Understanding Student Differences

RICHARD M . FELDER complaints being variations of'They can memorize and plug num-
Department of Chemical Engin bers into formulas but they don't know how to think!" And yet,
North Carolina State University most engineering departments have one or more faculty members
who manage to get many of those same students to perform at re-
REBECCA BRENT markably high levels, displaying first-rate problem-solving and crit-
Education Designs, Inc. ical and creative thinking skills. Skill deficiencies observed in engi-
neering graduates must therefore aiso be attributable in part to what
instmctors are doing or failing to do.
ABSTRACF An implication of these observations is that to reduce enroll-
ment attrition and improve the thinking and problem-solving skills
Students have diíFerent levels of motivation, different attitudes of engineering graduates, engineering schools should attempt to
about teaching and learning, and different responses to specific improve the qualityoftlieir teaching, which in turn reqtiires under-
elassroom environments and instructional practices. The more standing the learning needs of today's engineering students and de-
thoroughly instructors understand the differences, the better signing instruction to meet those needs. The problem is that no two
chance tliey have of meering the diverse leartiing needs of all of students are alike. They have different backgrounds, strengtlis and
their students. TTiree categories of di\'ersit}' that have been shown weaknesses, interests, ambitions, senses of responsibility, levels of
to ha\'e important implications for teaching and learning are motivation, and approaches to stud\'ing. Teaching methods also
differences in students' learning styles (characteristic ways of vary. Some instructors mainly iecture, while others spend more
taking in and processing information), approaches to learning time on demonstrations or activities^ some focus on pnnctples and
(surfïtce, deep, and strategic), and intellectual development levels others on applications; some emphasize memory and others under-
(attitudes ahout the nature of knowledge and how it should be standing. How much a given student learns in a class is govemed in
acquired and evaluated). This article reviews models that have part by that student's native ability and prior preparation but also by
been developed for each of these categories, outlines their the compatibility ofthe student's attributes as a learner and the in-
pedagogical impÜcadons, and su^ests areas for fiirther study. structor's teaching style.

This is not to say that instructors should determine their stu-

Keywords: leaming styles, approaches to leaming, intellectual dents' individual leaming attributes and teach each student exclu-
de\'elopment sively in the manner best suited to those attributes. It is not possi-
ble to discover everjtbing that affects what a student leams in a
'Instnjction begins when you, the teacher, leam fi-om the class, and even if instmctors could, they would not be able to
leamer. Put yourself in his place so that you may understand figure out the optimum teaching style for that student—the task
what he learns and the way he understands it." would be far too complex. Moreover, even if a teacher knew the
—Kierkegaard optimum teaching styles for all students in a class, it would he im-
possible to implement them simultaneously in a class of more
than two students.
I. THREE FACETS OF STUDENT DIVERSITY If it is poindess to consider tailoring ijistmcdon to each individ-
ual student, it is equally misguided to imagine that a single one-
Declining interest in engineering among high school students in size-fits-all approach to teaching can meet the needs of ever}' stu-
recent years has led to steep enrollment decreases in many engi- dent. Unformnately, a single approach has dominated engineering
neering programs. Although the problem has heen exacerbated by education since its inception: the professor lectures and the students
the high student dropout rates that have characterized engineering attempt to absorb the lecttire content and reproduce it in examina-
cumctila for decades, many engineering faculty members continue tions. That particular sÍ2^ fits almost nobody: it xiolates virtually
to view the attrition positively, believing the dropouts are mainly every principle of effective instruction established by modem cogni-
weak students who are unqualified to become engineers. This helief tive science and educational psychology [2-5], Any other approach
is wrong. In their classic study Talking ahout Leaving [1], Seymour that targets only one type of student wotild probably be more effec-
and Hewitt showed that grade distributions of students who leave tive, but it would still fail to address the needs of most students. It
technical curricula are essentially the same as the distributions of follows that if completely individualized insuiiction is impractical
those who stay in. While many of those who drop out do so because and one-size-fits-all is ineffective for most smdents, a moœ bal-
of academic diÉGculties, many others are good students who leave anced approach that attempts to accommodate the dix'erse needs of
because of dissatisfaction vnth their instruction, a fact made graphi- the students in a class at least some ofthe time is the best an instmc-
cally dear in comments quoted by Se)'mour and Heviitt. torcando.
Faculty complaints about students who remain in engineering Diversity in educadon usually refers to the effects of gender and
through graduation are also commonly heard, with many of the ethnicity on student performance. Those effects are important and

January 2005 Joumal of Engineering Educati. 57

;u-e considered elsewhere in this journal issue [6]. This artide exam- obtained in repeated assessments) and valid (the instrument mea-
ines three other important aspects of student diversity: sures what it is intended to measure).
• Learning Styles: Learning styles are "characteristic cognitive, In anotber paper in tbis issue. Olds, Moskal, and MiUer [9] offer
affective, and psychological behaviors that serve as a good introduction to reliability and validity analysis. Some of the
relatively stable indicators of how learners perceive, interact measures of reliability and validity they discuss that are applicable to
witb, and respond to the learning environment" [7], The instmments ofthe types we will describe are tbe following.
concept of learning styles has been applied to a wide variety • Test-retest reliability—the extent to which test results for an
of student attributes and differences. Some students arc com- individual are stable over time.
fortable witb theories and abstractions; otbers feel much • Intemal consistency reliaèility~^he homogeneity of items in-
more at bome with facts and observable phenomena; some tended to measure the same quantity, that is, the extent to
prefer active learning and otliers lean toward introspection; which responses to the items are correlated.
some prefer visual presentation of information and others • Scale ortbogonality—the extent to which the different scales of
prefer verbal explanations. One learning style is neither the instalment (ifthere are two or more scales) are independent.
preferable nor inferior to another, but is simply different, • Construct validity—the extent to which an instrument actual-
vi'ith different characteristic strengtlis and weaknesses. A goal ly measures tbe attribute it purports to measure. Tbe instru-
of instruction sbould be to equip students with tbe skills asso- ment scores are said to have convergent validity if they corre-
ciated m t b every learning style category, regardless of tbe late with quantities with which they should correlate and
students' personal preferences, since they will need all of divergent OT discriminant vahàity if they fail to correlate with
those skills to fianction effectively as professionals, quantities with wbich there is no reason to expect correlation.
. Approaches to Learning and Orientations to Studying: Reliability and validity data of these types are readily obtainable
Students may be inclined to approach tbeir courses in one of for some ofthe instmments to be discussed, while for others (no-
tbree ways [8], Those with a reproducing orientation tend to tably several ofthe learning styie assessment instruments) they are
take a swface approach to learning, relying on rote memoriza- difficult or impossible to find. At the end of each of sections III
tion and mechanical formula substitution and m'aking littie or (Learning Stj'les), IV (Approaches to Learning), and V (Levels of
no effort to understand the material being taugbt. Those widi Intellectual Development), we offer lists of potential researcb ques-
a meaning orientation tend to adopt a deep approach, probing tions. To each list might be added tbe foUcnving two-part question:
and questioning and e.xploring tbe limits of applicability of new If an assessment instrument is used to study any ofthe preceding ques-
material. Tbose witb an acbieving orientation tend to use a tions, 'what reliability and validity data support ¡is use (a) in general,
strategic approach, doing wbatever is necessary to get tbe bigb- and (b)for thepeculation studied?
est grade they can, taking a surfece approach if that suffices and
a deep approach when necessary'. A goal of instruction should
be to induce students to adopt a deep approach to subjects that III. LEARNING STYLES
are important for their professional or personal development.
• Intellectual Development: Most students undergo a develop- Students are characterized by different learning styles, preferen-
mental progression from a belief in the certainty of knowl- tially focusing on different types of information and tending to op-
edge and the omnisdence of authorities to an acknowledg- erate on perceived information in ciifferent ways [10,11]. To reduce
ment ofthe uncertainty and contextual nature of knowledge, attrition and improve skill development in engineering, instruction
acceptance of personal responsibility for determining trutb, should be designed to meet the needs of students whose learning
inclination and ability to gather supporting evidence for styles are neglected by traditional engineering pedagogy [12-14].
judgments, and openness to cbange if new evidence is forth- Several dozen learning style models have been developed, five of
coming. At tbe bigbest developmental level normally seen in which have been tbe subject of studies in tbe engineering education
college students (but not in many of them), Índi\áduals dis- literature. The best knovm of tbese models is Jung's Tbeory of Psy-
play thinking patterns resembling those of expert scientists chological Tj-pe as operationalized by the Mj'ers-Bri^ Type Indi-
and engineers, A goal of instruction should be to advance cator (MBTl), Stricdy speaking, tbe MBTI assesses personality
students to tiiat level by the rime they graduate. tjpes, but MBTI profiles are known to bave strong learning st)4e
In this artide, we outline models of student learning style prefer- implications [14-16], Tbis instrument was the basis for a multi-
ences, orientations to studying, and levels of intellectual develop- campus study of engineering students in the 1970s and 1980s and a
ment; review the impUcations ofthe models for engineering educa- number of otber engineering-related studies since tben [17-24],
tion; and suggest promising avenues for iuture smdy. Before doing Other models that have been applied e.\tensively to engineering are
so, we briefly discuss the topic of assessment instrument validation, those of Kolb [12, 14, 25-31], and Felder and Silverman [13, 14,
aresearch issue central to all three of these diversity domains. 32-40], We discuss tbese three models in the sections that follow.
Two other models that have been used in engineering are those of
Herrmann [14, 41-43], and Dunn and Dunn [44-tó]. Relatively
II. A NoTii ON VALIDATION little assessment has been performed on the applicability of these
models to instmctional design in engineering, and we do not discuss
Much (if this paper describes assessments of various student at- die models iurtber in this paper. For information about them, see
tributes and inferences that have been drawn from the data. Before tbe cited references.
too much stock is placed in such inferences, the instalment used to Before we look at specific models, we should note that the
collect the data should be shown to be reliable (consistent results are concept of learning styles is not universally accepted. The simple

58 Journal of Engineering Education January 1 ' ï

mention ol" the term !U"OLtses strong emotional reactions m manv ing" courses in the curriculum (such as engineering economics,
members of the academic community {notablj' but not exclusively which tends to be formula-driven} the sensors scored higher.
the psychologists}, who aigue that learning style models have no In a longitudinal study carried out at the University of Western
sound theoretical basis and that the instruments used to assess learn- Ontario b)' Rosati [22, 23], male introverts, intuitors, thinkers, and
ing sR'les have not been appropriately validated. On die other hand, judgers at the low end of the academic spectrum were found to be
the studies summarized in tlie sections that follow paint a dear and more likely to succeed in the first year ofthe engineering curriculum
consistent picture of le;u-ning style differences and their effects on stu- than were tlieir extraverted, sensing, feeling, and perceiving counter-
dent performance and attitudes. Additionally, instruction designed to parts. Rosati also observed that the introverts, thinkers, and judgers in
address a broad spectnjm oflearning styles has consistendy proved to tlie low-performance m;ile population were more likely than the ex-
be more effective dian traditional instruction, which focuses on a nar- travetts, feelers, and perceivers to graduate in engineering after four
row range of styles. We therefore propose taking an engineering ap- years, although die sensors were more likely than the intuitors to do
proach to learning styles, regarding them as usefiil heuristics for un- so. No statistically significant type differences were found for acade-
derstanding students and designing effective instruction, and
mically strong male students or for female students.
continuing to use them until demonstrably better heutistics appear.
As part of another longitudinal study, Felder [24] administered
the MBTI to a group of 116 students taking the introductory
A. The Myers-Briggs Type Indicator chemical engineering course at North Carolina State Univereity.
People are classified on the Myers-Briggs Tjpe Indicator® That course and four subsequent chemieal engineering courses were
(MBTI} according to their preferences on fotir scales derived from taught in a manner that emphasized active and cooperative learn-
Jung's Theory of Psychological Types [15]: ing, and type differences in various academic performance measures
• extraverts (try things out, focus on the outer world of people) and attitudes were noted as the students progressed through the
or introverts (think things through, focus on the inner world curriculum. The results were remarkably consistent with expecta-
ofideas). tions based on type theory:
• sensors (practical, detail-oriented, focus on facts and proce- • Intuitors performed significantly better than sensors in
dures} or iiituitors (imaginative, concept-otiented, focus on courses with a high level of abstract content, and the converse
meanings and possibiBties}. was observed in courses of a more practical nature. Thinkers
• thinkers (skeptical, tend to make decisions based on lo^c and consistendy outperformed feelers in tbe relatively impersonal
rules) ot feelers (appreciative, tend to make decisions based on environment ofthe engineering curriculum, and feelers were
personal and humanistic considerations). more likely to drop out ofthe curriculum even if they were
• judgers (set and follow agendas, seek closure even with in- doing well academically. Faced with the heavy time demands
complete data} or perceivers (adapt to changing circum- of the curriculum and the eorresponding need to manage
stances, postpone reaching closure to obtain more data). their time carefully, judgers consistently outperformed
Lawrence [15] charactetizes the preferences, strengths, and weak- perceivers.
nesses of each ofthe sixteen MBTI types in many areas of student a Extravetts reacted more positively than introverts when first
functioning and offers numerous suggestions for addressing the confronted with the requirement that they work in groups on
learning needs of students of all types, and Pittenger [16j reviews homework. (By the end of the study, both groups almost
research based on the MBTI. unanimously favored group work.}
Most engineering instruction is oriented toward introverts (lec- • The balanced instruction provided in the experimental
turing and individual assignments rather than active class involve- course sequence appeared to reduce or eliminate the perfor-
ment and cooperative learning), intuitors (emphasis on science and mance differences prewously noted between sensors and in-
math fundamentals rather than engineering applications and opera- tuitors and between extraveits and introverts.
tions}, thinkers (emphasis on objective analysis rather than interper- • Intuitors were three times more likely than sensors to give
sonal considerations in decision-making}, and judgers (emphasis on themselves top ratings for creative problem-solving ability and
following the syllabus and meeting assignment deadlines ratber to place a high value on doing creative work in their careers.
than on exploration ofideas and creative problem solving}. In 1980, • The majotity of sensors intended to work as engineers in
a consortium of eight universities and the Center for Applications large corporations, while a much higher percentage of intu-
of Psychological Type was formed to study the role of personality' itors planned to work for small companies or to go to gradu-
ti-pe in engineering education. Predictably, introverts, intuitors, ate school and work in research. Feelers placed a higher value
thinkers, and judgers generally outperformed extraverts, sensors, on doing socially important or beneficial work in their careers
feelers, and perceivers in the population studied [19,21]. than thinkers did.
In work done as part of this study, Godleski [20] reported on Very few results failed to confirm expectations from type tlieory,
grades in four sections of the introductory chemical engineering and most of the failures involved type differences that might have
course at Cleveland State University taught by three different in- been expected to be significant but were not. The condusion was that
structors. The emphasis in this course is on setting up and solving a the MBTI effectively characterizes differences in the ways engineer-
wide variety of problems of increasing complexity', witb memory ing students approach learning tasb, respond to different forms ofin-
and rote substitution in formulas playing a relatively small role. In- struction and dassroom environments, and fonnulate career goals.
tuitors would be expected to be at an advantage in tbis course, and
the average grade for the intuitors in all sections was indeed higher B. Kolb's Experiential Learning Model
than that for sensors. Godleski obtained similar results fbr other In Kolb's model, students arc dassified as having a preference fbr
courses that ennphasized intuitive skills, while in Ehe few "solid sens- (a) concrete experience or abstract conceptttatization (how they take
January 2005
Journal oJ Engineering Educatto 59
information in) and (b) active expeiimentation or reflective obsena- teaching around tbe cycle (in their term, "bolistic instruction ), an
tion (how they process information) [12, 25]. The four tyj)es of otlier was taught traditionally, and the course grades earned by the
two groups were compared. Although tbe results were not conclu-
• Type 1 (concrete, reflective)—the divei-ger. Type 1 learners re- sive, they appeared to indicate that Types 1 and 4 students were
spond well to explanations of hovi- course material relates to more likely to get low grades tban tbe more numerous Types 2 and
their ejcperience, interests, and ftiture careers. Their charac- 3 students when teaching was traditional, and that holistic instruc-
teristic question is "Why?" To be effective with Type 1 stu- tion may have helped a more diverse group of students to succeed.
dents, tlie instructor should function as a motivator. Spurlin et al, [29] report on an ongoing smdy comparing fresbman
• jy/ii 2 (abstract, reflective)—the assimilator. Type 2 learners engineering students of the four Kolb types. Their preEminary re-
respond to information presented in an organized, logical sults also show Types 2 and 3 srudents doing better academically,
fesliion and benefit if they are given time for reflection. Their atid they are conducting fiorther studies intended to pinpoint rea-
characteristic question is "WhatT To be effective, the instruc- sons for the relatively poor performance and high risk of attrition of
tor should fonction as an ex^pert. the T)'pes 1 and 4 students,
• Type 3 {abstract, active)—the converger. Type 3 learners re- Julie Sharp ofVanderbiit University has used the Kolb model in
spond to having opportunities to work actively on well- several ways as the basis for instmctional design. Her work includes
defined tasks and to learn by trial-and-error in an environ- the development of a variet)' of "wriring to learn" assignments that
ment that allows them to fail safeiy. Their characteristic should be effecrive for each ofthe four Kolb types [30] and applica-
question is "How?" To be effective, the instructor should tions ofthe model to instmction iti communicarions and teamwork
fiinction as a coach, proxiding guided practice and feedback in [26,31].
the methods being taught,
• Type 4 (concrete, active)—the accommodate. Type 4 learners C. The Felder-SUverman Model
like applying course material in new situations to solve real l) Model categories: According to a model developed by Felder
problems. Their characteristic question is "'What if?" To be and Silverman [13,32], a student's learning style may be defined by
effective, the instructor should pose open-ended questions tbe answers to four questions:
and then get out ofthe way, maximizing opportunities for the 1, Wliat type of information does the student preferenrially per-
students to discover things for themselves. Problem-based ceive: sensoryi (sights, sounds, physical sensations) or intuitive
leaming is an ideal pedagogical strateg)' for diese students. (memories, thoughts, insights)? Sensing learners tend to be
Preferences on this scale are assessed with the Leaming Style In- concrete, practical, methodical, and onented toward facts and
ventory® (McBer and Company, Boston) or the Leaming Type hands-on procedures. Intuirive learners are more comfortable
Measure (About Learning Inc., Wauconda, 111.). Most smdies of with abstractions {theories, mathemarical models) and are
engineering smdents based on the Kolb model find that the majori- more likely to be rapid and innoi-ative problem solvers [47].
ty ofthe subjects are Types 2 and 3. For example. Sharp [26] reports This scale is identical to the sensing-intuitive scale of the
that of 1,013 en^neering students she tested, 40 percent were Tj^e Myers-Briggs Type Indicator,
3, 39 percent Type 2, 13 percent Type 4, and 8 percent Type 1. 2. What type of sensory information is most effecrively per-
Bernold et al, [27] found that ofthe 350 students in their study, 55 ceived: visual (pictures, diagrams, flow charts, demonstra-
percent were Type 3,22 percent Type 2,13 percent Type 4, and 10 tions) or verbal{wntten and spoken explanations)?
percent T>pel. 3. How does the student prefer to process informadon: actively
Traditional science and engineering instruction focuses almost (through engagement in phpical activity or discussion) or re-
exclusively on lectiuing, a style comfortable for only Type 2 learners. flectively (through introspection)? This scale is identical to the
Effective instruction involves teaching around the cycle—motivating active-reflective scale ofthe Kolb model and is related to the
each new topic (Type 1), presenting the basic information and extra\'ert-introvert scale of tbe MBTI.
methods associated with the topic (Type 2), providing opportuni- 4, How does tbe smdent characterisdcally progress toward un-
ties for practice in the methods (Type 3), and encouraging explo- derstanding: sequentially {in a logical progression of incre-
ration of applications (Type 4), mental steps) or globally (in large "big picture" jumps)? Se-
A faculty training program based on the Kolb learning style quential learners tend to think in a linear manner and are
model was initiated at Brigham Young University in 1989 [28], able to function witb only partial understanding of material
About a third ofthe engineering faculty was trained in teaching they have been taught. Global learners think in a s}'stems-
around the cycle. The volunteers implemented the approach in oriented manner, and may have trouble applying new mater-
their courses, reviewed videotapes of their teaching, and discussed ial until they fiilly understand it and see how it relates to ma-
their successes and problems in focus groups. Many courses were terial they already know about and understand. Once they
redesigned; instmctors—induding a number who did not partici- grasp the big picmre, however, their holisric perspective en-
pate in the original training—used a variety of teaching methods in ables them to see innovarive solutions to problems that se-
addition to formal lecturing; discussions about teaching became a quential learners might take much longer to reach, if tbey get
regular part of department faculty meetings; and several facult)' there at all [48].
members presented and published education-related papers. Arti- More detailed descriptions of the attributes of the different
cles describing the program do not indicate the extent to which the model categories and the nature and consequences of leaming and
modified instruction led to improved learning. teaching style mismatches are g^ven by Felder and Silverman [13]
Berrold et al. [27] describe an experiment at Nordi Carolina and Felder [32]. Zyvmo and Warden [36] report on tiie develop-
State University in which one group of students was subjected to ment and successfiil implementation of hypermedia instnicrion

60 Journal of Engineering Education Januaiy2i : -

 POPULATION' A S Vs N Refer nee
[own State. IMntci'iiils Engr. 63% 67% 85% S8% 129 Constant 51

Ovlord Brookes Univ., Business 64"o 70% 68% 64% 63 De Vita|53|

lirilish studenls 85% 86"'.. 52% 76% 21
Imemaiional sludcnts 52% 62'^u 76% 52% 42
Rycrson Univ., Elee. Engr.
Students (20ÜÜ) 53% 66% 86% 72% 87 Zywno & W alen [36Í
Students (2001) 60% 66% 89% 59% 119 Zy«no 1381
Students (2OÜ2) 63% 63% 89% 58% 132 Zywno [54]
Facultv 38% •12% 94% 35% 48
Tu In ne. Engr.
Second-Year Students 62% 60% 88%

Universities in Bclo Horizonte Lopes [55

(Brazil)'
65% 81% 79% bT'/o 214
Humanities 52% 62% 39% 62% 235
Univ. of Limerick. MffrEner. 70% 78% 91% 58% 167 Seerv et a . 161
Univ. of Miehignn, Chem. Engr. 67% 57% 69% 71% 143 ,Montsomer\ 1571
Univ, of Puerlo Rico-Mayaguez
Biology (Semester!) 65% 77% 74% H3% 39 Bu\eda& \4oore [58]
Biologj'(Semester 2) 51% 69% 66% 85% 37
Biolog}'(Semester 3) 56% 78% 77% 74% 32 --
Elect, &. Comp. Engr, 47% 61% 82% 67% Buxedaetal, 591
Univ, of Sao Paulo. Engr," 60% 74% 79% 50% 351 Kuri & Truzz [60]
Civil Engr. 69% 86% 76% 54% 110
Elec. Engr. 57% 68% 80% 51% 91
Mech. Engr. 53% 67% 84% 45% 94
Indust, Enpr. 66% 70% 73% 50% 56
Univ. of Technology 55% 60% 70% 55% ? Smith etal, ¡ •1]
Kingston, Jamalen
Univ, ofWestern Ontario, E n g r ' 69% 59% 80% 67% 858 Rosati [35
First year engr 66% 59% 78% 69% 499 Rosati [34

Engr, faculty 51% 4()'''(, 94% 53% 53

Engineering Student Average 64% 63% 82% 60% 2506
E n e i n e e r i n s Fnculty Average 45% 41% 94% 44% 101
'Rows in boldfaee denote studies using the cun-ent version o f t h e Index of 1 earnine St\ks with native
English speakers.
''Portuguese translation ofthe ILS used.
'Data collected wilh Version 1 ofthe ILS, (All oiher studiesuseii Ve sion 2,;
Table 1. Reported leamingstylepreferences-

designed to address t h e learning needs of stj'les less favored by m t h results summarized in Table 1 [50]. Unless othenwse indicat-
traditional instruction, and Sharp [40] descrihes an instructional ed, the population samples shovm in Table 1 are undergraduates.
module based o n the Felder-Silverman model that makes students T h u s , for example, o f t h e 129 undergraduate engineering students
aware of differences in learmng styles and h o w they may afïêct per- w h o completed the I L S in a study conducted at Iowa State Univer-
sonal interactions, teamwork, interactions with professors, a n d sity-, 63 percent were classified as active (A) iearnet^ (and by impli-
learning difficulties and successes. cation 37 percent were classified as reflecta'C learners), 67 percent
2) The index of learning styles: T h e Index of Learning Styles® were sensing (S) learners (33 percent intuitive learners), 85 percent
(ILS) is a forty-four-item forced-choice instrument developed in were visual (Vs) learners (15 percent verbal), and 5 8 percent were
1991 by Richard Felder and Barbara Soloman to assess preferences sequential (Sq) learners (42 percent global).
on the four scales of the Felder-Silverman model. In 1994 several Table 1 iUustiates several o f t h e mismatches described by Felder
hundred sets of responses to the initial twent)'-eight-itern version of and Silverman [13] between learning styles of most engineering u n -
the instrument were collected and subjected to factor analysis. Items dergraduates and traditional teaching styles in engineering educa-
that did not load significantly on single factors were discarded and tion, Si)rt\'-three percent o f t h e undergraduates vi'ere sensore, while
replaced by Tie.vi items to create the current wrsion, which was put tradidonal engineeting instruction tends to be heavily otiented t o -
on the W o r l d W i d e W e b in 1997 [49]. T h e I L S is availahle at no ward intuitors, emphasizing theory and mathematical modeling
cost to individuals who wish to assess their own preferences and to
over experimentation and practical apphcattons in most courses^
instmctors or students who wish to use it for dassroom instmction or
82 percent o f t h e undergraduates were i^sual learners, while most
research, and it may be licensed by non-educational o i ^ n i z a t i o n s .
engineering instruction is overwhelininglv verbal, emphasizing
Learning style preferences of numerous students and faculty written explanations and mathematical formulations of phi.'sical
members have been determined using the Index of Learning Styles, p h e n o m e n a over demonstrations and visual illustrations^ and 64

J a n u a r y 2005 Joumal oJ Engin 61

percent of the students were active, while most engineering courses teaching style will not be helped to develop critical skills in their iess
other tlian laboratories rely almost exclusively on lectures and read- preferred learning style categories [13, 14], The optimal teaching
ings as the principal vehicles for transmitting information. style is a balanced one that sometimes matches students' prefer-
Table 1 also shows that 60 percent ofthe students assessed were ences, so their discomfort level is not too great for them to learn ef-
sequential and traditional engineering education is heavily sequen- fectively, and sometimes goes against their preferences, forcing
ti;tl, so tliis dimension does not involve the same type of mismatch them to stretch and grow in directions they migbt be inclined to
ohserved for the others. Global students constitute a strong and im- avoid if given the option.
portant tninorit)', however. They are the multidisciplinary tlunkers, The preceding paragraph su^ests what we believe to be the most
whose broad vision may enable them to become, for example, important application of learning styles, which is to help instmctors
skilled researchers or chief executive officers of corporations, Unfor- design a balanced teaching approach that addresses the learmng
mnately, traditional engineering education does litde to prowde needs of all of their students. Designing such an approach does not
stttdents with the systemic perspective on individual subjects they require assessing the students' leaming style preferences: it is enough
need to tlinction effectively, and the ones who take too long to get it for instructors to select a model and attempt to address all of its cate-
by themselves lu-e at risk academically. gories (in Kolb model terms, to teach around the cycle), knowing that
Section II briefly discussed the issue of instrument validation. everydassprobablycontainsstudentswith every preference [14].
The Index of Learning Styles is one ofthe few instruments men- Assessing the leaming style profile of a class with an instrument
tioned in this paper for which reliability and validity data have been such as the Myers-Briggs Type Indicator, the Kolb Leaming Style
collected for engineering student populations [37, 50, 54]. We will Inventory, or the Index of Learning Styles—without being overly
not provide details of the reliabilit)' analvses here; suffice it to say concemed about which students have which preferences—can pro-
that all three ofthe studies just cited conclude that the ILS meets or vide additional support for effective instmctional design. For exam-
exceeds accepted reliabilitj' standards for an instmment of its type. ple, knowing that a large majority of smdents in a dass are sensing
Felder and Sptirlin [50] summarize results from several studies that and visual leamers can—and should—motivate the instructor to
provide evidence of both convergent and di^'ergent construct \'alidi- find concrete and \asual \vays to supplement rhe presentation of ma-
ty. Profiles of engineering students at different institutions sbow a terial that might normally be presented entirely abstracdy and ver-
high degree of consistency wath one another and differ substantially bally. Many specific suggestions for designing instmction to address
and in a predictable manner from proftles for engineering faculty the fiill spectrum of leanung styles are given by Felder and Silver-
and humanities students (see Table 1). Another indication of con- man [13] and Lawrence [15],
vergent validity is that preferences for sensing and active leaming What about identifying individual students' leaming styles and
measured on the ILS ^vere found to correlate witli preferences for sharing the results with them? Doing so can provide them witli valu-
sensing and extraversion measured on the Myers-Briggs Type able dues about their possible strengths and weaknesses and indica-
Indicator [33]. tions of ways they might improve their academic performance. Pre-
As noted previously, the conventional lecture-based teaching cautions should be taken if students are told their leaming st\'les,
approach in engineering education favors intuitive, verbal, reflec- however. The instructor should emphasize diat no leaming stj'le in-
tive, and sequential learners. In yet another demonstration ofthe stiunfient is infallible, and if the students' perceptions of how they
constmct validit}' of the ILS, Zywno and Waalen [36] found that leam best differ from what the instrument says, they should not dis-
on average the performance in conventionally taught courses of count their own judgment. They should also be assured that their
each of tlie favored types was superior to that of the less favored leaming style preferences are not reliable indicators of what they are
types and they also found that the use of supplemental hypermedia and are not capable of domg, and that people with way possible
instruction designed to address the needs of all types decreased the leaming style can succeed in any profession or endeavor. If a student is
performance disparities. Felder and SpurUn [50], Livesay et al. [37], assessed as, say, a sensing leamer, it says nothing about his or her intu-
and Zywno [54] conclude that the ILS may be considered reliable itive skills (or sensing skills, for that matter); it does not mean that he
and valid for assessing learning styles, although all three papers rec- or she is unsuited to be an engineer or sdentist or mathematidan; and
ommend continumg research on the instrument. it does not excuse the low grade he or she made on the last exatn. In-
stmctors or advisers who use leaming st)'les as a basis for recommend-
D. Pedagogical Implications and Potential Misuses ofLcaming Styles tng curriculum or career choices are misusing die concept and could
Studies have shown that greater learning may occur when teach- be doing serious disservices to their students and advisees.
ing styles match learning styles than when they are mismatched [11,
13,62,63], but the point of identiiying learning stj'les is not to label E. Questions for Further Study
individual students and tailor instruction to fit their preferences. To As previously noted, leaming styles are controversial, vnth ques-
iunerion effectively as engineers or memhers of any other profes- tions commonly being raised regarding their meaning and even
sion, smdents will need skills characteristic of each type of learners: their existence. Much work needs to be done to resolve these ques-
the powers of observation and attention to detail ofthe sensor and tions and also to determine the \Mlidit)' of different learning style
the imagination and abstract tliinking ability of the intuitor; the models for engineering students and to confirm or refit te daims re-
abilities to comprehend information presented both visually and garding the effectiveness of a balanced teadiing approach. The fol-
verbally; the systematic analysis skills of the sequential learner and lowing questions merit investigation:
the multidisciplinary synthesis skills of tlie glohal leamer; and so on. 1. Does an assessed learning style preference indicate (a) the
If instruction ts heavily biased toward one category of a learning t)'pe of instruction a student is most comfortable vidth or
style dimension, mismatched students may he too uncomfortable to (b) the tjfpe of instmction most likely to lead to more effec-
learn effectively, while the students whose learning styles match the tive learning? To wbat extent are the two coincident?

62 Journal of Engineering Education January 2(105

 2, Do any learning style preferences depend on smdents' eth- They have an intrinsic motivation to learn, witb intellectual curiosi-
nic and cultui-al backgrounds? Wliich preferences, and what ty ratiier than the possibility of external reward driving tbeir efforts.
arc the namre and extent ofthe dependences? They cast a critical eye on each statement or formula or analytical
3. To what extent does teaching exclusively to a student's procedure they encounter in dass or in the text and do whatever
learning stj'le preference lead to (a) greater student satisfac- they think might help them understand it, such as restating text
tion, (b) impro\'emeiit in skills associated witb that prefer- passages in their own words and trying to relate the new material to
ence, (c) lack of improvement in slôlls associated witb the tilings they have previously learned or to everyday experience. Once
opposite preference? the information makes sense, they try to fit it into a coherent body
4, Does a curriculum liea\ály biased toward a particular learn- of knowledge.
ing style increase the inddence of dropouts of students with Students who adopt a strategic approacb do wbatever it takes to
conflicting styles? To what extent does more balanced in- get tbe top grade, Tbey are weD organized and efficient in tbeir
struction reduce attrition and improve academic perfor- studying. Tbey carefully assess tbe level of effort tbey need to exert
mance of those students? to acbieve their ambition, and if they can do it by staying superficial
5, Is the provision of choice over learning tasks an effective tbey will do so, but if tbe instructor's assignments and tests demand
strategy for accommodating different learning style prefer- a deep approacb tbey will respond to the demand.
ences? How mucb choice should be provided and what kind? A smdent may adopt different approaches to learning in differ-
6, How effective is instructional technology that provides al- ent courses and even for different topics within a single course. An
ternative pathways through a body of material, witb tbc orientation to studying is a tendency to adopt one ofthe approaches
pathway's being designed to appeal to different learning in a broad range of situations and learning environments [5, 8]. Stu-
style preferences? dents who habimally adopt a surface approacb have a reproducing
7. How should learning style preferences be incorporated in orientatioir, those who usually adopt a deep approach have a mean-
adwsing? How effective are interventions that take learning ing orientation; and those inclined to take a strategic approacb have
st}'le into account? an achieving orietitation. The Lancaster Approaches to Studying
8. Does mixing learning styles wben forming project teams Questionnaire (LASQl [65] is a sixty-four-item questionnaire tbat
lead to better team products? Does it lead to increased in- involves twelve subscales relevant to tbe three orientations and four
terpersonal conflict? If the answer to eacb question is "yes," adcUtional subscales. Shorter forms ofthe LASQ_tbat provide less
do tbe improved products compensate for tbe greater con- detailed information are referenced by Woods et: al. [66], and an al-
flia risk? Does making team members aware oftbeir learn- ternative to the LASQ_is the Study Process Questionnaire devel-
ing style differences lower the potential for conflict? oped by B i ^ [67].
9, How belpfijl to students is discussion of learning styles in Woods et al. [66] report on a study in which one of the short
dass? forms ofthe LASQ^was administered to 1,387 engineering stu-
10. To wbat extent are preferences on comparable scales of dif- dents. The strongest inclination ofthe students was toward a strate-
ferent instruments correlated? ^ c approacb, followed in order by a surface approacb and a deep
11. To wbat extent do the answers to any of the preceding approach. Bertrand and Knapper [68] report LASQ_results for stu-
questions depend on the strength of students' learning style dents in other disciplines. Chemistiy and psychology students went
preferences? firom a preference for strategic learning in their second year to a
preference for deep learning in their fourth year, w t h botb groups
displajdng consistcndy low inclinations toward a surface approach.
IV. APPROACHES TO LEARNING AND Bertrand and Knapper [68] also report on three groups of sm-
ORIENTATIONS TO STUDYING dents in two multidisciplinar}' curricula—smdents in the second
and fourth years of a project-based environmental resource studies
A. Definitions and Assessment program and smdents in a problem-based program on the impact of
Marton and Säljö [64] define three different approacbes to new materials. All three groups showed relatively strong inclina-
learning—a surface approach, a deep approach, and a strategic approach. tions toward a deep approach. There was little difference in the pro-
Students who adopt a surface approach to learning memorize files ofthe second- and fourth-year students, su^esting that the re-
facts but do not try to fit them into a larger context, and they follow sults might reflect the orientations ofthe students selecting into the
routine solution procedures without trying to understand their ori- programs more than the influence ofthe programs.
gins and limitations. These students commonly exliibit an extrinsic There are similarities between orientations to studying and
motivation to leam {I've got to leam this to pa.'^s tbe course, to graduate,
learning styles. Both represent tendencies tbat are situationally de-
to get a goodJob) and an unquestioning acceptance of everytbing in pendent, as opposed to fixed traits like gender or handedness that
the textbook and in lectures. To tbem, studying means scouring always characterize an individual. Just as a student who is a strong
their texts for worked-out examples that look like the homework inmitor may function like a sensor in certain situations and vice
problems so tbey can simply copy the solutions. They either ignore \'ersa, a student with a pronounced meaning orientation may under
the text outside ofthe examples or they scan through it with a high- some circumstances adopt a surface approacb to learning, and a
lighter, looking for facmal information that the instmctor might strongly reproducing student may sometimes be motivated to dig
consider important, which they will attempt to memorize before deep. Similarly, just as students may be reasonably balanced in a
the exam. learning style preference, fi:equently functioning in ways cbaracter-
Students who take a deep approacb do not simply rely on mem- istic of, say, both sensors and intuitors, some students may be al-
orization of course material but focus instead on understanding it, most equally likely to adopt deep and surface approaches in

January 2005 Journal of Engineering Educ 63

diflerent courses and possibh' within a given course. We will shortiy 3. Assessment methods that emphasize conceptual understand-
say more about instructional conditions that influence the choice. ing encourage a deep approach; methods that emphasize re-
call or rhe application of routine procedural knowledge dis-
B. Effects of aDeep Approach on Learning Outcomes courage it.
Researchers have assessed student approaches to learning and 4. Teaching methods that foster active and long-term engage-
corrdatcd the results with various learning outcomes [3, 5, 69]. In ment with learning tasks encourage a deep approach.
studies cited by Ramsden [5], students who took a deep approach to 5. Opportunities to exercise responsible choice in the content
reading created comprehensive and integrated summaries of mater- and method of study encourage a deep approach.
ial they had read, interpreting the information rather than simply 6. Stimulating and caring teaching encourages a deep approach;
repeating it, while those who took a surface approach were more apathetic or inconsiderate teaching discourages it. A corollaiy
likely to recite fragments ofthe reading content ahnost randomly. is that students who perceive that teaching is good are more
The deep approach also led to longer retention of information^— likely to adopt a deep approach than students with the oppo-
presumably because the information was learned in context rather site perception.
than by rote memorization—and to consistently higher grades on 7. An excessive amount of material in the curriculum and an
examinations and in courses. unreasonable workload discourage a deep approach.
For example, Prosser and Millar [70] examined first-year 8. Previous experiences with educational settings that encour-
physics students' understanding of force concepts before and afrer aged deep approaches fijrther encourage deep approaches.
their introductory mecbanics course. Eigbt out of nine students A similar statement can be made regarding surface
who took a deep approach and oráy two of twenty-tliree wbo used a approaches.
surface approach showed significant progress in understanding WelJ-established instructional strategies can be used to achieve
force concepts, moving away from Aristode and toivard Newton. these conditions. Inductive teaching methods such as problem-based
Meyer et al. [71] found that eng^eering students who adopted a Slid project-based learning [72-77] can motivate students by helping
deep approach in a course were very likely to pass the course (in fiict, to make the subject matter relevant to their prior experience and in-
none of their subjects in this category failed}, while students who terests (addressing item #1 above) and they also emphasize concep-
adopted a surface approach were very likely to fiul. The students tual understanding and de-emphasize rote memorization (item #3}.
who adopted a deep approacli also generally expressed greater satis- An excellent way to make expectations dear (item #2) is to articu-
faction with their instruction. late them in the form of instructional objectives [78-80]—statements
of observable actions students sbould be able to do (define, explain,
C. Motivating a Deep Approach to Learning calculate, derive, model, design) once they have completed a secrion
The approach a student might adopt in a particular situation de- of a course.
pends on a complex array of factors. Some are intrinsic to the stu- Several student-centered teaching approadies accomplish the
dent (e.g., possession of prerequisite knowledge and skills and moti- goal of actively involving students in learning tasks (item #4}, no-
\'ation to learn the suhject}, while others are determined more by tably active learning (engaging students in dass activities other than
tbe instructional environment (e.g., the content and clarity ofthe listening to lectures) and cooperative learning (getting students to
instrLictor's expectations and the nature and quality ofthe instruc- work in small teams on projects or homework under conditions that
tion and assessment}. hold all team members accountable for the learning objectives asso-
Biggs [3] proposes that achieving desired learning outcomes re- dated with the assignment) [Sl-84]. Trigwell et al. [85,86] found a
quires constructive alignment of the elements just listed. Alignment positive correlation between an instructor's use of such instructional
means that the factors under the instructor's control are all consis- methods and students' adoption of a deep approach to learning.
tent with the goal: the desired outcomes are dearly communicated Other references provide numerous examples of teaching in a stim-
to the students as expectations, instructional methods known to ulating eating manner (item #6}, providing dear feedback by de-
favor the outcomes are employed and methods that work against signing appropriate tests, among other wavs (item #2}, and provid-
them are avoided, and learning assessments (homework, projects, ing choice in learning tasks (item #5) [4', 87-91]. Several ofthe
tests, etc.) are explicitly directed toward tbe outcomes. Constritctive references cited in this paragraph and the preceding one also sum-
means that the instructional design adheres to the principle of con- marize research connecting die instructional methods mentioned
structivism, which holds that knowledge is constructed by the wiEhavarietyofpositive!eaniingoutcomes[72,82,84].
leamer, as opposed to being simply transmitted by a teadier and ab-
sorbed. The teacher's job is to create conditions that lead students to D. Questions for Furtlier Study
construct accurate representations of the concepts being studied, Of the three diversity domains discussed in this paper, ap-
first abandoning prior misconceptions if any e.\jst. proaches to learning may be the one witli the most solid research
Certain features of classroom instruction bave been found to be base [3, 5, 69, 92], However, Uttle has been done thus far to apply
constnictively aligned with the adoption of a deep approach to and extend the research to engineering. Following are some ofthe
learning, while other feamres have the opposite effect [3,5,69): questions that might profitablj' be studied:
1. Interest in and background knowledge ofthe subject encour- 1. What percentages of students in traditional engineering cur-
age a deep approach; lack of interest and inadequate back- ricula are characterized by reprodudng, meaning, and achiev-
ground discourage it. ing orientations to studying?
2. Clearly stated expectations and dear feedback on progress 2. Do approaches to learning and otientations to studying de-
encourage a deep approach; poor or absent feedback discour- pend on students' ethnic and cultural backgrounds? What are
the nature and extent of tlie dependences?

64 Journal ol Etwtneerini' Educ January 2'"'5

 3. Does tbe adoption of a deep approach to learning in an engi- tive progression of stages intended to characterize women's devel-
neering course lead to improved learning as it has been shown opment. Baxter Magolda's Model of Epistemological Develop-
to do in other disciplines? If so, for which learning outcomes ment [97,110] integrates die preceding models by defining altema-
can improvements be demonstrated? tive patterns for all levels but the highest one, with one pattern
4. Do the instructional condirions and methods (e.g., active chiu'acterizing more men than women and tbe other more women
learrting, cooperative learning, and problem-based leaming) than men. Table 2 shows the levels and patterns of the Baxter
that purportedly motivate tbe adoption of a deep approach Magolda model and the correspondences between tbat model and
do so in engineering? How and to what extent can smdents the other three. The paragraphs that follow discuss primarily the
with a reproducing orientarion be motivated to adopt a deep models of Baxter Magolda and Perry.
approach? The developmental pattern described by all four models has the
5. Would one need to reduce the content or extend the length of foUowing general form, Smdents at the lowest levels {Baxter
the engineering curriculum to reduce the heavy time de- Magolda's absolute knowing and Perry's dualism) believe that every
mands on smdents that have been shown to discourage the intellecmal and moral question has one correct answer and their
adoption of a deep approach? professors {at least the competent ones) know what it is. As the sm-
6. How do engineering smdents with meaning, reproducing, dents confront challenges to tbeir belief s)^tems in their courses and
and achieving orientations to leaming compare in high-level through interactions witb peers, they gradually come to believe in
thinking skills, such as critical thinking and creative thinking? the validity of multiple viewpoints and concurrentiy decrease their
7. Might discussing approaches to learning with engineering reliance on the word of authorities {Baxter Magoida's transitional
smdents promote their adoption of a deep approach? and independent knowÍ7!g and Perry's multiplicity). Baxter Magolda's
highest level, contextual knowing, which parallels Perry's contextual
relativism (Level 5) and the early stages oí commitment in the face of
uncertainty (Level 6 and perhaps Level 7), is characterized by final
V. LEVELS OF INTELLECTUAL DEVELOPMENT
rejection ofthe notions of certainty of knowledge and the omni-
science of authorities. Contextual knowers take responsibility for
Many students enter college in what fCroU [93] refers to as a state
constructing knovvledge for themselves, relying on both objective
of "ignorant certaintj'," believing that knowledge is certain, beliefs
anal)^is and lnmition and taking into account (but not accepting
are either right or wrong, tbe authorities {e.g., their professors and
without question) the ideas of others whose expertise they acknowl-
the authors of their textbooks) have the answers, and their job is to
edge. They move away from the idea commonly held by indepen-
memorize those answers and repeat them on tests. As they gain ex-
dent knowers (Level 4 on the Perry scale) that all opinions are
perience, most gradually progress toward a state of {again in KroU's
equally valid as long as the right method is used to arrive at them,
terminology) "intelligent confosion," in which they recognize that
and they acknowledge the need to base judgments on tbe best avail-
all knowledge is contexmal, take responsibility for making their
able evidence witbin the given context, even in the face of uncer-
own judgments on the basis of evidence rather than rel)'ing on the
tainty and amhigmty.
word of authorities, and become relatively sophisticated at gather-
ing and interpreting ev-idence from a wide range of sources. In other
words, those who attain that state (which relatively few do by the B. Assessment ofDevelopment
time tbey graduate) come to think hke expert scientists and engi-
In the method tradition'olly used to assess developmental levels,
neers. This progression has been referred to as intellectual {or cogni-
trained interviewers conduct structured open-ended interviews, the
tive or epistemológica!) development.
interviews are transcribed, and nrained raters analyze the transcripts
Different levels ofintellecmal development constitute the third and assign levels to the interviewees. While this method is univer-
categorj' of smdent diversity to be discussed here. In this section sally considered tbe most valid and reliable approach to assessment,
we re\dew several models ofintellecmal development, discuss their the cost of implementing it has motivated the design of pencil-and-
appücabilit)' to engineering education, survey existing apphca- paper instruments that can be more easily administered and scored.
rions, and suggest areas for fiirther exploration. Much ofthe ma- The Measure of Intellectual Development (MID) for the Perry
terial presented is drawn from a pair of artides recently pubhshed model [111] and the Measure of Epistemological Reflection
inthisjournal[94,95], {MER) for the Baxter Magolda model [112,113] call for smdents
to write essays on topics derived firom the intenaew protocols and
A. Models of Intellectual Development the essa)'s are rated in the same manner as the tntervtew transcripts.
Four models ofintellecmal development are described in the lit- The Learning Environment Preferences {LEP) questionnaire
erature. The first, Perr^''s Model of Intellectual Development [114] and Reflective Thinking Appraisal [115] arc Likert-scale in-
[96,97], is the only one that has had widespread application in engi- struments for assessing le\'els on the Perr\' and King-Kitchener
neering education [98-106]. The low and intermediate levels of models, respectively.
Perry's model are almost identical to the low and intermediate levels Wliile pencil-and-paper instmments are easier and faster to ad-
of the King-Kitchener Model of Reflective Judgment [97, 107, minister than interviews, the ratings obtdned tend to be one to two
108], which may be the most widely used and validated ofthe foiu- positions lower than ratings obtdned wth interviews and correlate
models outside engineering educarion. (The two models diverge at moderately at best witb inteniew ratitigs [100, 104]. To improve
their highest levels, which are rarely attained by college smdents.) ln the correlation, PaveUch, Miller, and Olds [104] de\'eloped an on-
Women's Ways ofKnowing, Belenlg' et al. [109] suggest that Perry's line tool called Cogito, vvhich asks questions about scenarios related
model largely characterizes men (its formulation was based almost to four controversial issues, asks follow-up questions based on tbe
entirely on interviews v/ith male smdents) and propose an altema- responses, and uses a neur.J net to idenrify response pattems and

January 2005 journal of Engm mgEdnc 65

 Ab olutc sitionni Independent ContextuHl
Baxter Mngoldii Kn Knowins' Knowing''

•I
Miistery Kece vine 1 111 personal InUTpersonal Individual Interinüividual

Perry M Ulli pi icily M Ulli pi icily ^ ^ ^ ^ ^ H Co textual Relalivism

Subordínale Prd
BL'k-riky ^ Procctliiral Procedural
(Women's Wiiys Ruci.
ofKi.o«irg)

King-
Kitcliener
H H Ihi
kiw« cdce

1 ate
Prea
Th nking
Paiiern

Thinking
Paiiern
Quasi-Rclleclive
knowlcdye

Relleetivc
1 hinking

,viiig. A and the responsihility to

and die e questions to make sure
Iheir inlbrmalion is corrL'ci and L-hallcngc devialitms Irom Iheir view oTihe trulh. Receiving pattern (r ;n): Students take in and
nicord informalicin pajsively, willioul iiucslionin^ or challenging il.
" Transitional knowing. Some knowkdgi: is cenain and some is nol. Authorities have lhe responsibility to communicale the certainiies. and the
students an; responsible fur making Ihcir own judfiments regarding the uncenainlies. Impersonal pattern (more men than nomenl. Skidents make
iudgments uiiine a loeical procedure prescritied by auihorilies, I'ull credil is deserved for following the right procedure, regardless of the elarlly ofthe
rcnsoniiis .wi'.i ÚK L|ii;ility of lhe supponing e\'idence. Interpersonal pattern (more \vr»men than men); Students base jud^meTils on ir^tuilion and
p,i .11 I..II •• .ii.irusi logical analysisandahsirLict reasoning.
' liiiii|Hiuii HE kiiu\ii!iE. Most knowledge is uncertain. Studenls lake responsihility for their own learning rather than relying heavily on authorities or
|\ I :• ii : I \w\ colled and use evidence lo suppon judgmenls. but ollen superficially, and elieve ihal when knowledge is uncertain all
Lij,^i„.,uii. ,,^.irj:ii^ il are equally good if the riglii procedure is used lu reach them. Individual paiiern [more men than women): Rely on objective
lugK. trititLiI ill i liking, Eind challenging their own Liiid others' posT tions lo cstuhlish trtilii and make moral judgmeiils, Interindividual pattern (more
woiiKn IhLin men): Students rely un caring, empalhy, and underslanding of otht-rs' piisiliiins as bases for judgmenls,
Contextuikl kno^t'ing. All truths are contextual. Sludenls liike responsihility for making: |ud9mcnls, «lckno^S'ledging lhe need lo do so in lhe face of
uncertainty and anibiguily. They use all possible sources of evidence in the proeess—objeelive analysis and intuition, their own thoughts and
feelings and ideas of others whose expertise they acknowledge—and they remain open to changing their deeisions if new evidence is forthcoming.

Tai>k2. Models of intellectual develefment[94j.

Year (n) Average Perry level (SD) % at Level S Reference

1(45) 3.27 (0.44)
2 (34) .1.71 (0.53) Pavelich a; Moore [102]
4(46) 4.28 (0.70) 25%
1(21) 3.27 (0.40) 0%
3(21) 3.33(0.35) 0% Wise el al. |IO61
4(21) 4.21 (0.50) 33%
Table 3. Perry levels of en^neering students.

assign levels to them. The neural net is trained on a set of responses the subjects. Most ofthe entering students were near Pern' Level 3,
submitted by individuals with known levels on the Reflective Judg onl)' beginning to recognize that not all knowledge is certain and
ment and Periy models (based on structured interviews). In initial still relying hea\'ily on autliorities as sources of truth. The average
tests, the maximum correlation coefficient of about 0,5 between the change after four years of college was one level, with most of the
interview-based levels and the Cogito-as signed levels was indeed change occurring in the last year. Neither instructional approach
higher than the best values obtained for the penciJ-and-paper in- met its goal of elevating a significant number of students to Level 5,
struments, but was still well below the desired minimum value of As discouraging as these resitlts might seem, one could speculate
0,S. The authors speculated that 0.5-0.6 might be an upper bound that a cunicuJum lacldng such features as the experiential leaming
to the correlation coefficient between ratings obtained using inter- envirotiment at Mines or the project-based first-year experience at
views and objectively-scored instruments. Penn State (in Wankat's term, a "dualistic curriculum" [91]) would
lead to even less growth than was observed in tlie two studies in
C. Levels ofDevelopment of Engineering Students quesáon.
Table 3 summarizes results of two studies in which the Periy Wise et •ú- [106] also report Peny ratings of eight male engineer-
levels of beginning and advanced engineering undergraduates were ing students and eight female engineering students who completed
measured. Pavelich's study [102] was carried out to assess the effect the first-year project-based design course. There was initially no ap-
on intellectual development ofthe strong experiential learning en\'i- preciable difference between the two groups in average Peny rating
ronment at the Colorado School of Mines. The study by Wise et al, or SAT scores. At the end of die first year, tlie arerage Peny rating
[106] was intended to determine the effect ofa first-year project- was 3.50 for die men and 3.16 for the women; at the end ofthe third
based design course at Penn State. The studies are remarkably con- year the ratings were 3.50 (men) and 3.00 {women); and at tlu- end
sistent in their assessments ofthe initial and final average levels of ofthe fourth year the ratings were 4.00 (men) and 4.50 (women)

66 Journal of Engineering Education January 2( •:

None of the differences were statistically significant R, QuestionsforFurther Study
altliough the differences for the diird year came close (p = 0.054). The study of the inteüectual development of engineering stu-
The lack of significance cotild be an artifact ofthe small sample stze. dents is still in a preliminary stage, with many basic questions as yet
To the extent that the observed differences are real, tliey support the unaddressed. Several of tlie questions follow.
contentions of Belenlg' et al. [109] and Baxter Magolda [110] that 1. What intellectual development level distributions character-
men and women exhibit different patterns of development. ize most engineering students at different stages ofthe cur-
riculum? Are there differences between smdents at different
D. Promoting Intellecmal Development types of schools? Do levels vaiy with demographic or socio-
A necessary condition for students' intellectual growth is chal- logical factors or academic predictors such as SAT scores,
lenge to the beliefe that characterize their current developmental lev- and if so, how? Do levels correlate with course grades? Are
els. An absolute knower who is never confronted -mth open-ended the contrasting gender-related patterns of Baxter Magolda's
questions that have multiple sokttions cannot be expected to accept mode! observed for engineering students? What levels and
the realitj' of multiplicity and move to transitional knovring sponta- patterns characterize engineering facult}'?
neously. Similarl)', an independent knower who is not challenged 2. To what extent do levels on the different models of intellec-
for inadequate use of evidence in making judgments is not likely to mal development actually correspond in the manner shown
make the shift to contextual knovñng. in Table 2? (Those correspondences are based entirely on the
The challenge cannot be too great, however. If students are descriptions ofthe levels and not on comparative data.)
confi-onted with tasks that call for thinking too far above their cur- 3. To what extent do the instmctional conditions listed In Table
rent developmental level (in Vygotsky's term, outside their Zone 4 promote intellectual development? What other instmction-
of Proximal Development [116]), they may not be capable of un- al conditions or methods do so, and to what extent?
derstanding what is being required of them. Moreover, challenge 4. Is Vygotsky's Zone of Proximal Development a reality in the
alone—even at an appropriate leve!—may not be sufficient to context of intellectual development? In other words, are as-
move students to higher levels of development. Students con- senions that smdents cannot cope with instmction more than
fronted with challenges to their fundamental beliefs may feel (say) one Perry level above their current developmental level
threatened and either persist at their current developmental levels vaüd, or can suitable support enable them to bridge broader
or retreat to even lower levels. To avoid these outcomes, instruc- cognitive gaps?
tors should provide appropriate support to help their students meet 5. What are the effects of introducing students to the concept of
the challenges.
intellectual development? For example, would being abie to
Felder and Brent [95] propose five instructional conditions that identify theix own attitudes in the context of devdopmental
should provide the balance of challenge and support needed to pro- levels promote their intellectual growth, or might explicit de-
mote intellectual growth and suggest numerous ways to establish scription ofthe different stages of development lead to resent-
the conditions. The conditions are listed in Table 4. Most of the ment and increased resistance from students at lower levels?
methods suggested in [95] are supported by extensivelj' dted refer-
ences on teaching and leaming [2. 3, 5, 87, 88, 90, 91], and the
student-centered approaches of Condition D have repeatedly heen VI. TEACHING TO ADDRESS A L L TÍÍREE
shown to have positive effects on a wide variety of leaming out- FORMS OF DIVERSITY
comes [119-123]. However, until a researcher implements the rec-
ommendations and assesses the intellectual development of the Teaching strategies have been recommended to help instructors
subjects {ideally comparing their growth with that of a control meet the needs ofthe full spectrum of leaming styles [13, 15, 26],
group that goes through a traditionally taught curriculum), the ef- induce students to adopt a deep approach to learning [3,5,69], and
fectiveness ofthe conditions in Table 4 at promoting grovrth will promote students' intellectual development [95]. The prospect of
remain speculative. implementing three different teaching approaches simultaneously
to achieve all three goals could be intimidating to instructors, but
commonalities among the three diversity' domains and the instruc-
A. Vartclv ÜFIJ chuce ii/tectrnrijg /íívjt^
tional methods that address them make the task manageable. The
d m d basis ofthe discussion that follows is the set of recommendations
B
for promoting intellectual development presented in Table 4.
Assigning a variety of leaming tasks (part of Condition A of
Table 4) is foremost among die methods that have been recom-
mended to address leaming goals in all three diversity domaitis.
D. Variation enables instmctors both to challenge the beliefs about
d knowledge and its acquisition that characterize different develop-
A
3 Measu es la deru.se resistance tosludenl-cf nterc:d insiruciion
mental levels and to ensure that students are confi-onted with some
assignments that require a deep approach to leaming. Varietj' in as-
c of canil ^ about siudcnl
„.,,.,cl,ofd.,.l.p™n, signments is also a cornerstone of recommendations fbr addressing
2 A »are

whilçj the flill spectrum of learning stj'les, with some problems emphasiz-
ing practical considerations and requiring carefiil attention to details
Tahle 4. IlStructional conditiot U thatfacilitate intellectual
(sensing strengths) and odiers calling fbr tlieorerical interpretation
growth [95],
and mathematical modeling (intuitive strengths), some involving

January 2005 Journal of Engineering Educatio 67

individual efforts (reflective) and others requiring teamwork 4. To what extent do eacb oftbe conditions listed in 1 al' •
(active), and so on, including tht: use of s tu dent-centered instructional [ I'-lf'
A deiu' similarity exists between the characteristics of a deep ap- such as cooperative learning and problem/project-based
proadi to learning and the defuiing attributes of Baxter Magolda's learning—promote intellectual grovrth, tbe adoption of a
contextual knowledge level of intellecmal development (Perry Level deep approach, and tbe development of skills assodated with
5 ;ind above). Both a deep approach and contexmal knowing involve different learning styles in engineering students? Are there
tiiking responsibility for one's own learning, questioning autborities instmctional methods or conditions not covered in Table 4
rather tiian accepting their statements at face value, and attempting that would adiieve the same goals?
to understand new knowledge in the context of prior knowledge and
e.xperience, A reasonable assumption is that conditions known to
promote a deep approach should also promote intellectual growth. VII. SUMMARY
As we noted in section IV-C, Conditions A3, B, C, DI, D2, and El
of Table 4 have been shown to encourage a deep approach. Students differ from one anotlier in a wide variety of ways, in-
Inductive instructional approaches such as problem-based learn- duding the types of instniction to whicb tbey respond best (learn-
ing (Conditi,on D of Table 4) should also be effective for addressing ing styles), the ways tbey approach tbeir smdies (orientations to
die learning goals associated witb all tliree domains. Open-ended studying and approaches to learning), and their attitudes about the
problems that do not have unique well-defined solutions pose a seri- nature of knowledge and their role in constructing it (levels of intel-
ous challenge to students' low-Ieveî beliefs in the certainty of knowl- lectual development). While much has been written about all tbree
edge and the role of instmctors as providers of knowledge. Such categories of diversity in the general education literature, relatively
problems by tbeir very nature also require a deep approacb to learn- little solid research specific to engineering education has been per-
ing (rote memorization and simple algorithmic substitution being formed. We bave suggested a number of promising areas for study.
dearly inadequate strategies for them), and solving tiiem eventually • Validating instruments used to assess ¡earning styles, orientations
requires skills associated with different learning styles: the imagina- to study, and levels of intellectual development of engineering
tion and capacity for abstract tbinking oftbe intuitor and the atten- students: Most of the instruments listed in this paper have
tion to detail of the sensor; the holistic vision of tbe global learner been subjected to reHabiUty and validity analysis, but few of
and the systematic analytical approacb oftbe sequential learner.
the vabdation studies involved engineering student popula-
Requiring students to modifj' tbeir fundamental bebefs about tions. While results obtained witb an instrument that has not
tbe nature of knowledge can be unsettling or threatening, as can been rigorously validated may be informative (especially if
calling on them to adopt a deep approacb to learning when tbey are tbey are consistendy replicated in independent studies), con-
inclined to a surface approach or to complete assignments tbat call clusions can be made and generalized witb much greater
for abilities not nonnally associated with their learning style prefer- confidence if the instrument has been shown to be reliable
ences. It is reasonable to speculate that the conditions in Table 4 in- and valid for tbe population being smdied.
volving support for students should help students respond success- • Charactenzing students: Learning style profiles, orientations
fiilly to these types of challenges. Offering a cboice of learning tasks to Study, and levels of intellectual development of engineer-
(part of Condition A of Table 4), expÜdtiy communicating e.xpec- ing students should be assessed and analj-zed. Differences in
tations (Condition B), modeling and providing practice and feed- any oftbe three should be identified among (a) smdents at
back on bigh-level tasks (Condition C), and shovÁng respect for different levels of a single engineering curriculum, (b) stu-
students at all levels of development (Condition E) are all ways to
dents in different branches of engineering, (c) students at dif-
provide support.
ferent t>pes of schools (research-intensive and teaching-in-
While these linkages among the domains may appear logical, tensive, public and private, small and large), (d) engineering
they must be considered speculative in the absence of rigorous con- students and students in other disdplines, and (e) students
firmatory analysis. Here, tben, is our final list of suggested questions and faculty,
to explore, • Establishing correlations among the three diversity domaitis:
1. How strong is the hj'pothesized link between orientation to Correlations among learning styles, orientations to study,
studying and level of intellectual development? Put another and levels of intellectual development should be identified.
way, to what extent does a student's level of intellectual devel- Correlations could be useflil for instructional design—so
opment correlate with his or her tendency to adopt a deep ap- that, for example, if the anticipated correlation between a
proach to learning? meaning orientation to study and a contextual knowing level
2. What correlations e.xist between learning styles and ap- of development on Baxter Magolda's scale (Perry Levd 5) is
proaches to learning and/or levels of intellectual develop- verified, instructors wishing to promote the intellectual de-
ment? For example, are intuitors more likely than sensors and velopment of their students could tee! more confident in
global learners more likely tban sequential learners to adopt a using mediods known to promote a deep approacb to learn-
deep approacb? Are tbere developmental level differences be- ing. Moreover, confirming the e.xistence of antidpated corre-
tween students with different learning style preferences? lations would support the construct validit}' of the instru-
3. Are tbere gender-related patterns in learning style prefer- ments used to assess the positions or preferences being
ences or orientations to studying comparable to the patterns compared.
in Baxter Magolda's Model of Epistemological Dc\'elop- • Evaluating the effectiveness of instructional methods and pro-
ment? Are tbere cultural differences in any ofthe three diver- grams: Most engineering faculty would agree that to be eftec-
sity categories? tive, instruction should address the needs of students across

68 Journal of Engitiening Education Januaiy H)\)$

 the hill spectrum of learning styles, promote adoption of a [11] Schmeck, R.R., ed.. Learning Strategies and Learning Styles,
deep approach to learning, and help students advance to New York, N.Y.: Plenum Press, 1988.
higherlevels of intellectual development. Many authors have [12] Stice, J.E., "Using Kolb's Learning Cycle to Improve Student
proposed instructional methods for achieving one or more of Leüming," Engineering Education, Vol. 77, No, 5,1987, pp. 291-296.
those goals. What is needed is soUd evidence that either sup- [13] Felder, RM., and Silverman, L.K., "Learning and Teaching
ports or refijtes daims ofthe effectiveness of those methods Sti'les in Engineering Education," Engineering Education. Vol. 78, No. 7,
in achieving the desired outcomes.
We began this paper with an admonition by Kierkegaard that 1988.pdf,
true instruction begins when instmctors understand their students. [14] Felder, R.M., "Matters of Style," ASEE Ptism, Vol. 6, No. 4,
An impottant component ofthat understanding is awareness ofthe 1996, pp. 18-23. Online at -w-mw.ncsu.edti/felder-public/Papers/LS-
different attitudes students have toward learning, the different ways Prism.htm.
tliey approach it, and how instmctors can influence both their atti- [15] La\vrence, G., People Types and Tiger Stripes: A Practical Guide
tudes and approadies. The research summarized in this paper and Learning Styles, 3rd ed., Gainesville, Fla.: Center for Applications of Psy-
the researdi that remains to be done can help instruetors gain that chological Type, 1993.
awvueness. The more successful they are in doing so, the more effec- [16] Pittenger, D.J., The Utility of the Myers-BHggs Type Indicator,"
tively they can design instmction that benefits all of their students. In Review of EdticationalResearch, Vol. 63,1993, pp. 467-488.
turn, tbe better students understand the strengths and weaknesses [17] McCaulIey, M.H., "Psychological Types of Engineeting
associated with their attitudes and preferences, the more likely they Students—Implications for Teaching," Engineering Education, Vol. 66,
will be to leam effectively wliile they are in school and throughout No. 7,1976, pp. 729-736.
[18] Yokomoto, C.E., and Ware, J.R., "Improving Problem Solving
Performance Using the MBTI," Proceedings, 19S2 ASEE Conference and
Exposition, Washington, D.C.: Ametican Society for Engineering
ACKNOWLEDGMENTS Education.
[19] McCaulley, M.H., Godleski, E.S., Yokomoto, C F . ,
The authors are grateful to Mike Prince, Kenny and Gary Harrisberger, L., and Sloan, E.D., "AppUcations of Ps>'chologica] Type in
Felder, and the Journal of Engineering Education reviewers for their Engineeting Education," Engineering Education, VoL 73, No. 5, 1983,
insightful comments on preliininary versions of this paper. pp. 394-400.
[20] Godleski, E.S., "Learning Style Compatibility of Engineering
Students and Faculty," Proceedings. 1984 Frontiers in Education Conferen
REEERENCES Washington, D.C.: ASEE/IEEE.
[21] McCauUey, M.H., Macdaid, G.P., and Granade, J.G., "ASEE-
[1] Seymour, E., and Heivitt, H., Talking about Leaving: Wby Under- MBTI Engineering Consortium: Report ofthe Fint Rve Years," Proceed-
graduates Leave the Sciences, Boulder, Colo.,: Weswiew Press, 1997. ings, 1985 ASEE Annual Conference,\'Vashmg:on, D.C.: American Sodety
[2] Bransford, J.D., Brown, A.L., and Cocking, R.R., eds., HotuPeople for Engineeting Education.
Leam: Brain, Mind, Expei-ience, and Scbool, Washington, D.C.: National[22] Rosati, P.A., "Student Retention from First-Year Engineering
Acidemy Ptess,2000.Onüf\e At bttp.-//i'oots-nap.edu/catalog/9853.btm/. Related to Personalitj'Type," Proceedings, 1993 Frontiers in Education Con
[3] Biggs, J., Teaching for Quality Learning at University, 2nd ed., / ^ « « , Washington, D.C.: ASEE/IEEE.
Buckingham: The Society for Researdi into Higher Educarion and Open [23] Rosati, P.A,, "Psychological Types of Canadian Engineering Stu-
University Press, 2003. dents,'7u«njo/o/Pí)íí-éu%Vfl/ri^í, Vol. 41,1997, pp. 33-37.
[4] McKeacfiie, WJ., McKeacbie's Teaching Tips: Strategies, Researcb, [24] Felder, R.M., Felder, G.N., and Dietz, E.J., "The Effects of Per-
and Tbeory ñtr College and Universtty Teachers, 11th ed., Boston, Mass.: sonality Type on Engineering Student Performance and Attitudes,"/oa/--
HouEhtonMifflin,2002, nal of Engineering Education, Vol. 91, No. 1, 2002, pp. 3-17. Online at
[5] Ramsden, P., Uaniing ta Teach in Higher Editeaiwa. 2nd ed.,bttp://iuwv:.ncsii.edu/fetder-ptMc/Papers/longmbti.pdf.
London: Taylor md Fmd., Inc., 2003. [25] Kolb, DA., Experiential Learning: Experience as thi Source of
[6] Chubin, D,E,, May, G.S., and Babeo, E.L., "Diversifying die En- Learning and Development, Englewood Cliffi, N.J.: Prendce-Hall, 1984.
gineering VJoitíoKtL: Jaamal af Engineering Edueation, Voi. 94, No. 1, [26] Sharp, J.E., "Teaching Teamwork Communication wth Kolb
2005, pp. 73-86, Learning Style Theor>'," Proceedings. 2001 Frontiers in Education
[7] Keefe, J,W,, -Learning Styk An Overview," in Keefe, J.W,, ed.. Conference. Washington, D.C.: ASEE/IEEE.
Student LeamtTig Styles: Diagnosing and Prescribing Programs, Reston,[27] Va.i Bernold, L.E., Bingham, W.L., McDonald, P.H., and Atria,
National Assodation of Seeondary School Prindpals, 1979. T.M., 'Impact of Holisric and Leaming-Oriented Teaching on Academ-
[8] Entwistle, N., "Motivational Factors ¡n Students' Approaches to ic Success," Jottmal of Engineering Education, Vol. 89, No. 2, 2000, pp.
Learning," in Schmeck, R.R., ed.. Learning Strategies and Learning Styles,191-199.
Ch. 2, New York, N.Y.; Plenum Press, 1988. [28] Harb, J.N., Durrant, S.O., and Teny, R.E., "Use of tbe Kolb
[9] Olds, B.M., Moskal, B.M., and MiUer, R.L., "Assessment in En- Learning Cycle and the 4MAT System in Engineering Educarion,"/o'"'-
gineering Education: Evolution, Approaches, and Future Collaborations," nat of Engineering Education, Vol. 82, No. 2,1993, pp. 70-77.
JoumaloJEngineeringEducatio>i,Vo\.94,NoA,2<}05,pp,U-25. [29] Spurlin, J.E., Bemold, L.E., Crossland, C.L., and Anson, CM.,
[10] Corno, L., and Snow, R.E., "Adapting Teaching to Individunl "Understanding How Freshman Engineering Students Think They
Differences Among Learners," in M. C. Wittrock, ed.. Handbook of Re- Leam," Pimeeditigs. 2003 ASEE Conference and Exposition, Washingto
search on Tiacbing, 3rd ed.. New York, N.Y.: Macmilkn and Co., 19S6. D.C: Anierican Societ}' for Engineering Education.

January 2005 ^ _ ^ _ ^ ^ Journal of Engineering Educat 69

 [30] Sharp, J,E„ I lurb, J.N,, ¡ind Teriy, R,E., "Cümbininí; Kolb [48] Felder, R.M,, "Meet Your Students, 2, Susan and C_,
Learning Styles and Wriüng to Learn in EngiiicL-rins Chi^eCJournal of Chemical Engineering Education,Vol 24,^0. 1, 1990, pp. 7-8, O • m, :it
EngineeringEdiication, Vul, 86, No, 2,1997. pp. 93-101, Ti'-ïi'ïi.'- fii^u.ii iu/feldi 'r-ptihlic/Coturnns/Susanglenda. html.
[31] Slr.up, J.E,, "Learning Styles and Technic;d Communication: [4V| IVldcr, R.M., and Soloman, B,A,, Index ofLearnin,^ Styles,
Improving Communication and Teamwork Skiils." Procecding.^, 1998 htip://v.>wu.>. iicsii. edii/felder-piiblic/ILSpage. html.
Frontiers in Education Conjhence, Washington, D.C: ASEE/IEEE. [50] Felder, R.M., and Spurlin, J., "Applications, ReUability, and Va-
[32] Felder, R.M,, "Reacliing the Second Tier: L.;arning and Teach- lidity of the Index of Leaming Styles," InternationalJournal ofEn^neering
ing Sti'lcs in College Sdence Y.àuc,\uor\;''Joumal ofCollege Science Teaching,ciCiucation, in press,
Vol. 23, No, 5, 1993, pp. 286-290. Online at wwiu.ncsu.edii/felder- [51] Constant, K.P,, "Using Multimedia Techniques to Address Di-
piMc/Papei-s/Secondtier.html. verse Learning Styles in Materials Education," Journal ofMaterials Educa-
[33] Rosati, P.A,, and Felder, R.M,, "Engineering Student Responses /lon. Vol. 19,1997, pp. 1-8.
to :u] Index of Leaming Styles," Proceedings, 1995 ASEE Conference Exposi- [52] Paterson, K.G,, "Student Perceptions of In temer-Based Leaming
/;Df7,Wüshington,D,C.; American Society for Engineering Education. Tools in Environmental Engineering Ed\ic2.ûon,'' Journal af Engineering
[34] Rosati, P,A., "Comparisons of Learnir^ Preferences in an Engi- Education, Vol. 88, No, 3,1999, pp. 295-304,
neering Program," Proceedings, 1996 Frontiers in Education Conference, [53] De Vita, G., "Leaming Stj'les, Ctilture and Inclusive Instruction
Washington, D,C.: ASEE/IEEE. in the Multiculmral Classroom: A Business and Management Perspec-
[35] Rosati, P.A., "Specific Differences atid Similarities in tlie Learn- tive," Innovations in Education and Teaching Intel-national, Vol. 38, No. 2,
ing Preferences of Engineering Students," Proceedings, 1999 Frontiers in 2001, pp, 165-174,
Education Conference, Washington, D.C: ASEEAEEE. [54] Zywno, M.S., "A Contribution of Validation of Score Meaning
[36] Zywno, M.S., and WaiUen, J,K., "The Effect of Hypennedia In- for Felder-Soloman's Index of Learning Styles," Proceedings. 2003 ASEE
struction on Achievement and Attitudes of Students with Different Leam- Conference and Exposition, Washington, D . C : American Society for Engi-
ing Styles," Proceedings, 2001 ASEE Confei-cncs and Exposition, Washing-neering Edtication,
ton, D.C: American Society for Engineering Education, [55] Lopes, W.M.G,, ILS—Inventario de Estilos de Aprendizagem de
[37] Livesay, G,A., Dee, K.C, Nauman, EA., and Hites, Jr,, L.S, Feldei-Soloman: ¡nvestigiiçâo de sua Validade em Estudantes Universitarios de
"Engineerittg Student Learning Styles: A Statistical Analysis Using Bclo Horizonte. Masters Thesis, Univeisidade Federal de Santa Caterina,
Felder's Index ofLeaming Styles," Presented at the 2002 ASEE Conference Brazil, 2002,
and Exposition, Montreal, Quebec, June 2002. [56] Seei>-, N,, Gaughran, W.F., and Waldmann, T., "Mulri-Modal
[38] Zywno, M.S., "Instructional Technology, Leartiing Styles, and Leaming in Engineering Education," Proceedings. 2003 ASEE Conference
Academic: Achievement," Proceedings, 2002 ASEE Conference and Exposi-and Exposition, Washington, D . C : American Societ}' for Engineering
tion. Washington, D,C,: American Society for En^neering Educarion. Education,
[39] Dee, K.C,, Livesay, G.A,, and Nauman, E.A,, "Learning Styles [57] Montgomery, S,, "Addressing Diverse Student Leaming Styles
cf First- and Second-Year Engineering Smdents," Proceedings, 2003 through the Use of Multimedia," Proceedings. 1995 Frontiers in Education
ASEE/IVFEOInternational Colloquium, Washington, D,C,: American So- Conference, Washington, D . C : ASEE/IEEE,
cietj'for Engineering Educarion, [58] Buxeda, R., and Moore, D.A,, "Using Learning Styles Data to
[40] Sharp, J,E,, "A Resource for Teaching a Leaming StylesA'eam- Design a Microbiology Cornie," Joumal of College Science Teaching, Vol. 29,
work Module ivith the Soloman-Felder Index of Learning St}'les," Proceed- 1999, pp. 159-164.
ings, 2003 Frontiers in Education Conference, Washington, D,C,: [59] Buxeda, R,, Jimene-z, L., and MoreU, L,, "Transfortning an Engi-
ASEE/IEEE, neering Course to Enhance Student Learning," Proceedings. 2001 Interna-
[41] Herrmann, N,, The Crcati'ue Brain, Lake Lure, N . C : Brain tional Conference on Engineering Education, Arlington, Va,; International
Books, 1990, Network for Engineering Education and Research,
[42] Lumsdidne, E., and Voitle, J,, "Introducing Crearivity and Design [60] Kiiri, N.P,, and Tmzzi, O,M.S., "Learning Styles of Freshmen
into Traditional Engineering Design Courses," Proceedings, 1993 ASEE Engineering Srudents," Proceedings, 2002 Internaticnai Conference on Engi-
Conference and Exposition,'WiishingtQn. D,C,: American Societ)'for Engi- neering Education, Arlington, Va.: International Network for Engineering
neering Education. Education and Research,
43. Lumsdaine, M,, and Lumsdaine, E,, "Thinking Preferences of En- [61] Smith, N,G.. Bridge, J,, and CLtrke, E,, "An Evaluation of Students'
gineenng Students: Implicarions for Curriculum Restructuring,"yo//rí;¡í/o/ Performance Based on Their Prefcned Learning Styles," in Pudloivski, Z.L.,
Engineering Education. Vol, 84, No, 2,1995, pp, 193-204, ed.. Proceedings, 3rd Annual Conference ofthe UNESCO Intemational Centre
[44] Dunn, R,, Beaudn-, J,S,, and Klavas, A,, "Sun-ey of Research for EngineeringEdiication, Melbourne, Austnilia: UlCEE, 2002.
on Learning St)'les," Educational Leadership, Vol, 46, No. 6, 1989, [62] Hayes, J,, and Allinson, C W . , "Matching Leaming Style and In-
pp, 50-58, structional Strategy: An Application of tlie Person-Environment Literac-
[45] Dunn, R., "Understanding the Dunn and Dunn Learning Styles tion Paradigm," Perceptual and Motor Skills. Vol. 76,1993, pp, 63-79.
Model and the Need for Individual Di^nosis and Prescription," Reading, [63] Hayes, J,, and Allinson, C W , , "The Implications of Leaming
Writing, and Leaming Disabilities, Vol, 6,1990, pp, 223-247, Styles for Training and Development: A Discussion ofthe Matching Hy-
[46] Hein, T.L., and Budny, D,D., "Teaching to Students' Leaming pothesis," BritishJoirmal of Management. Vol, 7,1996, pp. 63-73.
Styles: Approaches That Work," Proceedings, 1999 Frontiers in Education [64] Manon, F., and Siiljö, R., "Approaches to Leaming," Ln Marton
Conference, Washington, D.C: ASEE/IEEE. c:a],[92].
[47] Felder. R.M,, "Meet Yotir Smdents. 1. Stan and Nathan," Chemi- [65] Ramsden,P., The Lancastei Approaches to Studying and Course Per-
cal Engineering Education, Vol. 23, No. 2, 1989, pp. 68-69. Online at ceptions Questionnaire: Lecturer's Handbook, Oxford: Educational Methods
www. ncsii.cdu/fsldei-public/Columns/Stannathan. html. Unit, Oxford Polytechnic, 1983,

70 Journal of Engineering Education January 2005

 |Ó6] Woods, D.R.. Hr>'mak. A.N., and Wriyht. H . M „ "Ap- ¡86] Trig\veQ, K,, Prosser, M , and Waterhousc. F.. "Relations be-
projiches to LearniriLç anti Learning Environments in Proble m-bused tween Teachers' Approaches to Teaching and Students' Approaches to
vs. Lecture-based Learning," Proceedings, 2000 ASEE Conference and Leiiming," Higher Education, Vol, 37,1999, pp, 57-70,
Exposition, Washington, D . C ; American Society for Engineering Ed- [87] Chiclcering, A.W,, and Gamson, Z.F,, Applying the Seven Princi-
pies for Good Practice in Undcrgraduate Education, New Directions for
[67] Biggs, J.B., Study Process Questionnaire, Aiisn-:t!ia: University of Teaching and Leaming, No. 47, San Francisco, Cal,: Jossey-Bass, 1991.
Neweasde, 1979. [88] Eble, K.E,, Tbe Craft of Teaching, 2nd ed., San Francisco, Cai.;
[68] Bertrand. D,, and Knapper, C.K., "Contextúa] Influences on Stu- Jossey-Bass, 1988,
dents' Approaches tn Leimiing in Three Academic Departments," Unpub- [89] Felder, R.M., "Designing Tests to Mawmize Leüming," Joumal
lished honors thesis. Psychology Department, University of Waterloo, ofProfessional Issues in Engineering Education and Practice, Vol. 128, No. 1,
Waterloo. ON. Canada, 1991 (circd by Woods et a]., [66]). 2002, pp, 1-3, Online at http://iu'.i!V}.ncsu.edu/felder-puliHc/Papers/
[69] Prosser, M., and Trigwell, K,, Understanding Leaming and Teach- TestingTips.htm.
ing, Buckingham: The Soeiety for Research into Higher Education and [90] Lowman, J., Mastering the Techniques of Teaching, 2"'' ed., San
Open Utiiversity Press, 1999, Francisco, CaL: Jossey-Bass, 1995.
[70] Prosser, M,, ,ind IVIiUar, R,, "The 'How' and Wliy' of Leaming [91] Wankat, P., TbeEffective, Efficient Professor: Teaching, Scholarship,
Physics," European Journal of Psychology of Education, Vol. 4, 1989, pp, and Service, Boston, Mass,: AJlyn and Bacon, 2002,
513-528. [92] Marton, F., Hounsell, D,, and EnRvistle, N,, eds., The Experience
[71] Meyer. J,H,F„ Parsons, P,, and Dunne, T.T., "Individual Study of Leaming, 2nd ed,, Edinburgh: Scortisb Academic Press, 1997.
Orchestrations and their Association widi Learning Outcome," Highei^Ed- [93] Kjoll,B.M,, Teaching Hearts and Minds: College Students Refîect on
«Mto«, Vol. 20,1990, pp. 67-89, tbe Vietnam War in Literature, Carbondale, III.: Southern llhnois University
[72] Dochy, F., Segcrs, M., Van den Bossche, P., and Gijbels, D., "Ef- Press, 1992,
fects of Problem-Based Learning: A Me ti-Analysis," Leaming and ¡nstrtic- [94] Felder, R.M,, and Brent, R,, 'The Intellectual Development of
rim. Vol. 13,2003, pp. 533-568. Science and Engineering Students, 1, IVlodels and Chuhnns^ts," Joumal ff
[73] Duch, B.J., Groh, S.E., and Allen, D.E., eds.. The Poiner of Engineering Education, Vol, 93. No. 4,2004, pp, 269-277,
Proble,n-Based Leaming, Sterling, Va.: Stylus, 2001. [95] Felder, R.M., and Brent, R., "The Inteilecma! Development
c ' J C • ' C J T "T" U• D
[74] de Graaf, E, and Kolmos, A,, "Characteristics of Problem-Based f
Learning," International Journal ofEngineering Education, Vol. 19, No. 5, jLicnce anQ engineering otuoenrs, ¿, leacnin^ to jr rom ote
2003, pp. 657-662. Growth," Joiimal of Enginen^ing Education, Vol, 93, No, 4, 2004, pp,
[75] Kolmos, A,, Fink, F,Ii., and Krogh, L,, eds,, TheAalborg PBL 279-291.
Model: Progress, Di'versity, and Challenges, Aalborg, Denmark: Aalborg [96] Perry, W,G,, Fori/is of Intellectual and Ethical Development in the
University Press, 2004. College Years:A Scbeme, San Francisco, Cal,: Jossey-Bass, 1988. (An updat-
[76] McMaster University Problem-Based Learning Web Site. ed reprint ofthe original 1970 work.)
httpy'/iirww.chemeng.mcmaster.ca/plii'PBL.HTM. [97] Love, P,G., and Guthrie, V,L,, Understanding undApplying Intel-
[77] University of Delaware Problem-Based Learning Web Site, lectual Development Tbeory, New Directions for Student Services, No, 88,
http^J/imim!. iidel. edv/pbl SanFrandsco, Cal.: Jossey-Bass, 1999,
[78] GioTA\iná,'^.E.,HowtoWriteandUseInstructionalObjecáves,Suí [98] Culver, R,S„ and Hackos, J.T., Teriy's Model of InteUecmal
ed., New York, N.Y.: MacmiUan, 1994. Development," £«^ne£Tin^c£í/;íroí/ofí, Vol, 72,1982, pp, 221-226.
[79] Mager, R.F., Preparing Instructional Objectives, 3rd ed,, Atlanta, [99] Fitch, P., and Culver, R.S,, "Educational Activities to Stimulate In-
Ga.: Center for Effective Performance, 1997, tellectual Development in Perry's Scheme," Proceedings, 19S4 ASEE Confer-
[80] Felder, R.M., and Bient, R., "Designing and Teaching Courses erice and Exposition, Washingtor\, D,C,: American Society for Engineenng
to Sadsfy the ABET Engineering Catena," Joumal ofEngineering Educa- [100] Pavelich, M,J„ and Filch, P,, "Measuring Students' Develop-
tion, Vol, 92, No. 1, 2003, pp. 7-25, Online at bttp./Zimvw.ncsu.edu/J'elder- ment Using the Pen^- Model," Pi^oceedings, 19SSASEE Conference and Ex-
public/Papcrs/ABET_Paper_UEEj.pdf position, Washington, D,C,: American Society for Engineering Education.
[81] Felder, R,M,, and Brent, R., "Leaming by Doing," Cheinica!En- [101] Pavehch, M,J„ "Helping Smdents Develop Higher-Level
gineering Education, Vol, 37, No. 4, 2003, pp. 232-283. Online at Thinking: Use ofthe Perry Model," Proceedings, 2996 Frontieis in Educa-
http://Tmmi!.ncsu.edu//i:lder-pub!ic/Columns/Active.fidf. tion Conference, WasHngton, D,C.: ASEE/IEEE.
[82] Prince, M., "Does Active Leaming Work? A Review of the Rt- [102] Paveüch, M.J,, and Moore, W.S,, "Measuring die Effect of Ex-
scarch," Journal of Engineering Education, Vol, 93, No. 3, 2004, periential Education Using lhe Peny MadA" Joumat ofEngineering Edu-
pp. 223-231, cation, Vol, 85, No. 4,1996, pp, 287-292,
[83] Johnson. D,W,,Johnson, R.T,, and Smith, K. A., Active Leaming: [103] Felder, R.M., "Meet Your Smdents, 7. Dave, Manha, and Rober-
Cooperation in the College Classroom, 2nd ed., Edina, Minn.: Interactionto," Chemical Engineering Education, Vol. 31, No. 2,1997, pp. 106-107. On-
Book Co, 1998, line •^^http://i}yuyií}.ncsit.edii/Jclder-public/Coliimns/Perry.htínl.
[84] Smith, K.A,, Johnson, D,W,, Johnson, R.W,, and Sheppard, [104] Pavelich, M.J,, Miller, R.L., and Olds, B,M,, ''Softvwire for Mea-
S.D., "Pedagogies of Engagement: Classroom-Bused Pzacúccs," Joumal of suring tlie Intellectual Development of Students: Advantai^es ,\nA Limita-
Engineering Education, Vol. 94, No. 1,2005, pp, 87-101. tions," Proceedings, 2002 ASEE Conference and Exposition, Washington,
[85] Trigwell, K., Prosser, M., Ramsden, P., and Mardn, E,, "Improv- D,C,: American Society for Engineering Education,
ing Student Learning through a Focus on the Teacbmg Conte.vt," m [105] Marra, R., and Palmer, B., "Encouraging Intellectual Growth:
Gibbs, G., ed.. Improving Student Learning, Oxford: G.rford Centre for Senior College Student Proii^tis" Joirmal ofAdult Development, VoL 11,
StafFDevclopment, 1998. No. 2,2004, pp. 111-122,

January 2005 Journal oJ Engineering Education 71

 [106] WiscJ,, Lee, S.H,, Litóngcr, T A . , Marra, R.M,, and Palmer, sured by the Pi^riy Scherciv," Joumal ofEngineering Education, Vol. ''•''•. i"^"-
B., "A Report on a Four-Year Longitudiri,il Study of liitcllectiiíd Develop- 1,2000, [ip.39-45.
mmt ofZugiiKmn^Under^dxiates," Journal of Adult Deveíopment,'Vo\. [119] Hake, R,R., "Interactive Engagement vs, Tradition.il Merh-
ll,No.2,20Ü4, pp. 103-110. ods: A Six-thuusand Student Survey of Mechanics Test Data for Intro-
( 107] King, P,M,, and Kitchener, K.S., Developing R^ectiveJudgment: ductory Physics Coutscs," American Journal of Physics, Vol. 66, 1998, pp.
Uudmtanding and Promoting Intellectual Growth and Critical Thinking in 64-74,
.í,/c/tw<í;/!rt)i,//;d'«//.t, San Francisco, Cal.: Josscy Bass, 1994. [120] Sptinger, L., Stanne, M.E., and Donovan, S., Effects of Small-
[108] King, P.M., atid Kitchener, K.S,, "The Reflective Judgment Group Leaming on Undergraduates in Science, Mathematics, Engineering, and
Model: Twent>' Years of Reai:arcli on Epistc-niic Cognition," In Hcfer, Technology: A Meta-Analysis, 1998, http://www.'wcer.wisc.edu/nise/CLI/
B.K. and Pintrich, P.R., i:ós.,PeiMnal£pistemology: The Psychology of BeliefsCL/resource/R2.htm.
ahout Knowledge and Knowing, Mahwah, N J.: Lawrence Erlbavim Associ- [121] Johnson, D.W.Johnson, R.T,, and Stanne, M.B., "Cooperative
ates, 2001. Learning Methods: A Meta-Analysis," May 2000, http://www.co-
[109] Beknlg-, M.R, Clinchy, B.M., Goldberger, N.R., and TanilL-, operation. org/pages/cl-methods. html.
J.M., Women's Ways ofKnowing: The Development of Self, Voice, and Mind, [122] Terenzini, P.T., Cabrera, A F . , Colbeck, C.L,, Parente, J.M.,
New York, N.Y.: Basic Books, 1986 (reissued 1997). and Bjorklund, S,A, "Collaborative Learning vs. Lecture/Discussion; Stu-
[110] Bitter Magolda, M.B., Knowing and Reasoning in College. San dents' Reported Learning GMr\s" Joumal of Engineering Education, VoL
Frandsco, Cal.: Jossey-Bass, 1992, 90,No. 1,2001, pp. 123-130,
[ i n ] Moore^y^.S., The Measure of Intellectual Development: An histnt- [123] Fügen, A.P., Crouch, C.H,, and Mazur, E., "Peer Instmction: Re-
ment Manual, Olympia, Wash.: Center for the Study of Intellectual Devel- sultsfix>ma Range of Classrooms," The Physia Teacher, VoL 40, 2002, pp.
opment, 1988, 206-209. Otilittc at http://ma7.ur-www.harvard.edu/puhlications/
[112] Baxter Magolda, M.B.. and Porterficld, W.D.,AssessingIntellec- Puh_2S6.pdf
tiial Development: The Link between Theory and Practice, Alexandria, Va.:
American College Personnel Association, 1988,
[113] Ba.\ter Magolda, M.B., "A Constructivist Revision ofthe Mea- AUTHORS' BIOGRAPHIES
sure of EpistemologicalReflection|"yDHI7^a/o/"Co//í^c'5íH¿fr/^íDI•'l'^.•/o^melI/,
Vol. 42, No. 6,2001, pp, 520-534, Richard M . Felder, Ph.D., is Hoechst Celanese Professor
[114] Moore, W.S., T h e Learning Environment Preferences: Ex- Emeritus of chemical engineering at North Carolina State Uruver-
ploring the Construct Validity of an Objective Measure of the Pert}' sit)'. He is co-autiior of the te.-a Elementary Principles of Chemical
Scheme," Journal of College Student Development, Vol. 30, 1989, Pioccsses (3rd Ed., Wiley, 2000), co-director ofthe ASEE National
pp. 504-514. Effective Teaching Institute, and a fellow ofthe ASEE.
[115] Wood, P.K,, King, P.M., KitchL-ner, K.S., and Lynch, C.L., Address: Dept. of Chen:ûcal Engineering, N,C. State University,
Technical Manual for the Reflective Thinking Appraisal, Columbia, Mo.: Raleigh, N C 27695-7905; e-mail: [email protected].
University of Missouri, 1994.
[116] Vygotsky, L,S., Mindin Society: Tbe Development ofHigher Psy- Rebecca Brent, Ed.D., is president of Education Designs, Inc.,
chological Processes, Cambridge, Mass; Harvard Universitj' Press, 1978. a consulting firm speciaÜzing in university and college faculty de-
[117] Woods, D.R., et al., "Developing Problem-Soiling Skills: The velopment and assessment of pre-college and college teaching.
McMaster Problem Solving Program," Journal of Engineering Education, She is co-director of the ASEE National Effective Teaching
Vol, 86, No. 2,1997, pp. 75-92. Institute.
[118] Marra, R.M., Palmer, B., and Litzinger, T.A., 'The Effects of a Address: Education Designs, Inc., 101 Lochside Drive, Caty,
First-Year Design Course on Student Intellectual Dci'dopmcnt its Mea- NC, 27511; e-mail: [email protected].

72 Journal of Engineering Educatio. January:

Copyright of Journal of Engineering Education is the property of ASEE and its content may not be copied or
emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.