Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
CHAPTER Although postdoctoral education in the United States has been in existence almost as long as graduate education, very little quantitative information exists that describes the scope and intensity of the enterprise. In view of the large amount of educational data that has been com- piled in the past, it is surprising that so little attention has been paid to post- doctoral study. Most of the data we have pertain either to particular fields or to particular programs of support; if all of these sources were assembled the record would remain incomplete. A basic difficulty in securing information on postdoctoral education results from the fact that no formal conferring of a degree or certificate marks the completion of the postdoctoral experience. This is not to argue the desirability of such a recognition of accomplishment, but rather to suggest that in a pro- fession where milestones are easily counted, postdoctoral activity takes place so unobtrusively and ends so indeterminably that little note is taken of the event except by the participants. The lack of documentation also springs from a lack of consensus as to the purpose of postdoctoral education. Postdoctoral education has grown almost spontaneously (and independently) in many segments of the universities and in nonacademic environments. At most institutions there is no coordination and no contact between the postdoctoral activity in one field and that in another. As Robert Alberty, Dean of Science at MIT, remarked in a speech to the National Research Council in the spring of 1968: "The graduate students 39
40 THE STUDY have their deans and graduate deans have their national organizations. Many universities have Vice Presidents for Research, but very few people in univer- sities have any formal responsibilities for postdoctorals." The lack of consensus is reflected in (and partially caused by) the numerous and uncoordinated agencies, foundations, health organizations, and profes- sional societies supporting postdoctoral education. Each has its own well defined purpose (if the agency is complex there may be several different well defined purposes), but there is not necessarily agreement among these groups with regard to motivation. Sometimes the differences are sufficiently great that some program officers are unaware that they are supporting postdoctorals as such. The practice of hiring postdoctorals to work on research projects at univer- sities, supported by extramural (usually federal) funds has contributed greatly to the absence of statistics on the magnitude of the postdoctoral population. Not only do the employing institutions often fail to make a distinction be- tween the postdoctoral research associate and the other professional and semi- professional staff being paid from these restricted funds, but some granting agencies are indifferent to the backgrounds of those the professor selects to work with. One program officer asserted that his responsibility was to pur- chase research as efficiently as possible; who was hired to do the work was not his concern. The result is that, with the exception of one or two federal agen- cies, no count has been made of the number of people at each education level who have been paid from agency funds. The new annual inventory of person- nel being carried out by NSF for the interagency Committee on Academic Science and Engineering will supply these data in the future.1 For the past and present, however, such information is unavailable. The Available Facts The most comprehensive previous examination of the postdoctoral situation was made by Bernard Berelson2 in 1960 in preparation for a report to the As- sociation of American Universities. Berelson visited some 16 campuses, sent questionnaires to the forty-odd member institutions of the A AU, and held discussions with representatives of a number of federal agencies. In lieu of hard data, except for a few national fellowship programs, Berelson applied the 1 Unfortunately the information requested of universities in this CASE Phase II study will not include those postdoctorals whose stipend is paid by a nongovernmental source, but whose research expenses are supplied from the mentor's federal contract or grant. Bernard Berelson, Postdoctoral Work in American Universities, Journal of Higher Edu- cation, Vol. XXXIII, No. 3, March 1962, pp. 119-130.
41 DEFINITION OF POSTDOCTORAL APPOINTMENT formula of one postdoctoral research associate for each $100,000 of federal research funds at universities. With this and other rough approximations he es- timated that in 1960 there were 8,000 postdoctoral appointees in all fields at universities. Another study was made in 1958 by Dr. Arthur S. Cain and completed, after his death, by Lois G. Bowen.3 They reported on a comprehensive ques- tionnaire study made of the system of medical fellowships and their impact on both the recipients and the medical institutions. Valuable in its limited areas of concern, the study is now out-of-date. The decade that has passed has seen tremendous growth, and the climate in medical schools has changed radically. An unpublished pilot study examining many aspects of postdoctoral edu- cation at eight universities was undertaken by the National Science Founda- tion in 1965. Studies at individual universities have been made by H. W. Magoun at UCLA, Robert Alberty at Wisconsin, G. M. Almy at Illinois, and John Perry Miller at Yale. In addition Myron Rand has written a short history of ihe National Research Council Fellowships describing the development of that important program that "contributed to the spectacular rise from mediocrity to world leadership in scientific research which the United States has accom- plished during the one generation in which the fellowship experiment has been in progress." In none of these studies is there an overview of the extent and nature of postdoctoral activity in the United States. The present study was undertaken by the National Research Council to provide that overview. The first task was to establish the boundaries of the universe to be investigated. This was no easy task, since the definition of postdoctoral education was really to be the con- clusion of the study. Definition of Postdoctoral Appointment Strictly speaking, postdoctoral is an adjective that pertains to an individual who has attained the doctor's degree. Thus, a postdoctoral appointment in precise terms refers to any formal position to which a person is appointed fol- lowing his completion of the requirements for a doctor's degree. The word would most naturally be contrasted with predoctoral. Arthur S. Cain, Jr. and Lois G. Bowen, The Role of Postdoctoral Fellowships in Aca- demic Medicine, The Journal of Medical Education, Vol. 36, No. 10, Part 2, October 1961, pp. 1357-1556.
42 THE STUDY However, just as predoctoral is generally limited to describing that period before the doctorate but after the baccalaureate, the common usage of post- doctoral is restricted to those holders of the doctorate who are pursuing some special experience or training beyond their formal predoctoral work. Exactly which experiences should be included or excluded is a question on which there is little agreement, although most observers would admit that holders of fellowships who are carrying on research in association with a senior investiga- tor represent the paradigm. The problem of definition is complicated by the age and field of the individual, the variety of titles used, the institution at which the appointment is held, and, most critically, by the ambiguity in the purpose of many appointments. We shall discuss these difficulties after pre- senting the following definition used in this study: This study is concerned with appointments of a temporary nature at the postdoctoral level that are intended to offer an opportunity for continued education and experience in research, usually, though not necessarily, under the supervision of a senior mentor. The appointee may have a research doctorate (e.g., PhD, ScD) or professional doctorate (e.g., MD, DVM) or other qualifications which are considered equivalent in the circum- stances. A person may have more than one postdoctoral appointment during his career. In its inquiries, the Committee on Academic Science and Engineering of the Federal Council for Science and Technology inserts the restriction that the man be within five years of his doctorate. We have avoided such a restric- tion for several reasons. If we set a limit that might be appropriate in the case of PhD's, many MD's who take postdoctoral work following their internships and residencies would not be included. Furthermore, in fields such as the hu- manities and social sciences, the pattern is to delay postdoctoral work until a period of time has passed. Some of these people would also be missed. Finally, in terms of the impact on institutions, it makes little difference if the occupants of laboratory benches or library carrels are just out of graduate school or have been employed elsewhere for some time. Our data permit us to distinguish among the age groups when it is important. The first key word in the definition is temporary. There are a number of temporary postdoctoral appointments that we want to exclude or at least to amplify the conditions under which they may be included. The first is the appointment to instructor or assistant professor. These appointments are gen- erally temporary, but ordinarily should not be considered within our defini- tion, since they are understood to be part of a regular series of academic ap- pointments and lead, if all goes well, to permanent positions. On the other hand, at some institutions a person may be given a fractional professorial ap- pointment with the remainder of his support coming from a fellowship. Such people will be included in our study.
43 DEFINITION OF POSTDOCTORAL APPOINTMENT Another variation on this theme is the practice, especially in mathematics, of creating named instructorships, such as the Moore Instructorships at MIT, the Pierce Instructorships at Harvard, or the J. Willard Gibbs Instructorships at Yale, where young PhD's are given a reduced teaching load so that they may concentrate their efforts on research. It is likely that these people should be included in the study, although it is not known whether all such individuals or departments responded to our inquiries. Another temporary appointment is that of the Visiting Professor. In this case, although the individual has often accepted the appointment to make use of the research facilities and professional contacts at the host institution, we have excluded him from our definition if he is filling a regular faculty position in the host institution. Our reasoning here is that his impact upon the budget and facilities of the institution is small; the faculty member he temporarily replaced would have used essentially the same resources. The effect of this decision on our part is to reduce the apparent number of postdoctoral positions in those fields where, for lack of extramural funds, other postdoctoral oppor- tunities are rare. It may be one reason for the low representation of postdoc- torals in the humanities and in the social sciences. Some temporary appointments are so short as to be little more than visits. These clearly are not relevant to the study. However, it is less clear how long the visit must be before we become interested. The critical question is whether the duration is sufficient for research to be accomplished. We decided that the criterion ought to be whether a formal appointment has been made by the host institution. Another ambiguous group is what Clark Kerr4 has called the "unfaculty." These are the professional research personnel who are more or less perma- nently appointed to the research staffs of institutes and departments of univer- sities without having regular faculty appointments. At some institutions a parallel structure of research faculty appointments is established through which these people may progress without ever attaining tenure or other fac- ulty privileges. This group overlaps in an irregular and indefinite way with the postdoctoral population to the degree that it is difficult to draw the dividing line. From the point of view of the supporting federal agency and the director of the research group, both the professional research staff and the postdoc- torals are appointed to perform research under the rubric of the contract. There is no explicit intention in either case that the appointment provide an opportunity for continued education and experience in research, although this opportunity exists. The distinction between the unfaculty and the post- 4Clark Kerr, The Uses of the University, Harvard University Press, Cambridge, Mass., 1964, p. 67.
44 THE STUDY doctoral is in their respective perceptions of their goals and purposes. The postdoctoral is one who intends to leave the position after an interval, having received the continued education and experience in research that he sought. The second key word in the definition is research. There are a number of types of temporary positions that have the character of apprenticeships. These include internships and residencies for physicians, clerkships for lawyers, teaching internships in liberal arts colleges, and administrative internships in in the major universities. None of these is a postdoctoral appointment in our sense unless research training under the supervision of a senior mentor is the prime purpose of the appointment. Related to the restriction to research is the problem of the second doc- torate. What is to be done in the case of the physician who seeks a PhD or the PhD who heads for a professional doctorate? It was decided to admit the man to postdoctoral status if research was his main activity. This has the ef- fect of denying this status to the man seeking the professional doctorate (e.g., medicine or law) and of granting it to the physician pursuing the PhD degree. The situation for the young medical doctor is further complicated by the fact that some sources of support do not make the research distinction that we do. Thus, a man may hold a postdoctoral traineeship from the National Institutes of Health to obtain training in research or to obtain training in clini- cal practice. The former we include; the latter we do not. Up to this point, much of the discussion has dealt with the university scene. In industrial, governmental, and nonprofit laboratories and libraries around the country there are positions similar to those described above in the univer- sity environment. When such positions in nonacademic organizations have the character and objectives of postdoctoral appointments in the universities, we have included them in the study. Regardless of the host institution, a major problem in identifying postdoc- torals is the bewildering array of titles that are attached to them. Although there are only four basic types of postdoctoral appointments (see page 86 for fuller discussion), the titles are often unrelated. A man supported by a fellow- ship generally has the word "fellow" in his title: however, a man supported on faculty research funds may be called a "fellow," a "research fellow," a "re- search associate," a "research assistant professor," etc. At many institutions a research associate is a young postdoctoral supported by faculty research money, while at the California Institute of Technology a research associate is a dis- tinguished visiting scholar who does not teach, regardless of his source of support. Differences in semantic usage have made for difficulties in collecting data. When asked how many postdoctoral students were in his department, one chairman answered, "None," when, in fact, his department leads the country
45 STRATEGY OF THE STUDY in the ratio of postdoctorals to faculty (3:1). At his institution postdoctoral s are counted among the faculty and consequently are not students. In an attempt to anticipate some of the ambiguities mentioned above, a list of explicit exclusions and inclusions was attached to the definition distrib- uted along with all questionnaires and inquiries. This list is reproduced as follows: EXCLUSIONS 1. Although appointments to Instructor and Assistant Professor are temporary, they are excluded because they are understood to be part of the regular series of academic appointments and lead, if all goes well, to a permanent position. 2. Visiting professor appointments are excluded if they fill regular places in the host institution's academic staff. 3. Service Research appointments which are not intended to provide an opportunity for continued education in research are excluded. 4. Internships and Residencies are excluded unless research training under supervision of a senior mentor is the prime purpose of the appointment. 5. Holders of a doctor's degree who are studying for another doctorate that does not involve research as a primary activity are excluded. INCLUSIONS 1. Postdoctoral appointments, supported by whatever funds, that provide an oppor- tunity for continued education and experience in research are included. 2. Scholars on leave from other institutions are included if they come primarily to further their research experience. 3. Appointments of holders of professional doctoral degrees who are pursuing re- search experience are included even though they may be candidates for a second doctoral degree. 4. Appointments in government and industrial laboratories that resemble in their character and objectives postdoctoral appointments in the universities are included. 5. Persons holding fractional postdoctoral appointments are included. For example, a postdoctoral fellow with a part-time Assistant Professorship is included. 6. Appointments for a short duration are included if they are of sufficient duration to provide an opportunity for research and a formal appointment can be made. Strategy of the Study In order to provide information and opinions from the whole spectrum of persons connected in some way with postdoctoral education, we found it necessary to use a wide variety of instruments and techniques to sample the pertinent components of the population. Depending on the nature of his in- volvement, an individual may have been asked to respond to a formal question- naire, to an invitation to record free replies to broad inquiries, or he may have
46 THE STUDY been approached through an interview. The interviews included single indi- viduals and groups. A number of conferences were held following speeches by the Director and by members of the Advisory Committee and many reactions were obtained. It is felt that opinions of most major groups have been sampled. Two kinds of information have been gathered: factual counts of numbers of individuals, institutions, and responses; and statements of opinion. The former were required simply to provide scope to the study; the latter to place the scope in context. Let us first consider the instruments used to collect the facts. The fundamental question is: How many postdoctorals consistent with our definition exist? Immediately associated with that question are many others. Where are they located? In what fields do they work? Where did they get their formal education? What is their citizenship? By whom are they supported? What is the nature of their support? How much remuneration do they receive? What is the nature of their activities? Why did they seek such an appointment? What are their future plans? etc. Although some of this information can be partially gleaned from federal agencies and private funding sources, most of the data did not exist. For example, most agencies have only fragmentary in- formation on the number of postdoctorals supported on research grants, since the receptor institutions are allowed some freedom in the selection of the kinds of personnel hired with these funds. We decided that only a census of postdoctorals would permit us to answer the questions posed. Adequate re- sponses were received from 10,740 postdoctorals and we estimate that the total postdoctoral population in the spring of 1967 numbered 16,000.5 Another major question concerns the nature of the environment within which the postdoctoral is working and where he is likely to be employed fol- lowing his present appointment. In both cases the location is probably an insti- tution of higher education (as is evident from the postdoctoral census data). Accordingly, a questionnaire was designed to be answered by departmental chairmen to discover the answers to such questions as: How many faculty of what rank are in the department? What kind of background do the faculty have? How many graduate students are enrolled? How many graduate degrees are awarded? What positions do their doctoral recipients fill following their degrees? How many postdoctorals are in the department? What are the depart- mental policies regarding the postdoctoral? etc. Returns were received from 4,040 departments in 357 schools.6 The technique used in carrying out the census and the way in which the rate of return was estimated are discussed in Appendix A-l. In view of the uncertainties in the estima- tion procedure the estimate of 16,000 postdoctorals could be wrong by as much as 2,000 in either direction. See Appendix A-2 for details on sampling procedures and for analysis of the returns.
47 STRATEGY OF THE STUDY Faculty members play central roles in postdoctoral education. By directing inquiries to the faculty, views were sought from mentors of postdoctoral s and from those who, although involved in research, had no postdoctorals in their research group. Both views are important since the present evolution of post- doctoral study does not meet with the approval of all faculty. Answers were sought to such questions as: What is the composition of research groups in terms of graduate students, postdoctorals, professional staff, faculty co- workers, etc.? How many recent graduate degrees have been produced from the group? For what reasons are graduate students urged to take postdoctoral study? In what way do the various kinds of members of the research group contribute to the research and teaching? Does the nationality of the postdoc- toral make a difference? What are the time and space requirements of a post- doctoral compared to a graduate student? etc. Completed questionnaires were received from 2,195 postdoctoral mentors and from 564 doctoral mentors without postdoctorals in their research groups.7 The administrative point of view was elicited through an open-ended ques- tionnaire (see Appendix A-5) that was sent to each of the universities having postdoctorals. Questionnaires were sent to 165 schools and replies were re- ceived from 125. The many agencies and private organizations that support nationally com- petitive fellowship programs were asked three questions: How many fellow- ships in what fields have been awarded since the inception of their program? What was their budget for postdoctoral fellowships in fiscal year 1967 (July 1, 1966 to June 30,1967)? What purpose were they seeking to fulfill with their program? In addition we have had commentary from directors of nonprofit, govern- ment, and industrial laboratories on the effect of the growth of postdoctoral education on their activities. Interviews have been held with program officers in the several federal agencies supporting the bulk of the research in universi- ties to determine the part that consideration of the postdoctoral plays in their awarding research grants and contracts to universities. Twenty universities were visited and conversations were held with deans, departmental chairmen, faculty members, postdoctorals, and terminal doctoral candidates. Numerous discussions have been held with knowledgeable people in and out of the fed- eral government and close coordination has been maintained with a number of other related studies being carried out in the National Academy of Sciences and elsewhere. One other investigation has been made to determine the value of postdoc- toral education. Many observers are of the opinion that, for the most part, those who seek and receive postdoctoral appointments are among the better See Appendix A-3 for sampling details and for analysis of the returns.
48 THE STUDY doctoral recipients. This study confirms that opinion on the average. Separate studies such as the report of the Commission on Human Resources and Ad- vanced Education8 demonstrate that, as measured by the rate at which pub- lished work is cited by others, former postdoctoral s do more important re- search than those researchers who have not had postdoctoral appointments. The combination of these two concepts leads to the rather obvious conclusion that better PhD's do better research. Whether the postdoctoral experience is relevant to the subsequent success is left in doubt. We have attempted to improve on existing data by selecting two samples of doctorate holders of apparent equal quality. A group of former postdoc- torals was matched with an equal group of non-former-postdoctorals that was similar with regard to field distribution, to the "quality" of the PhD institu- tion,9 to the time lapse between the baccalaureate and the doctor's degree, and to the age of the individual. These two groups were sent questionnaires and citation information was gathered from the Science Citation Index. These then, in addition to published documents, are the inputs to the study. The exposition of our results and conclusions are found in the chapters that follow. 8Human Resources and Higher Education, Russell Sage Foundation, New York, in press, 1969. Allan Cartter, An Assessment of Quality in Graduate Education, American Council of Education, 1966. See Appendix A-4 for sampling details and response rates.