel wire that could suspend a typical Ameri-
can automobile.”
1
The largest-scale implementation of PBL in the United States in
undergraduate courses (including large introductory courses) is at the
University of Delaware in Newark, Delaware, where it is used in
many courses, including biology, biochemistry, chemistry, criminal
justice, education, international relations, marine studies, mathemat-
ics, nutrition/dietetics, physics, political science, and exercise science
[19, 20]. The initial PBL work at the University of Delaware was
supported by the National Science Foundation (NSF) and the Fund
for Improvement of Post-Secondary Education (FIPSE); more than
25 percent of the faculty have participated in weeklong formal work-
shops on PBL. Allen and Duch recently described their implementa-
tion of PBL problems for introductory biology [21].
Woods at McMaster University has described the university’s
implementation of PBL in engineering [17]. In the chemical engi-
neering program there, PBL is used as part of two courses: one topic
or problem in a junior-level course; and five topics in a senior-level
course [22]. PBL is used in a theme school program created at
McMaster University and in a junior-level civil engineering course
and a senior-level project course in geography. These are examples
of the use of small group, self-directed PBL where tutorless groups
of five to six students function effectively. The class sizes are in the
range thirty to fifty, with one or two instructors. The students con-
currently take conventional courses. Project-based learning, which
focuses on a project and typically a deliverable in the form of a re-
port or presentation, was emphasized in a recent publication on
project/problem-based learning at Aalborg University in Denmark
(all majors), Maastricht University in Maastricht , The Netherlands
(which implemented the McMaster PBL model in medicine in
1
Details of this example are available at www.ce.umn.edu/~smith. Many addi-
tional examples are available on the University of Delaware PBL Web site
www.udel.edu/pbl.
January 2005 Journal of Engineering Education 3
1974), and at universities in Australia. There is an excellent summary
of these programs in PBL Insight [23]. A comparison of problem-
based and project-based learning is available in Mills and Treagust
[24]. Project-based learning, which is often the basis for the senior
design courses in undergraduate engineering curriculum in the
United States, will not be further discussed in this paper; the reader
is referred to the work of Dym et al. [25].
Lest the reader think that the model of the teaching-learning
process illustrated in Figure 1(b) is a modern creation, consider the
long and rich history of the practical use of pedagogies of engage-
ment, especially classroom-based practices such as cooperative
learning and problem-based learning. Thousands of years ago the
Talmud stated that to understand the Talmud, one must have a learn-
ing partner. Confucius is typically credited with the Chinese
proverb “Tell me and I forget; show me and I remember; involve me
and I understand.” (However, Edgerton [1] and others attribute the
Lakota Sioux Indians). The Roman philosopher, Seneca, advocated
cooperative learning through such statements as, “Qui Docet Dis-
cet” (when you teach, you learn twice). J. Amos Comenius
(1592–1679) believed that students would benefit both by teaching
and by being taught by other students. In the late 1700s, J. Lancast-
er and A. Bell made extensive use of cooperative learning groups in
England and India, and the idea was brought to the America when
a Lancastrian school was opened in New York City in 1806 [26].
One of the more successful advocates of cooperative learning in
the United States was Colonel Francis Parker [27] in the late 1800s.
Parker started several schools and hosted many visitors to his
schools who in turn started or changed their own programs. In the
last three decades of the nineteenth century, Colonel Parker advo-
cated cooperative learning with enthusiasm, idealism, practicality,
and an intense devotion to freedom, democracy, and individuality
in the public schools. Following Parker, John Dewey promoted the
use of cooperative learning groups as part of his famous project
method in instruction [28]. John Dewey’s ideal school involved
a “thinking” curriculum aimed at deep understanding;
cooperative learning within communities of learners;
interdisciplinary and multidisciplinary curricula; and
projects, portfolios, and other “alternative assessments” that
challenged students to integrate ideas and demonstrate their
capabilities.
In the late 1930s, however, public schools began to emphasize
interpersonal competition and this view predominated for well over
forty years [29].
In the mid-1960s Johnson and Johnson began training K-12
teachers and a few post-secondary teachers how to use cooperative
learning at the University of Minnesota. The Cooperative Learning
Center at the University of Minnesota resulted from their efforts to
(a) synthesize existing knowledge concerning cooperative, competi-
tive, and individualistic efforts, (b) formulate theoretical models
concerning the nature of cooperation and its essential elements, (c)
conduct a systematic program of research to test the theorizing, (d)
translate the validated theory into a set of concrete strategies and
procedures for using cooperative learning, and (e) build and main-
tain a network of schools implementing cooperative strategies and
procedures throughout the world. From being relatively unknown
and unused in the 1960s, cooperative learning is now an accepted
and often the preferred instructional procedure at all levels of educa-
tion throughout the world in every subject area and from preschool
through graduate school and adult training programs [30].
Most of the work on developing and researching models of coop-
erative learning in the 1970s and 1980s focused on K-12 education.
For example, in the early 1970s DeVries and Edwards [31] at
Johns Hopkins University developed Teams-Games-Tourna-
ments (TGT) and the Sharans in Israel developed the group in-
vestigation procedure for cooperative learning groups [32]. In the
late 1970s Slavin and colleagues at Johns Hopkins University
extended DeVries and Edwards’ work by modifying TGT into
Student-Teams-Achievement-Divisions (STAD) and modifying
computer-assisted instruction into Team-Assisted Instruction
(TAI) [32]. Concurrently, Kagan [34] developed cooperative learn-
ing structures that involved detailed procedures, such as numbered
heads together. This was followed in the 1980s by Cohen develop-
ing a “complex instruction” version of cooperative learning [35, 36]
and Dansereau [37] developing a number of cooperative learning
scripts.
The 1980s and 1990s brought an expansion of cooperative
learning models into engineering. The concept of a cooperative
learning group was introduced to the engineering education com-
munity at the 1981 IEEE/ASEE Frontiers in Education (FIE)
conference in Rapid City, S.D. [38]. Goldstein also presented a
paper on cooperative learning at this conference and Goldstein
and Smith were subsequently invited to present a workshop
(probably the first) on cooperative learning at the 1982 FIE con-
ference. Also, in 1981 the first in a series of papers on cooperative
learning was published in
Engineering Education, “Structuring
learning goals to meet the goals of engineering education” [10].
In the mid-1990s the Foundation Coalition embraced the
cooperative learning approach, produced several one-page sum-
maries of concepts, and developed an extensive Web site on
Active/Cooperative Learning: Best Practices in Engineering
Education.
2
More recently, Millis and Cottell [39] adapted Kagan’s coopera-
tive learning structures for higher education faculty, and Johnson,
Johnson, and Smith began adapting the conceptual cooperative
learning model to higher education [40–42].
III. THEORY AND RESEARCH EVIDENCE
The underling precept of cooperative and problem-based learn-
ing is interdependence. The term interdependence was introduced by
Coleridge in 1822 and is defined, according to the
Oxford English
Dictionary
, as “The fact or condition of depending each upon the
other; mutual dependence.” Many of the early references to the
term, e.g., by Coleridge, Huxley, Spencer, were biology related.
Spencer introduced the “conception of [society] as having a natural
structure in which all its institutions, governmental, religious, in-
dustrial, commercial, etc., etc., are inter-dependently bound” (Ox-
ford English Dictionary
)
Research on cooperative learning has been guided primarily by
social interdependence theory. The theory was conceived of in the
early 1900s, when one of the founders of the Gestalt School of Psy-
chology, Kafka, proposed that groups were dynamic wholes in
which interdependence among members could vary. One of his col-
leagues, Lewin [43], refined Kafka’s notions in the 1920s and 1930s
while stating that (a) the essence of a group is the interdependence
2
See http://clte.asu.edu/active.
4 Journal of Engineering Education January 2005
among members (created by common goals) that results in the
group’s being a “dynamic whole” so that a change in the state of any
member or subgroup changes the state of all other member or sub-
group and (b) an intrinsic state of tension within group members
motivates movement toward the accomplishments of the desired
common goals. One of Lewin’s graduate students, Deutsch, formu-
lated the theory of cooperation and competition in the late 1940s
[44, 45]. One of Deutsch’s graduate students, D. Johnson (collabo-
rating with R. Johnson), extended Deutsch’s work into classroom
practices [46–48].
The social interdependence perspective assumes that the way so-
cial interdependence is structured determines how individuals in-
teract, which in turn determines outcomes. Positive interdepen-
dence (cooperation) results in promotive interaction as individuals
encourage and facilitate each other’s efforts to learn. Negative inter-
dependence (competition) typically results in oppositional interac-
tion as individuals discourage and obstruct each other’s efforts to
achieve. In the absence of interdependence (individualistic efforts),
there is no interaction as individuals work independently without
any interchange with each other [44].
Extensive research has been conducted on cooperative learn-
ing—defined in section II as the instructional use of small groups so
that students work together to maximize their own and each others’
learning. From 1897 to 1989 nearly 600 experimental and more
than 100 correlational studies were conducted comparing the effec-
tiveness of cooperative, competitive, and individualistic efforts in
promoting learning. Before 1970, almost all the reported studies
were conducted in college classrooms and laboratories using college
students as participants. The U.S. experimental research on cooper-
ative learning has its roots in Deutsch’s work in the late 1940s in a
study at MIT [49]. Between 1970 and 1990 the majority of the
studies were conducted in K-12 settings; however, in the 1990s, the
interest in investigating the use of cooperative learning at the college
level was rekindled.
Current meta-analysis work at the Cooperative Learning Center
at the University of Minnesota identified 754 studies that compare
the effectiveness of students working cooperatively, competitively,
and individualistically from 1897 to the present. Eighty five percent
were conducted since 1970; 43.5 percent had randomly assigned
subjects and 18.8 percent had randomly assigned groups; 41.4 per-
cent of the subjects were nineteen or older; 76.7 percent were pub-
lished in a journal; 31 percent were laboratory studies and 65 per-
cent were field studies [30]. These studies and others yet to be
coded will be analyzed in the coming months.
The next two sections summarize the research on cooperative
learning and problem-based learning at the post-secondary level,
that is, the studies of higher education and adult populations.
A. Cooperative Learning Research
Approximately 305 studies were located at the Cooperative Learn-
ing Center and were used to compare the relative efficacy of coopera-
tive, competitive, and individualistic learning in college and adult set-
tings, as reported in [41, 42]. The first of these studies was conducted
in 1924; 68 percent of the studies have been conducted since 1970.
Sixty percent randomly assigned subjects to conditions, 49 percent
consisted of only one session, and 82 percent were published in jour-
nals. These 305 studies form the research summarized below.
The multiple outcomes can be classified into three major cate-
gories: academic success, quality of relationships, and psychological
adjustment to college life. In addition, there are a number of studies
on students’ attitudes toward the college experience.
1) Academic Success: One of the most important goals for en-
gineering educators is that students succeed academically. Academ-
ic success is, above all, the college’s aim and the student’s aim. Be-
tween 1924 and 1997, more than168 rigorous research studies were
conducted comparing the relative efficacy of cooperative, competi-
tive, and individualistic learning on the achievement of individuals
eighteen and older. This represents the subset of the 305 studies
that focus on individual student acheivement. Other studies focused
on students’ attitudes, persistence (or retention), and other depen-
dent measures. These studies indicate that cooperative learning
promotes higher individual achievement than do competitive ap-
proaches or individualistic ones. The effect sizes, which indicate the
magnitude of significance, were 0.49 and 0.53 for competitive and
individualistic approaches, respectively. Effect sizes of this magni-
tude indicate significant, substantial increases in achievement. They
can be interpreted as saying, for example, that college students who
would score at the fiftieth percentile level on an individual exam
when learning competitively will score in the sixty-ninth percentile
when learning cooperatively; students who would score at the fifty-
third percentile level when learning individualistically will score in
the seventieth percentile when learning cooperatively [41]. For a
briefing on the meta-analysis procedure see [49].
The relevant measures here include knowledge acquisition, re-
tention, accuracy, creativity in problem solving, and higher-level
reasoning. The results hold for verbal tasks (such as reading, writ-
ing, and oral presentations), mathematical tasks, and procedural
tasks (such as laboratory exercises). There are also other subsets of
the 305 studies showing significant advantages for cooperative
learning in promoting meta-cognitive thought, willingness to take
on difficult tasks, persistence (despite difficulties) in working to-
ward goal accomplishment, intrinsic motivation, transfer of learn-
ing from one situation to another, and greater time spent on task.
The findings outlined above are consistent with results from a
recent meta-analysis focused on college level-one science, mathe-
matics, engineering, and technology (SMET) courses. Springer,
Stanne, and Donovan’s [49] study of small-group (predominantly
cooperative) learning in SMET courses identified 383 reports from
1980 or later, thirty-nine of which met the rigorous inclusion crite-
ria for meta-analysis. Of the thirty-nine studies analyzed, thirty-
seven (94.9 percent) presented data on achievement, nine (23.1 per-
cent) on persistence or retention, and eleven (28.2 percent) on
attitudes. The main effect of small-group learning among under-
graduates majoring in SMET disciplines was significant and posi-
tive, with mean effect sizes for achievement, persistence, and atti-
tudes of 0.51, 0.46, and 0.55, respectively.
Recent synthesis publications include Bowen’s [50] summary of
research on cooperative learning effects on chemistry and Prince’s
[51] summary of research on active and cooperative learning in
engineering.
Research that has had a significant influence on the instructional
practices of engineering faculty is Hake’s [52] comparison of stu-
dents’ scores on the physics Force Concept Inventory (FCI), a mea-
sure of students’ conceptual understanding of mechanics, in tradi-
tional lecture courses and interactive engagement courses. The
results shown for high school (HS), college (COLL), and university
(UNIV) students in Figure 3 show that student-student interaction
during class time is associated with a greater percent gain on the
January 2005 Journal of Engineering Education 5
![Figure 3. Plot of class average pre-test and post-test FCI scores using a variety of instructional methods [52].](/figures/figure-3-plot-of-class-average-pre-test-and-post-test-fci-akkiaatp.webp)
