Postdoctoral Education in America
by Steven B. Sample
President, University of Southern California
September 23, 1993
An address delivered at the annual meeting of the Association of Graduate Schools held at Chapel Hill, North Carolina.
I am here to discuss postdoctoral education, and why I believe it is time for the Association of American Universities and its daughter organization, The Association of Graduate Schools, to undertake a thorough study of the academic postdoctorate in America.
The AAU was formed in 1900 in response to a need to set standards for graduate education in general and doctoral education in particular. Postdoctoral education is a natural extension of that original mission of the AAU. I believe postdoctoral education is an arena in which we can be proactive and visionary, and perhaps even constructively influence the course of higher education in the U.S. and the world in the century ahead.
Postdoctorals have been a part of American higher education for over 100 years. The Johns Hopkins University began to support postdoctorals shortly after the institution was founded in 1876. Postdoctoral education grew only modestly during the first half of the 20th century. But the advent of the Cold War brought with it a boom in postdoctoral researchers. In fact, in the 1970s and 1980s, the number of men and women pursuing postdoctoral education increased more rapidly than the number working toward a doctorate. Thus the real increase in advanced graduate education over the last 20 years has been at the postdoctoral level.
There is reason to expect that this growth in postdoctoral education will continue. For one thing, there is an increasing market?driven need for specialization. The postdoctorate also serves as a holding pattern for new Ph.D.'s in a tight job market. Moreover in many fields, and most especially in the natural sciences, new Ph.D.'s are not yet ready to do truly independent research. Thus, for a young person in one of these fields who is interested in becoming an independent researcher, it is almost imperative that she complete a postdoctoral appointment.
The number of postdoctoral scholars in America is an elusive figure. Why? Because postdoctorals are largely an invisible group in American universities today, as were doctoral students at the turn of the century when the AAU was founded. Data on postdoctorals is the most unreliable of the figures that we collect in higher education, because postdoctoral students are counted differently from university to university, and even from department to department in the same university. For example, some universities count clinical fellows in medicine as postdoctoral students, while others count only Ph.D. postdocs. The National Science Foundation and the National Research Council both monitor the number of postdoctorates. But the NSF includes postdoctoral fellows from abroad and excludes those outside the sciences and those in industry and national laboratories, while the NRC excludes fellows from abroad but includes postdoctoral fellows outside the sciences and outside academia.
Obtaining good data on the financial investment being made in postdoctoral education is equally difficult. We don't know how many dollars are being spent for salaries and fringe benefits for posdoctoral students; such expenditures are never cited in state or university budgets. We know practically nothing about the salary structure for postdocs. What do universities pay people with Ph.D.'s who are not exactly students, but who are not exactly faculty either? Do postdocs receive adequate compensation? What does the term "adequate compensation" mean in this context? How do salary levels for postdocs vary from university to university, and from discipline to discipline? How do these salaries vary within the same institution? And of increasing importance, do we compensate domestic and international postdocs at the same level? (There is a suspicion that international postdocs are in many cases a form of slave labor within the American research establishment.)
Senior academic officers at most universities can't begin to talk in an informed way about postdoctoral education on their own campuses. How many of you came to this meeting knowing the number of postdocs at your own institution? How many of the postdocs on your campus are men? How many are women? How do they divide along racial and ethnic lines? What age groups do they represent? In what disciplines are they working? How are they supported? How many are domestic and how many are international students? What fraction of your postdocs earned their Ph.D.'s at your institution? What fraction earned their Ph.D.'s at another AAU institution or at some other U.S. university? Many of us could answer these questions fairly readily with respect to our doctoral students, but most of us would not have the foggiest idea of where to get answers with respect to postdocs.
The Chronicle of Higher Education, in its annual almanac issue, does not even mention postdocs, either as students or as faculty. And with good reason – it is tough to track people who do not fit into a common, recognized category. There is no standardized title for postdocs in American higher education. Sometimes they are called lecturers, sometimes instructors, sometimes research associates, and sometimes nothing at all. Some postdocs are categorized as staff, and others are considered faculty.
For purposes of definition, I do not think we should confuse the academic postdoctorate with clinical fellowships and residencies in the health sciences. In my judgement, the definition of a postdoc is a student who has just completed a Ph.D., and who is doing full?time research under the general oversight of a senior professor for a year or two prior to taking a permanent research position in academe, industry or government.
Table I comprises a list, based on NSF data, of the top 40 schools in terms of enrollment of science and engineering postdocs. As you can see, there are just over 31,000 postdoctoral students in the United States. Look at the extraordinary fall?off in the numbers of postdocs at each university as you go down through the list. Of the top 40, all but five are AAU institutions; moreover, those top 40 enroll about two-thirds of all the science and engineering postdocs in America.
The AAU – with 56 U.S. members out of 3,400 colleges and universities in the country – now produces two-thirds of the nation's postdocs. I think that puts a special responsibility on us to show some leadership in the field of postdoctoral education.
Let me caution again that the numbers in the table are probably very soft. I suspect that the counting of postdocs is relegated to an inexperienced clerk in most academic institutions. I would be the last person to believe that there is any real consistency among these data. I can look through this list and express grave doubt that the reported numbers at certain institutions are based on my definition of a postdoc ? that is, a person who has recently earned a Ph.D. and who is doing full?time research under a senior professor for a year or two before taking a permanent research position in academe or industry.
Figure 1 shows that the big growth over the last 15 years in postdoctoral education in America has been among international students. In science and engineering, the number of domestic postdoctoral students has stayed relatively constant, going from roughly 13,500 in 1977 to 15,500 in 1991. Over that same period the number of international postdoctoral students in American universities has nearly tripled, from about 6,000 to 16,000. Foreign nationals now comprise the majority of all science and engineering postdocs in the United States. Again, these numbers are probably soft, but my guess is that the fraction of postdocs in America who are international students may be even higher than these numbers indicate.
From what I can tell, a large number of the postdocs at my own institution are from abroad. Some groups of former USC postdoctoral students from overseas even have their own alumni associations in their home countries. They meet regularly to exchange ideas and resources. What brings them together? The fact that they were postdocs at USC? No. These people come together because they were all postdocs under one professor, a man by the name of George Olah, who is a well?known professor of chemistry, a member of the National Academy of Sciences, and the founder and director of the Loker Hydrocarbon Research Institute at USC. Many young Ph.D. chemists from around the world consider it a privilege to come for a year or two to Olah's laboratory at USC to widen their experience. They then return home and form "George Olah alumni clubs." I am told that Professor Olah will soon go to the Far East to talk to these "alumni" ? and perhaps encourage them to contribute funds to support other young postdoctoral students.
Up until now, when we in the academy have discussed international students, our focus has always been on undergraduates and graduate students. Do you remember how surprised many of us were to learn that most doctoral students in science and engineering at our most prestigious institutions are now international students? And do you remember it was somebody else ? a congressman or federal staffer ? who had to tell us? Now we know that the same thing holds true at the postdoctoral level. That fact alone merits our careful attention.
There are some very touchy issues surrounding the question of international postdoctoral students. We can certainly identify some advantages to having these students on our campuses. They help create an international network of scientists, which is a good thing intellectually and good for our domestic students. They advance our own scientific capability and serve the U.S. national interests in so doing. Some of the top scientific talent in the world comes to this country in order to pursue postdoctoral training, and then decides to stay here permanently; one cannot imagine a more highly selective and salutary form of immigration. Finally, international postdoctoral students provide a good deal of low-cost talent for our universities. They often serve as teachers in our research laboratories by supervising both doctoral and masters students.
But there are some potential disadvantages also. International postdocs may be just one more example of how the United States is supporting basic research for the entire world. The American taxpayer may be footing much of the bill for the education of advanced scientific talent for other countries. Scientific and technological expertise is clearly flowing out of our country when postdocs return to their homelands ? perhaps even more so than in the case of international students who earn undergraduate and graduate degrees at U.S. universities.
Moreover, there appear to be basic differences in the expectations and interests of domestic and international postdoctoral students. Domestic students are likely to use the postdoctoral experience as a stepping stone to a permanent position in academia. International postdocs, on the other hand, want most of all to become part of a scientific network, and they come here primarily to get plugged in with the American scientific establishment.
I believe there is one overarching question in regard to international postdocs: Is this the channel through which the major part of technology transfer takes place from our country to other parts of the world? International postdocs who return to their laboratories and universities back home may represent a much more rapid dissemination of U.S. research knowledge in those countries than any amount of publication in scholarly and scientific journals.
Aside from the fact that most of our postdocs come from abroad, we also know that postdocs are concentrated in the sciences. The typical ratio of scientific postdocs to engineering postdocs is 10 to one. The typical ratio of scientific postdocs to postdocs in the humanities is on the order of 100 to one. At USC, and I imagine at your universities as well, the heaviest concentration of postdocs is in the life sciences, followed by chemistry and physics. Of course, these facts are not surprising, because public and private research dollars are overwhelmingly directed to research in the sciences, and thus a form of support exists for postdocs in the sciences that simply does not exist in the humanities.
In America today, independent research by young people in the sciences is taking place much more at the postdoctoral level than it is during a student's doctoral work. In most disciplines outside of science and engineering, the doctorate is, and will continue to be, the highest level of graduate education. But in the sciences, and especially in the natural sciences and engineering, it is the postdoctorate that is becoming the de facto terminal credential. The Ph.D. in these fields, while still an important rite of passage, is becoming increasingly only an interim milestone in the development of an independent researcher. Moreover, the nature of the educational experience for doctoral students in the sciences is being changed by the presence of postdoctoral students in academic laboratories, just as the nature of undergraduate education in research universities has been changed over the years by the presence of graduate students serving as teachers.
I think this situation is not unlike that which has already occurred in the profession of medicine. We know that the medical residency has displaced the M.D. degree as the true preparation of an independent practitioner of medicine. Not so long ago, when a young person had finished his M.D. degree, he went directly into medical practice with perhaps a one?year internship intervening. But now, a three? to five?year residency is required before a young M.D. can enter the independent practice of medicine. In like manner, it is now virtually impossible for a young Ph.D. in the natural sciences to obtain a faculty position at a major university without first completing a one?year to two?year stint as a postdoctoral student.
Postdoctoral education at the end of the 20th century is in a somewhat similar position to the early Ph.D. programs of the late 19th century. One hundred years ago, doctoral programs were in their infancy and were largely unnoticed in the American academy. They were seen as a haven of spontaneity, specialization, and focused individual development. Ph.D. programs were subject to very little standardization and control, and thereby offered great freedom for gifted young people who wanted to do original research. Graduate education was being spurred on by increasing specialization in the academy, and the Ph.D. soon began to supersede the baccalaureate and master's degrees as the requisite entrance credential for appointments to the faculties of major universities. These developments at the turn of the century led to intense debate among academic leaders as to the purpose of, or even the need for, doctoral programs.
Today, the times seem ripe for a similar debate about postdoctoral education. Views differ sharply on what postdoctoral education represents, what it should represent, and what it is worth both to the university and to the individual participant. Some see postdoctoral education as a source of inexpensive talent. Some think it is divisive, creating disparity between the sciences and the humanities. Some argue that postdoctoral education is detrimental to the individual because it prolongs the time before she can enter a permanent professional position, thus delaying her productivity and her reaping of financial rewards.
Today there are voices calling for standardization of the postdoctorate, while others say that formalized postdoctorate programs are unnecessary and, in some cases, counterproductive. Some feel the postdoctorate is simply a holding pattern for those who can't find permanent employment, and that postdoctoral programs will wither away when the job market improves. But I don't believe the postdoctorate is simply a passing fancy in American higher education. I think it is here to stay ? and will grow in importance.
At USC we are beginning to make postdoctoral education a high priority. We want to build up and build on the strength of our postdoctoral programs. We want to recruit aggressively, and in a focused way, a larger proportion of the very best postdoctoral students.
We are asking ourselves what titles postdocs should have, and to what extent they should be treated as junior faculty. We are struggling with the question of compensation and benefits, and whether there should be some uniformity of compensation for postdocs (because just beneath the surface there is that slave labor question to which I referred earlier). When it comes to fringe benefits, we find that postdocs want health care benefits and don't care at all about retirement benefits. So we are looking for ways to give them what they want and save money in other areas.
I believe the AAU and the AGS should take responsibility for assisting the development of postdoctoral education. AAU's membership committee has already begun collecting data on postdoctoral enrollments and is now considering postdoctoral education as one factor in the assessment of potential new members for the AAU. Here again we see parallels with the evolution of medical education. As various groups assess the quality of medical schools, they increasingly look at the quality of residency programs, rather than simply at the quality of M.D. programs, because the M.D. has become only a way?station on the road to the residency.
In my judgement we must begin with a thorough study of postdoctoral education. So far as I know, this is an area which has never been studied in any depth at all in this country. Over the past year The Chronicle of Higher Education ran not one article on postdoctoral education.
Let me relate a little story that underscores this point. When I was preparing this talk, I asked USC's vice?provost for research to help me locate some figures on postdoctoral education. He decided to call Quantum Research, an independent analysis firm that disseminates information for the NSF. He was soon connected with the young woman who is in charge of compiling all the figures on postdoctoral education. She was thrilled to talk to him. She said, "You know, I've been working on this project all these years and I've never received one phone call about it until now. You're the first person ever to ask me a question about these data. I didn't know anybody was even reading this stuff. You've really made my day!"
I think we ought to begin this study by first trying to learn the history of the postdoctorate in America. Second, we should ascertain trends in postdoctoral education over the past two or three decades and projections of participation rates for the future. We ought to try to study the impact of postdocs on universities in general and on individual disciplines in particular. We should know something about the cost of postdocs and who pays those costs. We need to know the range of postdoctoral stipends paid by various disciplines and various universities. We need to study the impact of postdoctoral education on a person's prospects for permanent employment. We should explore the impact of international postdocs on U.S. universities and on the American research establishment. And finally, our study should look at the typical time limits placed on postdoctoral appointments; our own vice-provost for research tells me it is theoretically possible for a person to come to USC as a postdoc and retire in that position 40 years later!
I have suggested to the leadership of the AAU and the AGS that an appropriate committee to conduct a study of postdoctoral education could be relatively small, and should include presidents and chancellors, provosts, graduate deans, and perhaps a few full?time faculty who have supervised large numbers of postdocs.
There is no question that postdoctoral students play a significant role in the life of our institutions and our country. They advance our research mission; they assist in the recruitment and training of doctoral students; they are the academicians of the future; and they help maintain America as the scientific research center of the world.
Those universities that take postdoctoral educational seriously today are likely to emerge as the leading academic institutions in the early decades of the next century. Collectively, I believe it is incumbent upon us as AAU and AGS institutions to lead the way by exploring the issues surrounding postdoctoral education and by helping to delineate the course that postdoctoral education should take in the future.