Felder, Richard, "The Myth of the Superhuman Professor."
J. Engr. Education, 82(2), 105-110 (1994).
Richard M. Felder
Department of Chemical Engineering
North Carolina State University
Raleigh, NC 27695-7905
Your car has been making ominous clunking sounds lately and you've brought it in for a checkup. The mechanic says there are some minor transmission problems, which he'll fix while you wait. You position yourself in the grimy lounge with the stack of four-year-old issues of Road and Track and wait. And wait. After an hour you poke your head into the garage and ask what's going on, and the mechanic tells you he's right on top of the problem and is almost finished. After two more hours you storm out and find a maze of parts scattered all around and the mechanic thumbing through a repair manual and looking perplexed. You ask him what he's doing, and he says he thinks he needs to replace the torque converter but he's not sure where it is and the diagram in the book is no help. Seriously worried now, you ask him if he's really a trained mechanic, and he says that actually he's not - he's a body repairman - but the shop policy is that everybody is supposed to work on everything and if you'll just have patience he'll figure it out sooner or later...and while he's got your hood up he might just have a look at a couple of valves that seem like they're about to go. You start to whimper and plead with him to put your car together and let you go.Pretty frightening, eh? Wait - it gets worse.
You started experiencing severe chest pain a couple of hours ago and managed to get yourself to the emergency room. "Heart," they say, and in a little while you find yourself being trundled into the operating room...except that instead of putting you on a conventional operating table they roll you onto a large mahogany desk. The orderlies bustle around in their surgical gowns and masks, getting everything ready for the operation, and then the surgeon strides in, wearing a gray pinstripe suit and a power tie. He opens his briefcase, takes out a letter opener, and prepares to make the first incision. "Wait a minute," you say anxiously. "Are you really a surgeon?" "Well, technically no," he replies. "I'm the accountant from psychiatry. It's just hospital policy that all staff members have to pull their weight in the operating room. Now hold still - I've never done one of these before and it may sting a little." As you grab a ball point pen to fend him off you wake up.
These are of course absurd and unrealistic
nightmares (at least the second one is). Hospitals would never
require accountants to perform open-heart surgery - nor dermatologists,
for that matter. If I drive across a bridge I can safely assume
it was not designed by a civil engineer whose specialty was sewage
treatment. We expect professionals to perform jobs for which
they were trained: the idea of requiring them to perform every
task in their field, regardless of their training and experience,
is ludicrous and not subscribed to by any profession. Except
Consider the universal vision of the professor of the 90's. She does pioneering research in a critical area and brings in big bucks to support the research, including several six-figure NSF grants and 60% release time. She publishes 5-10 papers each year in the most prestigious journals in her field and is a shoo-in for the National Academy. She is a dedicated and stimulating instructor and wins teaching awards at her university and nationally. She does more than her fair share of the tedious but vital service chores that no one wants to do and does them excellently.
She is mostly imaginary. The classical academic
fantasy is that every professor should resemble this combination
of Leonardo, Socrates, and Mother Teresa, but the reality is that
very few can pull it off - certainly not enough to populate
every engineering department. Nevertheless, requiring every new
engineering professor to be first and foremost a researcher has
become standard academic policy in the past several decades, with
dramatic effects on every aspect of academia from the makeup of
the faculty to the structure and content of courses and curricula.
You might presume that there were compelling theoretical or empirical
reasons for so many universities to adopt a policy with such profound
ramifications, and that there must be equally compelling arguments
for maintaining the policy.
You would be wrong. The usual justification
for trying to make all professors researchers is the argument
that teaching and research are inextricably linked, to an extent
that the first cannot be done well in the absence of the second.
This argument is a strange one. Its proponents - usually
academicians, trained in scientific method and the rules of logical
inference - offer it with unbounded conviction, passion,
and a total absence of evidence. They argue that only researchers
are aware of recent developments in their field, so that courses
taught by nonresearchers must be irrelevant or obsolete. They
add that nonresearchers whom students rate as good teachers must
be merely "entertainers," providing style without
substance. When challenged to produce some evidence for the linkage
between research and teaching, they name professors they know
who have both admirable research records and teaching awards,
which is like claiming that you can only be a world-class organist
if you practice medicine in Africa and pointing to Albert Schweitzer
to prove it.
In this essay I want to take a closer look
at the purported linkage between teaching and academic research,
to see how it stands up to the tests of common sense and educational
research. I will argue that it stands up to neither. Before
I get started, though, perhaps I should clarify a point. I am
not saying that research and teaching are necessarily in
conflict; I cheerfully grant that in some cases the two activities
are indeed complementary. An advanced graduate course on a currently
hot research topic, for example, is likely to be taught best by
an instructor actively doing research on that topic. There is
no logical basis, however, for requiring active research involvement
to teach an introductory course on engineering mechanics or mass
and energy balances. My discussion of teaching in the balance
of this paper should therefore be understood to relate mainly
to courses that stress engineering fundamentals, practice, and
problem-solving methodologies - which is to say, most undergraduate
Rugarcia(1) points out several distinctions:
The principal goal of research is to discover
new knowledge, while that of teaching is to impart well-established
knowledge and provide training in problem-solving. Repetition
of previous work using standard procedures may be necessary in
research but what really matters is the result. On the other
hand, prior knowledge and solution algorithms are (or should be)
the focal points of undergraduate teaching. Glossing over them
in one's zeal to get to the results misses the point.
Ability to communicate is a desirable but
not a necessary condition to be a good researcher and a mandatory
condition to be a good teacher. Some
of the most eminent scientists in history - Gibbs and Einstein
come to mind - are well known for the obscurity of their
lecturing. Their lack of clarity in presentation in no way diminished
their stature as researchers. However, an outstanding teacher
who cannot communicate is inconceivable, a contradiction in terms.
The personality traits associated with outstanding
researchers are not the same as those associated with outstanding
teachers. Most excellent researchers
are intensely involved with their work. They feel the greatest
satisfaction when performing their experiments, interpreting the
data, struggling through their derivations. Many of them feel
compelled to minimize the time they spend on activities that distract
them from their research, such as teaching: they view having
to go over old material as a waste of time and may be impatient
with students who don't get it quickly.
Outstanding teachers are more outwardly directed.
They enjoy contacts with students and may get as much satisfaction
out of delivering a good lecture or seeing a student finally grasp
a concept as out of getting an experiment or derivation to work.
They may or may not be dynamic or entertaining in lectures, but
they share a clarity of expression and convey a sense of enthusiasm
that may be noticeably lacking in their research-oriented colleagues.
It is no secret that research is a major time
sink. It takes time - preferably in large uninterrupted
blocks - to define problems, generate support, collect, read,
and understand all relevant published work on the topic, plan
a method of attack, make false starts and wander down blind alleys,
wait out the inevitable unproductive periods, clean out logical
flaws or weak points, replicate experiments, explore possible
consequences and applications of results, write papers, and give
seminars. Doing all that is under any circumstances a full-time
job; doing it well enough to gain national recognition - now
the principal criterion for promotion and tenure almost everywhere - requires
an intensity of effort that tolerates few distractions.
That excellent teaching takes just as much
time and intensity of effort is not as well appreciated. Consider
the preparation of lectures. Most course notes and texts are
written from the point of view of someone who already understands
the concepts; the trick is to find a way to make the ideas clear
to someone approaching them for the first time. Just stating
a concept is likely to be useless. To make it comprehensible
to most students, the instructor must first provide examples to
establish relevance and motivate interest, then imbed the concept
in a web of alternative expressions and visual representations,
and finally provide more examples and participatory exercises
to solidify understanding. Finding a way to do all that for just
one relatively straightforward concept can take hours or even
days - and a course contains lots of concepts.
Making up good problems is another time-intensive
chore. Students almost never learn anything nontrivial in formal
lectures; they only start to get it when they try to solve problems.
For true learning to take place, however, the problems must vary
in scope and difficulty - some drilling basic concepts, others
integrating new and prior material, and still others challenging
the problem-solving skills and creativity of the best students.
Relatively few textbooks offer problems that provide the necessary
variety and scope; the burden on the instructor is to collect
problems from several sources and to make up and work out solutions
to others. Doing so takes immense amounts of time.
Reviewing studies done before 1965, Brown
and Mayhew(2) concluded that "Whenever studies of teaching
effectiveness are made as judged by students, no relationship
is found between judged teaching effectiveness and research productivity."
Finkelstein(3) and Feldman(4) reviewed more recent research studies
and found that the correlation between good teaching and strong
research was either nonexistent or, in a minority of cases, slightly
positive. Interestingly, quality of publications (as assessed
by frequency of citation) was considerably more likely than any
other publication measure to correlate negatively with
teaching effectiveness, and individual authorship of books and
first authorship of articles also showed strong negative correlations.
The implication is that professors doing individual research
good enough to gain widespread peer recognition are least likely
to be judged effective as teachers.
Perhaps the most telling indication of the
nature of the research-teaching interaction is provided by Alexander
Astin (5) in a landmark study conducted in the late 1980's.
Astin accumulated data on faculty members and almost 25,000 students
at 309 institutions of higher education. For each institution,
he assessed the faculty's research orientation (as
measured by research publications, research funding, time spent
away from campus on research-related activities, and self-rated
importance of engaging in research and being recognized for research
achievement) and student orientation (level of interest
in students' academic and personal problems, sensitivity
to minority issues, accessibility outside office hours, opportunities
for student-faculty interaction), correlating each orientation
with a variety of measures of student performance and attitudes.
The results are striking. Research orientation
of the faculty correlates negatively with completion of the bachelor's
degree, various other measures of academic performance, and student
satisfaction with quality of instruction and the overall college
experience (p. 338). Student orientation of the faculty correlates
positively with bachelor's degree completion, overall academic
attainment, student satisfaction with quality of instruction,
and self-reported growth in preparation for graduate school, writing
skills, leadership abilities, general knowledge, and public speaking
skills (pp. 341-342). Research orientation and student orientation
are negatively correlated (p. 338).
The quantitative results of the study led
Astin to reject the assertion that research and teaching are mutually
supportive. On the contrary, he concludes that "In certain
respects, the two poles of this factor [research vs. student
orientation] reinforce the commonly held notion that, in American
higher education, there is a fundamental conflict between research
and teaching" (p. 67) and that "Attending a college
whose faculty is heavily Research-Oriented increases student dissatisfaction
and impacts negatively on most measures of cognitive and affective
development. Attending a college that is strongly oriented toward
student development shows the opposite pattern of effects (p.
Certainly there are professors who are both
good researchers and good teachers, but their presence on faculties
(and hence the occasional slight positive correlation between
research and teaching performance) proves nothing, since they
are likely to get promotion and tenure where professors who are
excellent teachers and fair or poor researchers are not. The
real question is whether an institutional emphasis on research
activity improves or detracts from teaching quality. The evidence
clearly points to the latter.
Does any professor always do an optimal
job of teaching - continually updating and improving lecture
notes, providing concrete demonstrations of abstract concepts,
making up fresh assignments and tests that cover the full range
of thinking skills and problem-solving abilities? Probably not - no
more than any professor always replicates all data points and
reads all references he or she cites in research papers. There
are simply not enough hours in the day to do everything as thoroughly
as it should be done, and so shortcuts and compromises are necessary
and inevitable in academic life. Given this necessity, the question
becomes which activity to compromise.
Here the academic system stacks the deck.
Professors at research universities who choose to emphasize teaching
are likely to experience second-class citizenship and denial of
tenure and promotion. To move up the academic ladder they must
dedicate themselves primarily to research, doing what it takes
to meet minimal local teaching standards and no more. And since
the system uses the same performance criteria for every new faculty
member, the students experience a continuing succession of instructors
who have either voluntarily or reluctantly chosen to do a poorer
job of teaching than they are capable of doing.
The low position of teaching on the academic
scale of values manifests in several ways:
Few of us routinely take the time and put
in the effort required to teach as well as we could. It
doesn't take much effort to copy derivations from notes onto a
chalkboard, or to assign problems from the text as homework and
photocopy and post the solutions from the instructor's manual,
or to throw a test together a day or an hour before giving it
without working out the solutions. It's even easier to recycle
the same notes, homework problems, and tests every time the course
is subsequently given. The material may be outdated, the lectures
mechanical, the tests familiar to the students, but at least the
cost in professorial time and energy is minimal.
Our instructional environment is less and
less conducive to learning. As
institutions place increased emphasis on research, more teaching
is done in large lecture classes or by graduate students and adjunct
faculty members, more grading is done and corrective feedback
given by teaching assistants, and more advising is done perfunctorily
or by non-faculty members (5, p. 419).
We have largely abandoned our responsibility
to be mentors and role models to our students. Just
as it is hard work to prepare good lecture notes, homework assignments,
and tests, it takes considerable effort for professors to memorize
the names of all the students in their classes and take time to
listen to students' problems when they have no time to
attend to their own. Studies have shown that students with even
one teacher who does such things are much more likely to succeed
than students who never have one. Where are those teachers supposed
to come from?
We are not functioning as professional teachers
in the way that we function as professional researchers. Most
engineering professors do not read education journals, attend
education conferences, or belong to the ASEE. They do not develop
innovative teaching methods themselves or try proven methods developed
by others (e.g. cooperative learning, open-ended questioning,
in-class brainstorming and trouble-shooting exercises). They
especially do not write undergraduate textbooks. Why should they?
The system offers few incentives to do these things and imposes
severe penalties if taking the time to do them cuts down on research
We do not practice what we teach. We
are supposedly training people to design and construct manufacturing
processes and process equipment, devise and implement control
algorithms, supervise startups, identify and overcome product
quality problems, and assess environmental impacts of proposed
processes. Unfortunately, the number of us who have ever
done any of these things is small and shrinking. Since we are
most comfortable teaching what we know best, we teach less engineering
practice and more of the engineering science we know from our
own graduate study and research. In the words of Reuel Shinnar,
"We have become the only profession taught by nonpractitioners."
Research - like most human undertakings - is
performed best when it is motivated by a strong sense of mission,
if it is "the passionate pursuit of a problem or vision
that obsesses the researcher and will not let him/her rest."(6)
Professors who work on a research problem not out of a passion
to know and understand but simply to move up the academic ladder - or
worse, to raise funds - are unlikely to produce worthwhile
research. Rather, their goal will be to produce results in quantity
and haste, publishing lots of papers that can serve as the bases
of more proposals to raise additional funds to support more research.
They will accept superficial explanations of results without
critical scrutiny, ignoring contradictions or casually dismissing
them as "outliers" or "anomalies."
A glance through any research-oriented engineering
journal - at the complex mathematical models that will never
apply to real systems, and the experimental data that will never
be needed or could easily be obtained if the need ever arose - suggests
that this situation has already come to dominate academic research.
According to a recent study, 72% of the papers appearing in leading
engineering journals were never cited(7).
Imagine the educational uses that could
have been made of the money and time spent on the research described
in those papers.
Most university administrators claim that
their faculty must be outstanding at both research and teaching
to qualify for tenure and promotion. However, very few professors
have the ability and the time to do everything required to excel
at both activities; most must therefore give priority to one
activity and content themselves with doing an adequate job of
the other. Under the existing academic incentive and reward system,
the only viable priority for most professors is research. The
result is that much undergraduate teaching is done by professors
who either have little interest in it or cannot afford to take
the time to do it well, and much research is done by professors
who would rather dedicate themselves to education if they had
the choice. The quality of both teaching and research consequently
What is the solution? It is not for chancellors
and deans to proclaim yet again the supreme importance of undergraduate
education, perhaps creating one or two new teaching awards as
demonstrations of their sincerity. Such proclamations are hollow
as long as professors who do outstanding teaching and merely adequate
research are fated to be denied tenure, or if they are tenured,
to be relegated to second-class citizenship. In the words of
William Arrowsmith, "At present the universities are
as uncongenial to teaching as the Mojave Desert is to a clutch
of Druid priests. If you want to restore a Druid priesthood you
cannot do it by offering prizes for Druid-of-the-year. If you
want Druids, you must grow forests."
Neither should we drop most academic research
and go back to undergraduate teaching as the primary business
of the university. While this solution holds some attraction - particularly
considering the amount of relatively pointless research now being
done - it is regressive. Much of the basic research that
provides long-range benefits to American industry is done at universities,
and the future of American science and technology depends on its
continuation. From a less exalted but equally critical perspective,
most universities, engineering schools, departments, and professors
now rely heavily on research funding for most of their necessary
operations. Research funds support graduate research assistants,
teaching assistants for undergraduate courses, work-study students,
laboratory maintenance, professional travel, postage, telephone
calls, photocopying, and so on. If the research support dried
up, the quality of the entire educational program would suffer
until an alternative funding structure could be put in place - a
doubtful prospect in any case, an extremely long-range one at
No, we must continue to strive for both outstanding
research and undergraduate teaching programs, and the ideal faculty
member will always be the rare individual who can manage to do
it all well - carry out world-class research, win outstanding
teacher awards, and do his or her full measure of service. When
we find such individuals, we should accord them full recognition
and reward. But what we need is an incentive and reward system
not based on the myth that there are enough of these people to
The key to a solution is provided by Ernest
Boyer in his splendid monograph, Scholarship Reconsidered (8).
Boyer observes that the professoriate has four vital functions,
or scholarships: discovery (frontier research intended
to generate new knowledge), integration (interpreting and
applying new knowledge to existing problems, multidisciplinary
research), application (applying specialized knowledge
to socially consequential problems), and teaching. He
argues that while each of these functions is critical to the continued
well-being of both academia and society, the present academic
incentive and reward system values only the scholarship of discovery.
He then proposes establishing an alternative system that makes
it possible for professors to concentrate on any of the four functions
at different points in their careers.
One possibility for such a system is to establish
two broad pathways for faculty advancement: a research pathway
and an education pathway. The system might work as follows.
The research pathway would involve activity in fundamental research (discovery) and/or applied research and multidisciplinary studies (integration) and/or socially important research, e.g. in areas such as safety and environmental engineering (application).
The criteria for advancement on this pathway
would be those in effect now at all research institutions. Professors
concentrating on research would be expected to exhibit superior
research performance, as measured by external grants received,
publications of articles in refereed journals, research monographs,
review chapters and books, citations by other authors (an excellent
and currently underused criterion), and peer evaluation. They
would also be expected to teach both graduate and undergraduate
courses and to perform at a satisfactory level in their teaching,
although they might in some instances buy themselves out of undergraduate
teaching with release time.
The education pathway would be characterized by different expectations and different criteria for advancement. Professors on this pathway would be expected to
Of all full-time faculty slots in a department,
75-85% should normally be allocated to research-pathway
positions and 15-25% to education-pathway positions. The
cost of this policy to the department would therefore be minimal.
Moreover, as discussed below, the long-range effect of the policy
could be an increase in both research productivity and departmental
Education-pathway positions should normally
be filled by known outstanding teachers or experienced engineers,
and ideally by individuals who fall into both categories. New
Ph.D.'s with no industrial or teaching experience should not be
permitted to enter the education pathway directly: allowing them
to do so would defeat the purpose of the proposed system and could
also seriously jeopardize their career development if things did
not work out. They would be eligible to switch to the education
pathway only after demonstrating their potential to meet the performance
criteria listed above. If they have no industrial experience
they might be requested to acquire some through industry-based
sabbatical leaves or summer internships, either before or after
the switch becomes official.
No distinction should exist between the
two pathways in status, perquisites, or expectations of departmental
and university service. Education-pathway
professors should have the same opportunities for merit raises,
tenure, and promotion to full professor as their research-pathway
colleagues enjoy. The sole criterion for faculty recognition
and reward should be quality of performance: no professor should
ever experience second-class departmental citizenship because
of his or her career focus on either education or research.
Consider the potential benefits of this policy:
The quality of undergraduate education would
inevitably improve. Bordogna et
al. (9) recently proposed a new paradigm for engineering education,
with features that include a heightened instructional emphasis
on design and manufacturing, experiential learning, integration
of knowledge from individual courses and disciplines, dealing
with ambiguity, problem formulation, teamwork, and the societal
context of engineering problems. The faculty structure proposed
in this paper would both accommodate and facilitate the changes
Ernst recommends. New and better instructional materials, demonstrations,
problems, case studies, and teaching methods would be developed
and tested by the education-pathway professors. The results of
these efforts could then be communicated to the rest of the faculty,
who might not have the time to develop the materials and methods
themselves but might be willing to use some of them in their own
The education-pathway professors could serve
as mentors to other faculty members who wish to improve their
teaching. New assistant professors
and instructors could co-teach their first one or two courses
with education-pathway professors, seeing first-hand what excellent
teaching looks like and being exposed to techniques they can later
adapt to their own teaching styles.
Vital departmental functions for which research
credentials are irrelevant, including course and curriculum planning,
coordinating undergraduate advising, and undergraduate program
administration, would be done expertly by people who want to do
Research-oriented faculty members, with
their lighter teaching roles and freedom from unwanted undergraduate
administrative and advising responsibilities, should be able to
increase their research productivity. Between
the revenue generated by this increased productivity, grants brought
in by the education faculty for curriculum development and undergraduate
laboratory expansion, and funds for endowed education chairs provided
by industry (which often has more of a vested interest in high-quality
undergraduate education than in graduate research), the school
and departments could in the long run experience a financial gain.
Faculty members on the education pathway
could devote themselves to undergraduate education, free of the
need to dilute their efforts with research for which they have
little enthusiasm. The results
would include an increase in the average quality of engineering
research and a decreased competition for graduate students and
laboratory space in every academic department.
The courses most closely related to engineering
practice - the undergraduate laboratory and design courses - would
be taught by people with both the background and the enthusiasm
to teach them expertly rather than by professors dragooned into
teaching them on a rotation system. More important, industry-bound
undergraduates would gain teachers and advisors who would serve
as role models of experience and professionalism.
This proposal is neither radical nor unprecedented.
At least one department at a major research university has a
named full professor whose career has been devoted to engineering
education, whose presence in that chair has greatly enriched our
profession and the stature of his own department. Another major
college of engineering with a strong research program uses a flexible
evaluation system for determining merit raises: teaching, research,
and service are given variable weights according to each professor's
career emphasis, so that professors whose primary career activity
is teaching and those who are more research-oriented have equal
opportunities to advance.
In my own department, we brought on to our
faculty several years ago a man with 30 years of industrial experience,
who made it clear that he had no interest in research. It was
one of the best things we ever did for ourselves. He has taken
over, enthusiastically and expertly, many of the responsibilities
from which most of us instinctively recoil, including administering
the undergraduate program and transforming and modernizing the
undergraduate laboratory. His technical experience has proved
to be a rich resource for both his students and his colleagues,
and his managerial background and skill have enabled him to do
gladly and well what most of us would do reluctantly and, at best,
adequately. He was recently promoted to full professor, with
no dissent on any level, and the sky has not fallen.
In short, the system I propose would ensure
that we are not requiring body repairmen on our faculties to fix
transmissions or accountants to perform surgery. Both research
and teaching would be done by individuals with the desire to do
them and the enthusiasm and skill to excel at them. The quality
of the education program, the quality and productivity of the
research program, and the morale of the faculty would all increase,
and in the steady state departmental revenues would most likely
Why not try it? What could we lose?
This paper is based on the 1992 Phillips Award
Lecture given on May 1, 1992 in the School of Chemical Engineering
at Oklahoma State University.
If I were to name all the friends and colleagues
whose ideas have influenced mine, the list would be longer than
the paper. I will limit myself to just four of them. First,
this presentation in a sense follows the 1988 Phillips Award Lecture
by Hank Van Ness, who made many of the same points more gracefully
and less verbosely. Second, Armando Rugarcia's thoughts about
the contrasts between research and teaching inspired my choice
of the theme of the paper, and I have drawn heavily on those thoughts
and many more of Professor Rugarcia's insights about engineering
education when writing it. Finally, my thanks go to Rebecca Brent
and George Roberts, whose comments on a preliminary draft of the
manuscript led to substantial improvements in the final version.