The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director] (1969)

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.

10 CHAPTER Conclusions and Recommendations It is often said that research and graduate educa- tion are inextricably related. For predoctoral work this statement is most ap- plicable to the terminal or dissertation stage. However, there is no question but that the statement is true of postdoctoral education. In fact, it is fair to say that research and postdoctoral education are virtually identical. The validity of this description accounts for both the successes and the problems of postdoc- toral education as it has developed in this country. Proficiency in conducting research in most of the sciences is learned, or at least improved, in an apprenticeship to a master researcher. For a few who are exceptionally able and who take their graduate work with such a master, the graduate experience is sufficient to convert them from novice to proficiency status. For many, a longer apprenticeship is required. What form this extended experience should take depends, according to conventional wisdom, on the goal the apprentice seeks. If he desires to teach in an undergraduate college, he may want some teaching experience; further research is not as important. If he plans a career in industry, it might be wise to attach himself immediately to an indus- trial research laboratory where he can learn the appropriate styles of applied or project-oriented research by working with those who are committed to it. For the man who wants to become a master researcher, i.e., to train other stu- dents in research by joining the faculty of a graduate-degree-granting university, the postdoctoral appointment is the common route to follow. The problem with the above prescriptions is that they are too neat. As we have seen, only in some of the fields is postdoctoral work a major enterprise 241

242 CONCLUSIONS AND RECOMMENDATIONS and a prerequisite for employment in even the better universities. Some indus- trial and government laboratories find that they prefer employees with post- doctoral backgrounds. In fields such as engineering, many departments seem to want faculty with "postdoctoral" experience in industry. In short, we are dealing with a complex phenomenon concerning which every statement must be qualified. However, overemphasis on the exceptions should not be allowed to obscure the pattern. In the main, in fields like physics, chemistry, modern biology (in- cluding biochemistry), and medicine, postdoctoral education is virtually a neces- sity for subsequent employment in a highly research-oriented university. Further- more, the reasons are not simply that the postdoctoral system serves as a sieve that removes the less able, but that something positive happens and that the man who completes postdoctoral study is a better researcher than he was before. He has become better prepared and more likely to succeed as a teacher of grad- uate students. Whether other fields should embark on postdoctoral activities or expand them is a matter that must be decided field by field. There is danger of blind imitation, which should be avoided. The criterion should be whether only by postdoctoral study can the PhD recipient be expected to perform independent research in his chosen area of investigation. If the graduate or even the under- graduate curriculum can be arranged to make this unnecessary, then it ought to be so changed. Postdoctoral education should not be established to circum- vent a needed alteration of predoctoral training.1 Conversely, we find no evidence that postdoctoral education has resulted from a failure of graduate education to fulfill its function. One need only read the Proceedings of the Association of Graduate Schools, going back to the turn of the century, to realize that many of the problems and criticisms of graduate education are seemingly insoluble and unanswerable. If the date were not printed on the page, one would find it difficult to establish the year by the tenor and content of the discussion. As Berelson seems to imply,2 what is im- portant is the awareness of the problems; perhaps no solutions exist. If the function of a graduate education is to produce a finished independent re- searcher, it has always failed in some fields. It would be more surprising if it had succeeded not only today but even earlier. There is a tendency to look at the growth of knowledge today and to explain postdoctoral education in terms of the impossibility of absorbing all that need be learned during a graduate program of standard duration. There is a concomitant tendency to look back 1There is a special place for postdoctoral work when the field is undergoing a rapid evolu- tion. The recent surge of interest in mathematical methods in some of the social sciences, for example, has outstripped the ability of the schools to reorganize their curricula to cope with the change. 2 Berelson, Graduate Education in the United States, p. 41.

243 CONCLUSIONS AND RECOMMENDATIONS to earlier times and to conceive of them as simpler and of science then as being more easily grasped. This is likely to be more nostalgic than realistic. The major advances of science have been those that consolidated knowledge by the per- ception of unifying principles. Before the discovery of quantum mechanics physicists had to learn the bewildering variety of atomic spectra and myriad empirical laws of limited validity. Today, atomic spectra are relegated to tables and the physicist need only know in principle how their frequencies can be deduced from the equations of quantum mechanics. To be sure other vistas have opened up, but it is far from obvious that today things are complicated whereas yesterday they were simple. It is more likely that postdoctoral education has arisen in some fields be- cause those fields are so rich in subtleties of technique and sophisticated ideas that the single research project required for the doctoral thesis does not pro- vide the student with a sufficient grasp of his field to permit him to become an independent faculty member. On the other hand, not everyone who earns a PhD in those fields intends to continue in research on the frontier. To re- quire that everyone spend another two years to acquire the mastery that is essential for further research contributions is both inefficient and redundant. The present system allows the college teacher and the nonacademic researcher to get about their business and permits the potential academic researcher to have the additional benefit of experiencing research in a new environment. If this means that the theoretical definition of the PhD degree must be changed, that might be the direction in which to move. Our fundamental conclusion, therefore, is that postdoctoral education is a useful and basically healthy development. Although our discussion to this point has been concerned with the postdoctoral experience immediately following the PhD, the conclusion is valid for postdoctoral study at more senior levels as well. We shall return to this area in more detail later. Having stated our favorable attitude toward postdoctoral education, we are also convinced that current practices can be improved and that changes in atti- tudes and policies are desirable. The merging of research and training is critical for postdoctoral education, but when the training aspect is ignored or neglected the experience may not be as useful for the postdoctoral and for his subsequent employer as it could be. The origin of the difficulties lies in the indirectness of the support of much of postdoctoral activity, both by the federal agencies and by the universities. The problem is exposed most clearly when one tries to answer the question: "Are there too many or too few postdoctorals?" Lacking a clear statement of why there need be postdoctorals in the first place, such a question is in princi- ple unanswerable. There are two extreme cases where the dilemma can be re- solved. They are typified by considering the postdoctoral first as a "means" and second as an "end." The more realistic case where he is both means and end is more complicated.

244 CONCLUSIONS AND RECOMMENDATIONS If the postdoctoral is solely a means, i.e., he exists and is supported simply to assist a principal investigator in performing research, the number of postdoc- torals will be related to the level of research activity. Once it has been decided how much research is desirable and affordable and with what urgency the re- search is to be done, the number of postdoctorals there "should be" can be determined. Perhaps we should not in this case refer to them as postdoctorals but as professional research-staff members who hold the doctorate. Whatever one decides about the postdoctorals, such professional researchers might be desirable. There are PhD's for whom a career as a junior associate to a principal investigator is not only attractive but possibly constitutes the best use of their talents. Support for such full-time researchers may or may not be in the coun- try's interest, but they should not be confused with postdoctorals who are de- fined as seeking an appointment "of a temporary nature . .. which is intended to offer an opportunity for continued education and experience in research." At the other extreme, if the postdoctoral is solely an end, i.e., he exists and is supported simply to prepare him for a particular kind of position (or possibly several kinds of positions), then the number of postdoctorals would sensibly be related to the number of appropriate positions expected to be available at the conclusion of his appointment. The nature of the research activities under such an appointment would be such as to provide the postdoctoral with the techniques, the vision, and the independence that are required for the success- ful filling of the anticipated position. Under these conditions it might not be possible to have the research program of the mentor proceed as smoothly or as efficiently as under the concept of the postdoctoral as a means. Efficiency, however, would not be the point; it would be education. In practice neither extreme predominates, although some postdoctorals supported by faculty research grants approximate the former and some of those supported by training grants the latter. What is desired and what occurs much of the time regardless of the support mechanism is a combination of the two. The possibility of a mutually satisfactory relationship between the mentor and the postdoctoral is often realized, but grants and contracts in sup- port of research at universities should be consciously given with the purpose of achieving simultaneously both the research objectives and the training of pre- and postdoctorals. The consciousness should extend not only to the fac- ulty and administration of the university, but also to the granting agency. There may be some loss of efficiency implied in such a policy, but it would serve the mission of the university without hurting the mission of the agency. In some cases congressional action would be necessary to free the agency from current restrictions on support of training or education. The only criteria that the program officers may legally apply to requests for support for research assistance at either level must relate to the "level of effort" or to the need to achieve the research goals expeditiously. The university and, more particularly,

245 CONCLUSIONS AND RECOMMENDATIONS the faculty member is forced to focus its justification on these issues, not em- phasizing the educational possibilities that the research might involve. Where such a practice might be appropriate for an independent or industrial research laboratory, it is a distortion of the full responsibilities of the faculty member. The fact that many program officers do in practice concern themselves with support of graduate education despite the restrictions in no way vitiates the desirability of removing the restrictions. Education on both sides of the PhD should be supported by design rather than by accident. The training-grant approach to postdoctoral education appears to have all of the benefits and none of the drawbacks of the research-grant mechanism. Here the training is emphasized, although, since it is training in research, it im- plies a setting in which the faculty is fully involved in research. The trainees often play the part of research assistants and the research effort of the mentor is augmented. There is as well a more subtle, but important aspect of the train- ing grant proposal that makes it attractive. The department or proposed training- grant faculty must justify the awarding of the grant in part because of a need for people trained in the manner proposed. Thus the faculty have an awareness of what is happening to the manpower picture in their discipline and of their responsibility to respond to it. There is, however, a potential weakness in the training-grant approach that the research-grant mechanism does not share. Of crucial importance to the postdoctoral experience is the adequacy of the faculty member as a mentor. Unless the mentor is a master scientist capable of contributing not only skills but also a critical spirit to the relationship, the postdoctoral period may pro- vide the apprentice with merely more research experience and not necessarily better experience. The training grant is generally awarded to an entire depart- ment or to a group of faculty. Although usually there are exceptional men in the group, few departments can boast of having only such men. In many de- partments there is overwhelming pressure to spread the largesse of money and trainees among the entire group, without the hard decisions that would re- serve the postdoctoral support only for those investigators with something special to give. There is an aristocracy of excellence in science that is ignored only at the risk of mediocrity. The research grant tends to be awarded on the basis of such excellence. Those who construct and monitor federal programs should give thought to ways of combining the best of both approaches. Before returning to the question of how many postdoctoral positions there should be, we must consider the third important mechanism of support, the postdoctoral fellowship. Fellowships differ from the other modes in concen- trating attention on the postdoctoral himself. The great strength of the fellow- ships is that they identify the potential leaders in research and instruction. Since the fellow carries his own stipend with him, he is much better able to select his mentor and the mentor is usually able to accept him as an appren-

246 CONCLUSIONS AND RECOMMENDATIONS tice. For these exceptional people the fellowship permits, in principle, the exceptional experience. Again, however, the real world modifies the abstract and admirable princi- ples. Although the award is usually based not only on the scientific potential of the applicant but also on the proposed research, the grants do not in general have nearly enough support for research expenses to allow the fellow actually to carry out the anticipated research. He is forced to depend on the resources of his mentor, usually derived from research grants the mentor has won, to ac- quire the equipment and supplies necessary. Since the fellow is a superior indi- vidual, the mentor is usually happy to provide the funds if the purpose falls within the purview of his grant. At times, however, whether because of the restrictions on the mentor's grant or because of the mentor's own lack of inter- est in the research proposed by the fellow, the latter finds it to his advantage to shift his project to align it more closely with the mentor's research. The free- dom of the fellow to pursue his own research is thus frustrated; nor is it clear that additional research support alone would rectify the situation. The mentor should be brought into the decision-making process, perhaps by being asked to endorse the proposed research at the time of the application for the fel- lowship. Involvement of the mentor (now seen as the proposed mentor) in the appli- cation and judging process would have other advantages. Although the fellow has only himself to blame for choosing an inappropriate mentor, the review by the panels of the adequacy of the mentor as well as the quality of the appli- cant might avoid unfortunate experiences. Moreover, the group of possible mentors might be expanded. Present restrictions in the federal programs im- posed by legislation permit fellowships to be held only at universities and at certain nonprofit and governmental institutions. If the desire is to match the fellow with the mentor, it is conceivable that the best mentor for the particu- lar applicant is at an industrial research laboratory. Evaluation of the mentor as well as the applicant would go far to eliminate any fear that the postdoc- toral might be exploited or that the program might be compromised. We are not prepared to answer the question of how many postdoctoral posi- tions there should be in quantitative terms, but we do have some suggestions about what should be taken into account in determining that number. The first suggestion relates to the fact that, in spite of the differences in approach, the individual postdoctoral and his mentor do not attach the significance to the special properties of the fellowship, the traineeship, and the research associate- ship that the sponsors of these programs often do. They are all seen as means to the same end, namely, the postdoctoral experience. We believe that this fact of life should be accepted, without suggesting that the differences among the programs are unimportant or that these different mechanisms of support should not continue. Their importance lies, however, outside of the postdoc-

247 CONCLUSIONS AND RECOMMENDATIONS toral-mentor relationship and nothing would seem to be gained by trying to intrude these values into that relationship. It follows that, as far as postdoc- toral education is concerned, the numbers of postdoctorals is measured by considering the sum of the numbers on fellowships, on traineeships, and on project assistantships. A second suggestion is that a distinction be made between the person hired on a research grant who is looking for a permanent position as a research asso- ciate and the bona fide postdoctoral, who is seeking a temporary educational experience. Such a distinction represents a polarization rather than a dichotomy and probably can be made only by the mentor. It depends not only on the qualifications and goals of the "postdoctoral," but also on the qualifications of the principal investigator, qua mentor, and on the nature of the research activities to be undertaken. Host institutions and faculty members must take it on themselves to evaluate each situation and to ensure that the postdoctoral is not treated simply as an employee. The number of fellowships should be limited so that a distinctive element of the fellowship will be the recognition of exceptional quality. This means that the number of fellowships will have to be set at some modest fraction of the number of PhD's produced. The pattern in the biological sciences, where approximately one third of the postdoctorals are in each of the categories of fellowship, traineeship, and project associateship, might well be duplicated in the physical sciences. If this were done the number of fellowships in physics and chemistry would have to be increased over the number currently available and a traineeship program would have to be initiated. In addition, the total number of postdoctoral opportunities of all kinds should have some relationship to the number of people with postdoctoral backgrounds required by universities, by specialized industries, and by govern- ment laboratories and to the number of doctorate-holders who would benefit by the experience. Such a determination would necessitate some planning of manpower requirements. We do not agree with those who argue that manpower planning is unnecessary, that the market place will determine the numbers needed, and that the society will accommodate whatever numbers of postdoc- torals are available. Society will, of course, adjust to the number of postdoc- torals. However, unless this number approximates the number of subsequent opportunities to utilize their special aptitudes and training, we will have one of two consequences. If there are too many postdoctorals, we will have wasted the funds required to train them; we will have raised their expectations without being able to satisfy them; and we will have created pressures in the institutions that hire them to permit them the opportunities they desire, whether there is a social need or not. If there are too few postdoctorals, the consequences are more subtle. Universities and other natural employers of postdoctorals will ob- viously adapt to the situation, but we can expect a drop in quality and in pro-

248 CONCLUSIONS AND RECOMMENDATIONS ductivity that will be hard to measure. Discoveries not made and excellence not realized are never missed, but we are the poorer for their absence. An effort should also be made to ensure that a steady flow of foreign post- doctorals to the United States is maintained. We leave to those charged with foreign-policy management the task of justifying the flow in terms of our respon- sibility to the development of other countries less well endowed. Even if that were not an issue, the visiting and studying in our laboratories by foreign scien- tists could be justified by their contribution to American research alone. Ameri- can science is and has been improved by the ideas and techniques these people have brought from their home countries. Our graduate students, and indeed our faculty, are better for the association. The foreign postdoctorals who return home often constitute for the mentor a network for the informal exchange of ideas and scientific news that stimulates research long after the postdoctoral experi- ence itself. On the other hand, some control on the numbers of foreign postdoctorals needs to be imposed, both for their benefit and for ours. The essentially Ameri- can atmosphere of our graduate schools should not be lost through an exces- sive concentration of foreign scientists. Foreign postdoctorals of marginal qual- ity should not be encouraged to make the investment in coming to this country when their talents might be better used at home and, in general, foreign post- doctorals should be urged to return home. However, we should not allow too great a concern for the relevance of the American postdoctoral experience to the needs of the home country to prevent an exceptional foreign scientist from participating in our programs. The next Einstein may come from Indonesia or Mali; we should welcome that possibility. It is important that American PhD's have opportunities to work and study abroad. If the best mentor for a particular young scientist happens to be in a foreign country, then both the postdoctoral and American science will gain from his taking his appointment overseas. Familiarity with the best work being done in other countries is critical if American scholarship is not to become isolated. Moreover, the presence of American scientists in foreign laboratories will often stimulate research there. The recent reduction in the number of Ful- bright fellows and the elimination for at least a year of the National Science Foundation Senior Postdoctoral Fellowship Program are severe and regrettable blows to the international character of American scholarship. With regard to the overall support of postdoctoral activity, there is the need for more opportunities for study at the senior level. This need extends not only over all fields from the humanities to the natural sciences, but it encom- passes those in industry and government as well as those in the universities. There is ample evidence that innovation and renewal take place best when individuals move into new environments and interact with new stimuli. The senior postdoctoral appointment, usually in association with a sabbatical leave

249 CONCLUSIONS AND RECOMMENDATIONS with or without pay, is highly desirable both for the research and study that it permits and for the perspectives that it awakens in people who may have grown somewhat stale in their positions. This again is an area where we may not miss the benefits but we are the poorer for the lack.3 Finally, with regard to the numbers of postdoctorals, care must be taken that decisions made by Congress or the federal agencies to satisfy one purpose do not carry with them undesirable secondary effects. The case in point is the current budget squeeze that has resulted in a cutback in funds for research. Although the postdoctoral was not a target in this decision and the reduction of his numbers was not intended even as an accompanying side effect, there is evidence that he is one of the most vulnerable components of research budgets. In Table 58 we give the results of a survey taken in the fall of 1968 to measure the impact of federal research cutbacks on the postdoctoral population in physics and chemistry.4 Although the reduction in numbers is not as severe as had been anticipated, it must be remembered that the demand for postdoctoral TABLE 58 A Comparison of the Physics and Chemistry Postdoctoral Population in 1967 and 1968 Type of Academic Physics Postdoctorals Percent Chemistry Postdoctorals Percent Institution 1967 1968 Change 1967 1968 Change Ten leading 260 212 -18.5 379 356 -6.1 Twenty other major 311 330 +5.9 557 319 -6.9 Established 233 221 -5.2 406 433 +6.7 Developing 143 155 +8.3 358 415 + 16.1 Total 947 918 -3.1 1,700 1,723 +1.4 Source: NRC, Office of Scientific Personnel, follow-up survey for the postdoctoral study. The need for greater appreciation of the senior postdoctoral appointment is reflected in the decision of the National Science Foundation to drop their senior program temporarily in favor of the regular program during the present federal restrictions on funds. The sen- ior program, with only 55 fellowships, represented 6 percent of all senior postdoctoral appointments, while the regular postdoctoral program with its 120 fellowships supports only 3 percent of the postdoctorals within five years of their PhD.s. The relative impact of the decision on the senior postdoctorals is twice what it would have been on the more junior postdoctorals. 4The numbers in this Table cannot be compared with earlier data as the returns are not complete. The relative changes from 1967 to 1968 are real, however, and are probably representative. An attempt was made to obtain figures for biochemistry, but an insuffi- cient number of responses made the data unreliable.

250 CONCLUSIONS AND RECOMMENDATIONS appointments has been increasing. If the number of positions had remained constant, the effect would be a 7 percent to 9 percent reduction in available positions. Furthermore, most of the respondents testified that in the fall of 1969 the figures will show a significant downward change. Postdoctoral posi- tions are being excised from budgets coming up for renewal. Apparently the investigators, the agencies, and the agency review panels did not give postdoc- toral education as high a priority as predoctoral education. Most of the preceding comments and recommendations are directed at the supporters of postdoctoral education and, in particular, the federal supporters. The universities have concomitant responsibilities with regard to postdoctoral education. The primary need is for the recognition of postdoctoral activity as an activity that is as central to the university purpose as undergraduate or grad- uate education, on the one hand, or faculty research and public service on the other. Distinguishing again between professional researchers, who are employed more or less permanently in departments and institutes, and the education- seeking postdoctoral, the university must assure itself that it has created the proper environment for the postdoctoral-mentor relationship to take place. Because of the somewhat delicate nature of that relationship and because of the effectiveness of the informal nature of postdoctoral work, there is probably little that could be done to improve the relationship by making it more formal or by trying to structure it from the outside. Nevertheless, we have a few sug- gestions that should reduce abuses and possibly increase effectiveness. Conceiving of the postdoctoral as an "end," regardless of the nature of his support, implies that the experiences provided for him will be such as to pre- pare him for the future. It is not self-evident that every research project or every faculty member will or can provide the proper setting. The number of qualified postdoctoral mentors is smaller than the number of all faculty qualified to di- rect graduate research. The university has the responsibility of identifying these people either internally or with advice from outsiders in the disciplines. In part, this is done by the review panels who recommend the grants, but not always with this particular focus. To provide the proper setting, attention should be paid to the physical as well as intellectual environment. Because the growth of the postdoctoral popu- lation on most campuses has been relatively slow and because it was seldom planned but simply occurred, few universities have adequate space, facilities, or equipment for postdoctorals. The postdoctoral activity has had to "piggyback" on the graduate and research program, acquiring whatever space the faculty member could sequester or squeeze out of existing space. Because postdoctoral education has not received an institutional commitment, only a license to exist, the rate of acquisition of equipment or, conversely, the limiting of numbers of students and faculty members in accordance with the availability of equipment has not generally been determined with the postdoctoral in mind.

251 CONCLUSIONS AND RECOMMENDATIONS The universities are not solely to blame for these conditions. The donors and controllers of construction funds have been either indifferent or actually hostile to postdoctoral education. We know of no state legislature that permits its state university to include the anticipated number of postdoctoral s along with the number of faculty and students when planning new academic build- ings. Similar problems exist at private universities with their boards of trustees. These problems are not likely to be resolved until these bodies are educated by the universities concerning the importance of postdoctoral education to the university committed to research. Before that can happen, there must be a prior consensus within the university. We hesitate to suggest imperatives for other details of the postdoctoral ex- perience, because the making of a scientist-professor is such an individual mat- ter. Each postdoctoral comes with his peculiar background of experiences and insights and the most effective program will be one that is tailor-made. There are, nevertheless, some aspects that should be considered. These include the opportunity to teach with supervision, the participation in administrative problem-solving, and the setting of limits on the duration of the postdoctoral appointment. The compulsion to teach and to create knowledge in others is a strong one and one that is especially acute for the new PhD. For more than twenty years he has been taught, and he often wishes to return the favor. Some have had the experience as teaching assistants while in graduate school, but some have not. Even though the prime purpose of the postdoctoral appointment is a research apprenticeship, the ability to communicate one's new knowledge is also important. We recommend that the postdoctoral be given the opportunity to do limited teaching at some time during his appointment. It would also be helpful if his teaching could be criticized. Once he becomes a professor, he is less likely to receive peer criticism of his teaching. One of the first tasks the postdoctoral will have when he becomes an assist- ant professor will be to write a proposal to some agency or foundation for sup- port of his research. If he is successful, he will then be charged with administer- ing the grant. He will be much better prepared for such responsibilities if he has participated in grant administration while a postdoctoral, at least to the extent of sitting in while budgets are constructed or while expenditures are being planned. The question of how long the postdoctoral period should last is also diffi- cult to specify uniformly for all postdoctorals. In some fields for some individ- uals, a year is sufficient time to make the transition from student to professor. For most fields and most postdoctorals, two years will permit the achievement of the educational objectives. Occasionally, for the rare individual, a longer period would be effective, including possibly a change of mentor and host insti- tution. Again the question must be decided in terms of the individual. What

252 CONCLUSIONS AND RECOMMENDATIONS is important is that the postdoctoral not be kept any longer than is necessary. The decision should be made on the basis of the needs of the postdoctoral for further training, not on the needs of the faculty for further assistance. Another question is that of the concentration of postdoctorals at relatively few institutions. If this concentration reflects the concentration of superior faculty researchers at the same institutions (and it probably does), it is not only appropriate, but any pressure to spread postdoctorals among all universities in the name of equity of geographic distribution should be strongly resisted. Egali- tarian democracy cannot be the model for postdoctoral education. Only the best PhD's should be encouraged to pursue it and only the best faculty should supervise it. One of the more unfortunate ways in which a postdoctoral may be used as a means is to entice him to a weak department as a means of up- grading the department. The postdoctoral should follow excellence, not be responsible for creating it. The pattern of changes between 1967 and 1968 shown in Table 58 is not encouraging in this regard. There are several issues regarding postdoctorals that we mention here in the hope that others will consider them either in future studies or in the routine collection of statistics. As we have seen, postdoctoral activity makes a significant difference in the lives of the participants, in the universities that host postdoctorals, and in the flow of highly talented manpower among the universities and research institu- tions of the country. The collection of information on which these findings were based was a difficult process, requiring the creation of primary instruments to draw the necessary data from the sources. Very little information regarding postdoctoral work was available from compilations of statistics concerning higher education or scientific manpower. It would be a desirable consequence of this study if those responsible for collecting such information on an annual basis would include questions about postdoctorals. Some groups, such as the Graduate Traineeship Program at the National Science Foundation, the Ameri- can Chemical Society, the American Medical Association, and the Committee on Academic Science and Education of the Federal Council on Science and Technology, have recently been collecting such information. Similar activity by the U. S. Office of Education would be helpful. Similarly, recent changes in the form used by the Survey of Earned Doc- torates of the National Research Council have made the data on the backgrounds of new postdoctorals much more useful. We hope that the National Register of Scientific and Technical Personnel can include explicit questions on postdoc- toral experiences in its surveys of individual scientists. We have discovered that, as far as postdoctoral education is concerned, the presentation of information in tabular form is equally as important as its col- lection. In the course of the study it has become evident that certain variables are particularly significant in distinguishing among universities and departments.

253 CONCLUSIONS AND RECOMMENDATIONS Major differences in hiring practices, funding, graduate enrollments, distribu- tion of work loads, proportions of foreigners, etc., are exposed when data are distributed across these particular variables. The first is the reputation of the institution. Although valid arguments can be made against grouping by repu- tation, the correlation among reputation, federal obligations for research, and doctoral production is strong. The important point is that the behavior of the institutions at the graduate and research levels is much more strongly dependent on these variables than on the more classic ones of private versus public, secu- lar versus church-related, or, within limits, large versus small.5 To lump all universities or technical institutions together is to miss the diversity of higher education that exists within these categories and to present data that are mis- leading. The second variable that has been important in presenting data is the pres- ence or absence of postdoctoral s within a department. It would be a mistake to attribute the observable differences between departments to the postdoc- torals, but apparently the environment that attracts postdoctoral s also pro- duces other distinctions in the graduate and research programs. It would be interesting to determine the various correlates with postdoctoral presence. Much more needs to be understood about the subsequent behavior of post- doctorals. Longitudinal studies, now possible with our data base, will tell us where former postdoctorals go for employment and what their achievements are. We should be able to learn how important the postdoctoral experience is in determining the course of a scientist's career. The migration of the foreign postdoctoral could be plotted and the relationship between the "brain drain" and the availability of postdoctoral appointments could be more thoroughly understood. Beyond the longitudinal study, data should be collected periodically from postdoctorals to establish new data bases. One can expect some changes to occur in the postdoctoral picture as the means and extent of support change. More detailed information will be needed on the participants than simply a head count by discipline. It might be useful to establish a continuous record of postdoctorals similar to that made by the National Research Council's Sur- vey of Earned Doctorates. There is little doubt that the postdoctoral is here to stay. In fact, the cur- rent cutbacks in federal funds have awakened many to a realization of his im- portance in the academic world. If the academic community and the federal agencies respond to this awareness with coordinated programs of training and support, it will no longer be appropriate to refer to postdoctoral education as the "invisible university." 5 A welcome contribution to this suggested manner of presentation is the National Science Board.s 1969 publication, Graduate Education: Parameters for Public Policy.

254 CONCLUSIONS AND RECOMMENDATIONS Summary Postdoctoral education serves a variety of purposes, differing somewhat from one discipline to another. Nevertheless, certain common themes remain as long as we restrict ourselves to the sciences, where most of the activity takes place. Only among the senior postdoctoral s do the humanities play a comparable part in postdoctoral education. Throughout this chapter a number of conclusions and recommendations have been made. We summarize them here for the convenience of the reader. (Unless otherwise specified, the word "postdoctoral" refers to the immediate postdoctoral in the sciences.) Postdoctoral education is a useful and basically healthy development, both immediately following the doctorate and later for more senior investigators. Its major purpose at the earlier stage is to accelerate the development of an independent investigator capable of training others in research. At the later stage it serves as a means for concentrated pursuit of research and scholarship goals and of renewal for those whose regular responsibilities do not permit them to pursue these goals. All those connected with postdoctoral education are urged to conceive of the postdoctoral appointee as one who is in the process of development and not primarily as the means to accomplish other ends. For the agencies and foundations, this means recognition that the educational goals of the univer-. sity may be served explicitly through research support. For the university, this means that the postdoctoral is an important component of the educational scene. For the faculty member, this means that the postdoctoral should be given every opportunity and encouragement to develop his potential as an independent investigator. • Most, but not all, postdoctorals participate in teaching and many desire more opportunities to teach. Some postdoctorals are involved in research ad- ministration. Almost all postdoctorals spend no more than two years on the appointment; some appointments are as short as one year; and a few postdoc- torals find more than two years to be of benefit. Because of the individual nature of personal development, we believe that the participation of the post- doctoral in administration and teaching and the duration of the appointment should be determined in each individual case. The criterion should be whether the experience will enhance the postdoctoral's progress toward independence and excellence in research and graduate education. « Of critical importance to the training of a postdoctoral is the ability of his mentor to provide the proper leadership and environment. In some fields the

255 SUMMARY best possible mentor for a given postdoctoral may not be in a university or a national laboratory. Current restrictions should be removed to allow postdoc- toral fellows to choose mentors at industrial research laboratories. Few universities, whether public or private, have adequate space, facilities, or equipment for postdoctorals. Both boards of trustees and funding agencies, including state legislatures and budget offices, should be apprised of the im- portance of postdoctoral education in the university in which research is a significant part of the educational program. The allotment of existing space and the planning for new facilities should include explicit recognition of the anticipated postdoctoral population at both the immediate and senior levels. Postdoctoral fellowships should carry with them sufficient support for research expenses, so that the fellow need not depend on his mentor's sources of support to carry out his proposed research. The number of postdoctoral opportunities available at any time should be related to the number of Ph.D.'s and professional doctorate holders who can profit from the experience. The mix between fellowships, traineeships, and project associateships in the physical sciences might mirror that in the biologi- cal sciences, where approximately one third of the postdoctorals are in each category. A distinction should be made between the postdoctoral and the employee with a doctorate who is looking for a career as a research associate. Support for senior and intermediate postdoctoral opportunities should be increased in all fields. In the humanities and social sciences, the senior and the intermediate postdoctoral appointments are and probably will remain the dominant modes of postdoctoral activity. In the sciences, the faculty should be encouraged to take leaves for stimulation of their research interests and renewal of their perspectives. In addition, postdoctoral activity at these levels may have the greatest subsequent impact on the quality of teaching. Within the bounds of maintaining the essentially American character of our institutions, the foreign postdoctoral is a most welcome visitor. In addition to the contribution to international education, the presence of foreign postdoc- torals has enriched our science and has stressed the international nature of research. This exchange of persons can be stimulated by cooperating in pro- grams that are designed to encourage the foreign postdoctorals to return to their homelands. Travel of American postdoctorals abroad should be encouraged and the num- ber of opportunities increased. No\ only do our people learn what is happening in other countries, but they help to further research in those countries. The

256 CONCLUSIONS AND RECOMMENDATIONS recent severe limitation in Fulbright Fellowship opportunities is particularly unfortunate in this regard. * Postdoctoral fellows tend to go to those institutions where the scientific leaders are located. Postdoctoral project associates and trainees are likewise attracted to excellence in science, since the research and training grants are generally made with a view to the scientific capability of the principal investi- gator or the training faculty. As institutions that do not now host postdoc- torals are developed to excellence by the attraction of leadership-quality faculty, postdoctoral s will follow. Postdoctoral s should not be the means to the development of an institution, but the measure of its excellence.

APPENDIXES

• A APPENDIX A-1 Postdoctoral Census Questionnaire In an effort to make a census of all postdoctoral s in- the United States and all postdoctoral s abroad who were U.S. citizens, a questionnaire was designed to elicit information on the background of the postdoctoral, the nature of his appointment, and his subsequent plans. Since the identity of these people was unknown, it was necessary that the host institutions distribute the question- naire. A list of such institutions was compiled. These included all universities belonging to the Council of Graduate Schools in the United States (243), nonprofit institutions and government laboratories (164), independent hos- pitals receiving more than $25,000 in research funds from the National Insti- tutes of Health (43), member libraries of the Association of Research Libraries (73), other institutions receiving HEW Graduate Training Grants (182), and selected industrial laboratories (28). The president or director of each of these institutions was asked to designate from his staff a coordinator with whom we might correspond. Each of these coordinators was asked to distribute the questionnaires to the postdoctorals at his institution, to collect the completed forms from them and to return the forms to the Study office. Questionnaires were also sent directly to all holders of nationally awarded fellowships (both federally and privately financed) who were not at the above institutions. This census took place in the spring of 1967 and we received 10,740 com- pleted forms that were sufficiently complete and not excluded by our defmi- 259

260 APPENDIX A: THE QUESTIONNAIRES tion of a postdoctoral appointment. The question immediately arises: How many did we miss? To estimate this we have used counts from other sources. In the application form for its Graduate Traineeship Program, the National Science Foundation asks chairmen to indicate the number of postdoctorals in their departments. Robert H. Linnell has analyzed these applications1 and found a total of 6,352 postdoctorals in all sciences in the fall of 1966. Because the National Institutes of Health provides more funds in the health and life sciences, it is likely that many departments in these areas did not apply for training grants from NSF. Departments in the physical sciences and engineer- ing, however, must rely almost exclusively on the NSF for locally administered funds to support graduate education. Linnell feels that almost all eligible de- partments in these fields in the country made a traineeship application and thus the figure in these areas for the postdoctoral population is accurate.2 He found 3,967 postdoctorals in the EMP (engineering, mathematical, and physi- cal) sciences. It would have been preferable ta make comparisons by individual depart- ments to allow for differences among the return rates by discipline, but this was not possible. Our data distinguishes among fields of research; Linnell's among departments. Many postdoctorals in chemistry departments indicate that their field is molecular physics; they are included in our data as physi- cists. It is reasonable to assume, however, that people in EMP fields are in EMP departments. We had returns from 3,165 postdoctorals in the EMP fields at universities, which represents an 80 percent return rate. From the clinical fields at medical schools we received 2,207 returns, whereas the American Medical Association reported3 4,186 postdoctorals in these areas. In this much more diffuse area of postdoctoral activity where the definition is stretched to the extreme, our rate of return is 53 percent. If we take, as an average, a return rate of 65 percent for the basic medical sciences, assume that the fields generally associated with the arts and sciences at universities share the 80 percent return rate of the EMP fields, and assume that the return rate from postdoctorals outside of universities is the same as from those in universities, the total postdoctoral population comprised approxi- mately 16,000 persons in the spring of 1967. 1 National Science Foundation, Graduate Manpower Resources and Education in the Sciences, August 1967. 2 Assuming, of course, that chairmen always report accurate figures. 3Joumalof the American Medical Association, Vol. 202, No. 8, Nov. 20, 1967, p. 818.

261 POSTDOCTORAL CENSUS QUESTIONNAIRE THE STUDY OF POSTDOCTORAL EDUCATION Spomorcd by the NATIONAL RESEARCH COUNCIL NATIONAL ACADEMY op SCIENCES — NATIONAL ACADEMY op ENGINEERING If you have • poetdoctoral position (MM definition on the ettached shMt) pleeae fill out thil queetionneire end return it to your ueuejuueut ofiee (or other daeicneted office) u eoon u pomible. The iniarmetion provided on thif form will be held In rofiMenr. by the Neuoul Reeaerch Council, end naed lor ateurUcal purpceee only. Ye«rol L Noe 2. Birth i••«•i ILeM NBBMt trtnt Nw«) <MM*U N.awl ItT.M, 4. Kerltal Menu 12 O Herried 11 a Not merried lincludinf widowed, divorced) 5. Number of dependenle. Uae VS. income tax definition, but do not include youneU ( ) (Ml 1 Sodel Security Number (U.S) " * 7. Of whet country ere you • ciusen? I. Pleeee 111 out the tnlonneuon requeued below regerdinf your underereduete end greduete educetioo. Uae field r from the eUeched vecialtiee liu. » Pnrtou. portdoctonl ImUtution and Location Period of Appointment or IntMitut e PtM.W All in the information requested below regerding your present postdQrtor-l appaintment II. HUM and location of the institution 11. Department, Center, or Institute ~ _.. „ „ 12. Name of the profeMeor or other staff member with whom you are working "•-Ml ItJU.t MUM) iriMt M«J—) iM.MI. I«Hi,i) 13. Tttle of appointment If this is a part-tine appointment, what other position do you hold? 14 Deatgnation of your research area, using the name and number from the accampanyi&C trr-Hi«^ 1U1: U. When did your |»etdoctoral training begin? It When do you expect to complato your i.i-.-j {.•»...> portdoctoral trmiaing? It 1C What agmey has provided the funds for your preaent salary? <««-«> (If not sura, please aak your research sponsor) IT. Which of the following general types of appointment do you hold? <«•> Feliowahip Tn.niMli.ii g-M^+u-i PoaUion supported from project funda Olhar (apadfy) e tam over <A* page for tfW reet o

262 APPENDIX A: THE QUESTIONNAIRES a. Monthly aelary or •Kprnd on ihu natdn.laral appointment (opuonal) I ftAH*—.] monthly aalary tor taarhlnl actlviuee (optional) I I4e-ei II. Ant you now on lea»e Iron mother poaition? .«. Yes No If yea. Indicate the poutlon from »hkh you Of* en laava: i•>•••> ° ' •L An you • unaptly nceivine. a^ry support Irom your homa InMltution? .•• Yjs No 13. What u« )rmir main ra«*on« for taUng • pMtdoctonl appoiBtaMatT iaa-«i Ptan imnr tha followinf »itli raapM to tha natun of your potdocunl activltia. B. An you • cajidkUt e for mother cbetonl dagree? "i Yjs Ho Will you ulu or audit any raiular couraai durlac thla apmintmrnf •"' Yj. No » Da you partidpau In th» taaching of undargraduala or graduata atudaata? -« «.i UnbnraduMa r.raduata Yaa No Yaa Mo Couna lacturaa <••i Quiz a Non-eradlt oounai «•* M Do you wuh thai appolntmant pnddad mora opportunity tor taaching? Yaa No B. Hava you raaponaiblllty lor tha Inprovarnam of naaareh aqulpmanl' Yaa No <OTI 6 . m. with how many atari mambara do you hava awndf ant profaaalonal conuct. IT. Do you uoa tha library mora or laaa than you did aa a fraduata atudant? " Somewhat men About tna aama Laaa t your caraar axpaetattona attar mmplating your praaant poatdoctoral appolnlmanL at probably be employed after your preaent peatdoetoral Univenlty rolleja raderal Qovemment StaIe or local govenunent f •hlalnaaa or Induatry Non-proflt organlaauon Self-employed Other (apeotfr) » What la your moat probable location? <'••.» (atate or rountry) M. To whet aztant haa your poetdoctoral emperlence chanced your career eaplraHona? <"••<

263 DEPARTMENTAL QUESTIONNAIRE A-2 Departmental Questionnaire In order to determine the nature of the environment in which most postdoc- torals find themselves and where they are likely to be employed after their appointment, a questionnaire was designed to be answered by departmental chairmen at colleges and universities. Questionnaires were sent to the coordi- nators at all universities belonging to the Council of Graduate Schools (CGS) in the United States and to the presidents at a sample of all remaining col- leges and universities. This sample comprised all schools at which 50 percent or more of the faculty hold the doctorate and a 10 percent random sample of all other schools. The coordinators and presidents were asked to distribute these departmental questionnaires to those departments that deal with the fields listed and to return the completed questionnaires to the Study office. The distribution of returns is as follows: Other Colleges and Universities Field Humanities Social sciences Physical sciences Engineering Biological sciences Basic medical sciences Medical specialties Education Combined departments Total Departments More than Universities Half PhD in the CGS Faculty 425 592 658 307 354 238 150 201 199 17 66 12 645 Less than Half PhD Faculty 91 105 98 5 36 1 335 Other Institutions 14 209 244 19 486 Total 666 898 955 329 455 265 209 244 19 4,040 2,574 Note: Number of schools responding, 357; number of schools approached, 422.

264 APPENDIX A: THE QUESTIONNAIRES FIELDS OF STUDY COVERED BY THE DEPARTMENTAL QUESTIONNAIRE The questionnaire should be completed at each institution by that department chiefly responsible for each of the following fields of study. At many institutions one department may be responsible for several of the fields listed here, e.g., a department of applied science may include all the engineering fields or a depart- ment of social studies may include the fields of economics, political science and sociology. In these cases, even though the department is broader than the fields listed, the response should be for the entire department. On the other hand, sev- eral fields may not be represented in any department at a given institution. In such cases, of course, the fields should be ignored. If the field is represented by two departments, e.g., in the graduate school and in the medical school, please have both respond. If the departments are small and their circumstances similar, the form may be distributed to division, rather than department, chairmen. In extreme cases at small institutions, a single form may serve for the entire institution. 1. Agronomy 15. History 2. Animal husbandry 16. Internal medicine 3. Biochemistry 17. Mathematics 4. Botany 18. Mechanical engineering 5. Chemical engineering 19. Microbiology-bacteriology 6. Chemistry 20. Physics 7. Dentistry 21. Physiology 8. Economics 22. Political science-government 9. Education 23. Preventive medicine-public health 10. Electrical engineering 24. Psychiatry 11. English 25. Psychology 12. French 26. Sociology 13. Genetics 27. Surgery 14. Geology 28. Zoology

265 DEPARTMENTAL QUESTIONNAIRE DfcPARTMbNTAL QULSTIONNAtRb. POSTDOCTORAL I Department Institution ,. Telephonc Are« Code Number t.xlcnMon 1 \umher a} faculty memhrrt in ih, deportment« of the fall term. tVM I ull-tin a Professors (present and on leavel b. Associate professors (present and on leavel c. Assistant professors (present and on leave) d Lecturer> c. lustra tur• (other lhan graduate students holding this appointment) f Visiting professors, associate professors, assistant professors, and lecturers filling regular staff positions in the department g. Other ispecify) Total in above categories OV through "g" 3 Previous background and present }um tiont of recent appointeet to the full-timr /unior facultv a. The lait .'in members appointed to the full.time luntoe facuity iassistant professor, instructor, or equrvalent) came fro i!v lollnuinjr backgrounds (enter number in each appropriate category): iacuity appointments at other institutions Postdoctoral appointments at other institutions Had just completed work for a doctor's degree (PhD. M D . etc.) elsewhere Had IUM completed work for a doctor's degree at your institution Were engaged in graduate study elsewhere, without yet completing a doctorate Were engaged in graduate study ji your institution, without completing a doctorate .. Other (specify) Total - Ho« many of these five had completed work for a doctorate at the time of appointment? What tit the normal responsibilities of a newly appointed instructor or tuistani professor, measured in terim of the time Reseuch. including trarning student! in research Instruction, including lectures, seminars. tutorials, etc d. How many full-time faculty portions in the deputment. of the rank of aiuttant profeuor or mstructor were unfilled at the beginning of the fall term. 1966"* 4. Bockgrouniti amilunfttoni of pretent full-time facutivand stafl t Of the professors, associate professors, and aswtani pro feuor scoun ted in question 2 ai members of your full-time faculty in the fall of 1%6. ho<» many ace curiently on leave from the department for study or research'. have at any stage in their careers had a year or mote of postdoctoral study I supported either by others or by your institution)'. are actively engaged in rewarch? - _ are engaged in research supported in whole or in part by outside grants or contracts? _ b. txctuJitif Ihe faculty appointees counted in question 2, postdoctoral' as defined for the Study and counted below, and technician ho« many provisional rcsearch staff members were there in the department in the fall term. 1966? _ L How many of these professional rescarchers have a PhD. M.D., ot other doctor's degree? _ d. Pkax state lot estimatei the loul of research funds in your department from outwde grant> or contracts in the fiscal year 1966-67 I 01 the sake of uniformity, include overhead payments m the total. Do not include fellowship support or trarning grants 5. If you do not no* have postdoctoral students in the department, do you believe the department would benefit from the presence of such students? Greatly To some eKient Not significantly

266 APPENDIX A: THE QUESTIONNAIRES I OR DLPARTMLNTS WITH GRADUATL DLGRLL PROGRAMS 6 Pteax entei the luul number of f/*Juatc ttudentv full and pa/1-time. a1of the ii'i term. I9M A ••full-time" graduate Mudeni 11 defined here »a graduate undent who 11 engaged entirely in trarning actiniKs in hi* diicipline. thcie «* tivitiei may embran uiy appropriate combination of iiudy, teaching, and rciearch. Pleaie itate the numbei of full.tunr graduate uudenti now holding Teaching appointment! (e | . teaching aimunti, part.time uitlnictor>) _ Reiearcn appointmentt (e.g. re w arch tut tun It) weie awarded in the yew* I9A4-M ,nd I'M >uly I ihiough June 30, to itudenti maturing ir your depuimenC1 PhD'i at the equrvalent How many of yout PhD'i in 1964-45 and I96S-M entered each of the following oveupiiwniat lhei> I964-6S fradiutton> PhD'i AcMkmK •ppouitment M • cotIegc ot Hmvcnity Poitdocloral tiudy Rewvch in induitry Reievch in (oveinment or non-profit orgMtzaimu Mtlituy wrricc F.oreign country, my type of employment Othei lipecify) Tora! FOR DLPARTMLNTS WITH POSTDOCTORAL STUDLNTS 9a. Pteate enter the number of poii doc lord i in the deputmeni M of dw M HTM. I9M. U.S. F.ellowihip holders Truneei Appointees tupported on reicuch fundi „ Vititort wpporled by theu home midlutioni OtherKipecify) . Toial pott doc toral t b How many of your poitdoctoTali have the MD degrac or equrvklent? How many are in residency trarnmg? 10 How many of your profeuon, auocute profenon. and aiitUMt profeuoncounted m question 2 M MCMben of your faculty, are the mentori or iponiors of poitdoctorah? 1 1 Review of •ppomtmenti a ¥ftio reviewi ctitictlly iother than the individual The department head ;s5i.=bs r;'L±z^u«z..v,:;' * "«-•»".- «•—"- ocnt? (Qteck at many ai apply) A dean 01 vice-preudent An interdepartmental committee Other iipecify) No one b. It there any procedure for evaluatmi the progrcu and achwvcincnt of a postdoctoral stMlent after hi> term of appointment'. ................ Yei 1 1. How t lowly doei the monthly lalary paid to 1 poitdoctoral appointee by your departmeni approach the ialvy pard to faculty member! of the tame profeiuonal experience? Ptaue mdKate by a percentagc. tg. 1 10%. I«M. 85%. etc 1 3. How long may an individual continue in your department M a postdoctoral appointee? II ili-rr it a limit, is thit mititutional policy? May 19*7 Department H>ad Plrasr warn to thr rollr/tr oe urmentty coordinator nametl on thr firil p 14 .AH. there any limitations, aiide from thoic of fundi and floe* ipace. on the number of poitdociotali your department may appoint or admit in a given year ie.g. lo marntain a balance in the department, etc.)? ................ Ye> _ No 15 Do you have • departmental policy of involving poitdoctorali in leaching? ................ _ ......... V«i _ No.

267 FACULTY QUESTIONNAIRE A-3 Faculty Questionnaire In order to determine the relationship of the postdoctoral to the research group a questionnaire was designed to be answered by faculty members. Two groups of faculty members were selected to be in the sample: all those in the fields listed below who were mentors of the postdoctorals responding to the census questionnaire, and the faculty advisers of all students who received the PhD in 1966 in those same fields.4 The return rates by field are shown in the following table: Postdoctoral Mentors PhD Mentors Only Field Sent Returned Sent Returned Physics 654 430 488 127 Chemistry 785 625 614 217 Earth sciences 102 86 266 93 Social sciences 91 39 138 42 Internal medicine 561 250 - 7 Biochemistry 538 379 243 55 Biosciences 527 386 644 223 Total 3,258 2,195 2,393 564 The discrepancies and the different return rates are explainable in part by the fact that some faculty turned out to be mentors of postdoctorals who had not responded to our census. Thus those we thought were PhD mentors only were discovered to belong in the other group. In the social sciences a 10 percent random sample of the PhD advisers was taken to make the number comparable to the number of postdoctoral mentors.

268 APPENDIX A: THE QUESTIONNAIRES TIE STUDY OF POSTDOCTORAL EDUCATION - F«UI« Sponsored by the NATIONAL RESEARCH COUNCIL NATIONAL ACADEMY OF SCIENCES — NATIONAL ACADEMY OF ENGINEERING Please correct addms if changet I Academic Title. 2 Principal area in which you «ic currently conducting research (UM field number from the attached speclalties list) 3 •. How many students and ataff member* were punuing research with you or. if you are • member of • larger research group. In your group, as of April 1987? U.S. Foreign Citiiena Citisens Graduate degree candidate« (eg, for S.H., Ph.D., etc.) nipport«d on rexarch fundc of your own or of your research group .. ,. . Graduate degree candidatM iupportod from other Poftioctorab dncl*d\ng HD't pvrnitn0 reaeerch uMoW j^m, either by their tno>pe»aVitt plaaahtg or « part of their Tendency tratninoj fellow« Poatdoctoral trainee« («• g.. poatdoctorala appointed on an N.l.H. training grant) Portdoctoral appointees napported on your own or the group'a research fund« Postdoctoral visiton nipported by their home institutiona or, if foreign, by their home Other poatdoctorala (•peci/y) l Faculty eo-worken Profaarional research ataff including techniclans with a bachelor'^ degree but excluding Individuals aeparately counted above Other co~worfcen («p«:ify) b. Of your poatdoctorala counted above, how many: Have the Pfc-D. decree or equivalent? Have the ILD. degree or equivalent« . Of iheae M D '«, how many are doing research u part of their residency training? 4 One of our concenu u the relationship between reeearch training and rewarch iupporL Approximately what u your total reaearch budget this fiscal year, 1M8-CT? For the aake of uniformity, include overhead payments in the total. $ 5 How many atudento completed master'a theees or doctoral dissertations under your direction in the calendar year 1MI? How many are likely to finish in the calendar year 1M7? Estimated 1JM 1M7

269 FACULTY QUESTIONNAIRE 1. Purpote and Cftarortrr of Postdoctoral Study a. How strongly do you encourage your better graduate degree candidates, or your M D.'s in residency training if you are in a clinical fidd, to take an extra year or two of postdoctoral study . . . if they seek an academic career Strongly Fairly strongly Not strongly . - —, d they do not? Strongly Fairly strongly Not strongly b. If you encourage your better graduate degree candidates or residents to take an extra year or two of postdoctoral atudy, please check the three rroaonj which you feel are most compelling among thc*e listed below: Crad. degree candidates Residents To work with a particular scholar or scientist To acquire additional research techniques .... .. To gain further research experience To carry out a piece of research on their own To continue with research already started To sharpen the focus of their research To giv To put them at the growing edge of current research To support themselves in the academic world until a suitable faculty appoincment becomes available To gain some teaching experience To give them a breathing spell after their formal training To give them further time to mature To see the work being done at other centen To broaden their undentanding of the field To give them a chance to publish something 0*s» (specify) c. Should the character of predoctoral or of residency training be changed in the light of these reasons for promoting poit-Ph D and post-residency study? Yei PLEASE ANSWER THE FOLLOWING QUESTIONS tF YOU HAVE POSTDOCTORALS IN YOUR GROUP For the purpose of these questions, restrict the definition of postdoctoral to those within 5 yean of the doctorate. a. How do your postdoctorals contribute to the effectiveness of your research and teaching compared with the other students and staff working with you? Please put numben in the spaces below to indicate the character and scale of the contribution made by esch. 4 = very larg«; 3 = large; 2, small; 1 = very small; 0 = no contribution. . Graduate Post- A. fteaearcfc degree candidates doctorala Carry out complete sections of work Contribute stimulating new ideas Keep us in touch with research at other institutions Contribute necessary ancillary skills Open up new ares* of research Perform naciuary research routines to the quality and sophistication of the work a to the tempo of the work B.Tv+eking Help conduct laboratory counes

270 APPENDIX A: THE QUESTIONNAIRES I Piibhrahou b« poeXoctorell . How many book« and papen were publuhrf to IMS and IWt listing your postdoctorala Poetaoctorale with tht PS D. or equivalent ILD.'l independently punulng naearch training HO.'s In naldeney training b. What wa. the totel numb* of boob and pepen publlahed m 1HH and 1H7 by pereona under your direction? - 10. Your foreign peridoctoraUr > Pleaae 1*1 the home countria of your foreign poetdoctorala. with the number of pou- doctorale lran each. No. of Home country individuals 1 - 4. -- 5. - b Pln» put rfitck. In the approprUt e ap>m briow to cluneterlo your fonrign poM- doctorab: QiMJttV of prnioiu rMMrvh iremtne. Comparad with your US postdoctoral«, they h«ve Better theoretical training _ Better laboratory training Equally good theoretical training - Equally good laboratory training Leal good theoretical training _ Leat good laboratory training Relevance of thrlr portdortore! work to their mm country'« neeaj. Hieir poatdoctoral work hai: Much relevance Some relevance Little relevance— fntereet in retuming hom«: Moat plan to return home Many plan to return home Few plan to return • i • 11. Your US. pondoctoreb a. How many of your present U.S. poatdoctorala (excluding M.D.'a in rasldency training) have the qualifications you look for in an auistant profesaor, how many are •omewhat leal qualified, how many are distinctly leu qualified? Total number of US Number with Number who are Number who a poatdoctoral a (other qualifications aomewhat leaa distinctly leaf than M D 'i in resl- of faculty qualified qualified dency training) appointees b. Hal the quality of awerdees in the OS national fellowahlp programa (eg. N.SF.. N 1 H ) been of a auffldently high calibre? 12 How long ahould a poatdoctoral (other than an U.D. in raaidency training or an M.D. aaeking a Ph.D.) remain in your department . . . for hi« aake? . . . minimum yean maximum yean for the department's aake? minimum yean maximum yean IS. On the unit beala that 10 = the average time required to direct the reaearch training of a PkD. candidate, or of a resident if you are in a clinical fleld. what la your erumate of the average time required to direct the work of a poatdoctoral? 14. On the unit baais that 10 = the average office and/or laboratory spaee you aaalgn to a Ph.D. candidate, or to a reiident if you are in a clinical field, what u your aatimate of the average apaee nmipiaii by a poatdoctoral punuing further training?

271 FACULTY QUESTIONNAIRE 15. tf you have any opinion« about postdoctoral education which are not exprewed above, pie*.« comment below: It We would like to follow up on a nmple of the foreign postdoctoral* holding appointments in 1961-62, to obtain their opinion of what they accomplished as postdoctorals and to trace their iubaequent careen. Would you kindly list the naiMS, and If known, the preaent adrirewes. of as many as poiuible of your foreign pcMtdoctoral a in 1961••62? Name (PLEASE PRINT) Country of which Present addrett, if known they were a citizen Plaaw return to The Study of Portdoctoral Education, National Rasearefa Council. 2101 Constitution L. D.C. mu

272 APPENDIX A: THE QUESTIONNAIRES A-4 Postdoctoral Experience Questionnaire To determine the value of the postdoctoral experience to the individual and to compare careers of postdoctoral s with those who hold the doctorate but have not been postdoctorals, a questionnaire was designed to obtain such in- formation from a sample of PhD recipients some years after their degree. In order to avoid the possible bias that those who take postdoctoral appointments are already preselected, we attempted to make two samples of doctorate holders of apparent equal quality. For this purpose we took advantage of an existing study of the career patterns of doctorate holders5 by Lindsey Harmon. This study has followed up the careers of some 10,000 PhD holders who re- ceived their doctorates in five-year intervals between 1935 and 1960. Of this group approximately 1,600 had had a postdoctoral appointment. By restricting our sample to those who had received their degrees in 1950, 1955, and 1960, we were left with 779 former postdoctorals. This group was matched with an equally large group of non-former postdoctorals that was similar with regard to field distribution, "quality" of PhD institution,6 the time lapse between the baccalaureate and the doctor's degree, and age. These two groups were sent questionnaires and the return rate is given below. Some data on the nonre- spondents were collected from NSF's National Register of Scientific and Technical Personnel. Former Non-former Postdoctorals Postdoctorals Questionnaires sent to: 1950PhD.s 199 199 1955PhD.s 271 271 1960 PhD.s 309 309 Total 779 779 Questionnaires returned by: 1950 PhD.s 135(67.8%) 141(70.9%) 1955 PhD.s 175(64.6%) 186(68.6%) 1960 PhD.s 189(60.8%) 169(54.7%) Total 498 (63.9%) 496 (63.3%) It was subsequently discovered that the definition of a postdoctoral appoint- ment in Harmon's study differed from ours. This caused some switches between the two groups. Some, who had postdoctoral experience according to their response to Harmon's study, answered our questionnaire in the negative and 5NAS-NRC, Profiles ofPhD.s in the Sciences, Publ. 1293, Washington, D. C., 1965. 6 Allan Cartter, An Assessment of Quality in Graduate Education, American Council of Education, 1966.

273 POSTDOCTORAL EXPERIENCE QUESTIONNAIRE vice versa. Furthermore, we discovered that it was important to distinguish between those who had had an immediate postdoctoral experience and those who postdoctoral appointment was delayed. When we examine the returns and separate the respondents according to their replies we get the following distribution: Respondents Who Had Immediate Delayed No Postdoctoral Postdoctoral Postdoctoral 1950PhD.s 44 82 146 1955PhD.s 94 87 173 1960PhD.s 139 46 162 Total 277 215 481 These totals do not add to the numbers given in the previous table, since 19 respondents did not give sufficient information to allow themselves to be classified.

274 APPENDIX A: THE QUESTIONNAIRES THE STUDY OF POSTDOCTORAL EDUCATION Sponsored by the NATIONAL RESEARCH COUNCIL NATIONAL ACADEMY OP SCIENCES — NATIONAL ACADEMY OF ENGINEERING POSTDOCTORAL TRAINING AND EXPERIENCE 1 In your present position, how is your time divided between rsawch (including mining students in research), instruction, administration, uid clinical service? Research, Including training aiudcnu in research % Instruction, including lectures, seminan, tutorials, etc. . % Administration % Clinical service % ctmVif % TetaI 101 * 4. If you are currently engaged in research, what is your fold of specislisation now? Plaaae use field numbn- from the attached ^Mclalittw lutt Not currently i 9. If you have been ninicaffiil in obtaining mand* reeearch eupport u a principal investigator or co-investigator, when did you receive your £rpt grant or contract, and from what foundation or agency? Year II Granting Agency— Have you received subsequent grant (•) Yat «. How many books and papers have ,»» publiahed? 7. During your career, have you been • thesis adviser for any graduate students? If "Yes," how many at master's level at PhD level 8 What was your annual income in 1986 (or the most recent completed ft*"al year) from activities related : training? (optional) ... . . I

275 POSTDOCTORAL EXPERIENCE QUESTIONNAIRE t. Have you ever sought or •pplied for • postdoctoral appointment (Me attached definition)? • 11 v««, what reasons did you feel were the moat compaUin*' (For your Ant appointment, if more than one) Check up to ) To work with a particular scholar „ 01 To acquire additional rasearch techniques 02 To gain further meardi experience n To carry out a piece of research on your own 04 To develop further the reaearch you did during your pradoctoral or residency training 05 To Bharpen the foeua of your research 01 To give you a free period for research before you got aaddled with other responsibilities 07 To put youraelf at the growing edge of current research . 01 To support youraelf in the academic world until a niitable faculty appointment became To give you some teaching experience . .. .. 10 To give youraelf a breathing •pell after your formal training 11 To give you further time to mature 12 To aae the work being done at other centen . . . . . __.-... - - IS To broaden your undemanding of the field 14 To give you a chance to publlah aomething . . 15 Other upftify). if b. If no, waa a postdoctoral appointment unavailable in your case? Or did you feel that you would derive no benefit from it? Or were other opportunities at the time more attractive? 10. If you never had a postdoctoral appointment, do you now feel the lack of the experience it would have given you? Y« Please add any comments under question IS PLEASE ANSWER THE FOLLOWING QUESTIONS tF YOU HAVE HELD AN APPOINTMENT AS A POST- DOCTORAL AS DESCRIBED IN THE ATTACHED DEFINITION. IF NOT, THANK YOU FOR THE ABOVE DATA. PLEASE RETURN THE QUESTIONNAIRE TO THE STUDY OF POSTDOCTORAL EDUCATION, NATIONAL RESEARCH COUNCtL, 2101 CONSTITUTION AVENUE. WASHINGTON. D. C. 20418 11. Lut below in chronological order the postdoctoral appointments (excluding clinical residencies and internships) you have held, giving the information requested. Field Period of appointment Appoint- Speclal- (Including renewals) Institution merit type« R•• isation". .Insert the appropriste code from below: 1 — Research associate: appointment under research grant funds — Nffl Postdoctoral Trainee: appointment under training grant --N1H Postdoctoral Fellow: awarded in national competition — NSF Postdoctoral Fellow: awarded in national competition — NSF Senior Postdoctoral Fellow: awarded in national competition — Other government fellowship: awarded in national competition — Other non-government fellowship: swarded in national competition — Other ••Check if part of residency training following the M D '••Use Aeid numben from the attached specislities list

276 APPENDIX A: THE QUESTIONNAIRES 12. Did you have • postdoctoral appointment within on« year of receiving your doctorate? If yes. please anawer all questions below. If no. pleaae jump to question 16 and the following 13. What led you to choose the institution you attended for your J(nt postdoctoral appointment? Check three most important. Recommendation of faculty adviser Freedom to work in field of choice Opportunity to work with eminent scholar Stipend offered was most attractive Superior facilities, equipment and/or library Favorable geographic location To complete work started there .... The over-all reputation of the institution Penona] considerations . .,. .... Other (specify) i 14. Looking back, how do you rate your Af« postdoctoral appointment with regard to the following aspects: highly unsatfc- not appli- lauafactory satisfactory factory cable a. development of research skills b. your scientific adviser c. contact with other senior scholan d. your career advancement e. acquisition of knowledge f. work accomplished t opportunity to teach h. availability of facilities and equipment i. other (specify) 15. If you could have changed your jtrst postdoctoral experience, what would you have changed? check as many as apply a. changed nothing b avoided it altogether c. chosen a different institution d. chosen a different faculty i e. chosen a different field f. stayed longer g. cut It shorter h. waited until you had more experience j. sought more guidance k. other (specify) 16. Looking back, how do you rate your delayed or later postdoctoral appointment (s) with regard to: •atfcfactory factory a. development of research skills b. acquaintanceship with other scholan c. your career advancement d. acquisition of knowledge e. work accomplished f. opportunity to teach g. other (specify) Had no such appointment 17. Were your postdoctoral appointments periods of enhanced productivity?

277 POSTDOCTORAL EXPERIENCE QUESTIONNAIRE 18 Have you any comment* on postdoctoral education not covered in the above queationa? THANK YOU FOR YOUR ASSISTANCE PLEASE RETURN THE QUESTtONNAIRE TO THE STUDY OF POSTDOC- TORAL EDUCATION. NATIONAL RESEARCH COUNCIL, 2101 CONSTITUTION AVENUE. WASHINGTON, D. C. 20418.

278 APPENDIX A: THE QUESTIONNAIRES A-5 Institutional Questionnaire In addition to the above machine-processed questionnaires an open-ended questionnaire was designed to be answered by each institutional coordinator for the Study to determine institutional attitudes toward postdoctoral educa- tion. These were sent to the 165 schools whose postdoctoral s responded to the census questionnaire. Completed returns were returned by 125 adminis- trators.

279 INSTITUTIONAL QUESTIONNAIRE THE STUDY OF POSTDOCTORAL EDUCATION SPOMOCM! by A. NATlONAL RESEARCH COUNCIL NATIONAL ACAOKMT or SCIIHCM — NATIONAL ACADEMY or EncwuaiiM IM3TITUTIONAL QUESTIONKAIHE 1. Name of Institution 2. Name and title of person completing this questionnaire_ 3. What is the rationale of your institution in promoting postdoctoral study? U. The postdoctoral population a. Do you have reason to feel that you have too many or too fev postdoctorals at your institution at present? b. Do you feel that the proportion of foreign poetdoctorals in any cause of concern? Please give us any evaluation or recommendation you wish to make. 5- Selection and appointment procedures a. Would you like to see a change in the relative numbers of postdoctorals on national fellowships, postdoctorals on research grants, and postdoctorals on training grants? Should the funding agencies be encouraged to support one type of appointment more than another?

280 APPENDIX A: THE QUESTIONNAIRES b. Do you feel that your Institution.s procedures for the admission of post- doctoral! on fellowships or for the appointment of postdoctoral* on research or training grants are adequate to safeguard academic standards or do they need to be changed? Do you feel that your institution maintains sufficient control over the duration of postdoctoral appointments, or should this control be tightened? How long should a postdoctoral be allowed to stay? Research and teaching a. What do you feel has been the effect of your postdoctoral commitment on the quality of research at your- institution? Please cite any evidence you may have. b. Does your institution have any policy of involving postdoctorals in teaching Are there opportunities here which should be developed? c. What do you feel has been the effect of your postdoctoral commitment on the quality of your undergraduate and graduate programs? Please cite any evidence you may have.

281 INSTITUTIONAL QUESTIONNAIRE 7. Institutional arrangements in support of^poatdoctoral education. a. Does your institution finance any postdoctoral appointments out of its own funds? To what extent do you provide to your younger faculty members opportunities for their continued education in research comparable to the opportunities for research afforded by postdoctoral appointments? To what extent can Junior faculty appointments serve the same purpose as postdoctoral appointments? c. Have you developed any administrative structure (such as a school of advanced study) to .provide for the needs of postdoctorals? Have such arrangements proved effective? What do you feel needs to be done to integrate postdoctorals into the academic community? Funding a. What do you estimate is the net cost to your institution of accepting a postdoctoral who comes with hia stipend or salary paid but with no other support?

282 APPENDIX A: THE QUESTIONNAIRES b. Does your institution place any restriction on the number of such postdoctorals, or of other postdoctoral), a department may admit? If it doeg not, has it considered doing so? Do you charge, or have you thought of charging, a post- doctoral feet The federal agencies provide relatively «»«n grants ($500 to $1,000) towards the expenses of postdoctoral fellows. Can you make a case that these grants should be increased, or is it likely that the federal agencies. support in the aggregate of postdoctoral research appointees, postdoctoral trainees, and postdoctoral fellows covers the cost of their education? 9. Other comments:

APPENDIX of Data B-1 Fine Field Distribution of Postdoctorals Each respondent to the Postdoctoral Census Questionnaire was asked to spec- ify his postdoctoral field by using the three-digit code shown in the following Specialties List. This is the code used by the Survey of Earned Doctorates of the National Research Council. For the purpose of presentation of data, how- ever, the three-digit codes were grouped into larger subsets and identified by generic phrases. Since these subsets do not always correspond to the group- ings in the Specialties List, we present below the groupings used in this study: Field Inclusive Codes Mathematics 000-099 Astronomy 100 Physics 110-199 Chemistry 200-299 Earth Sciences 300-399 Engineering 400-499 Agricultural Sciences 500-509 Basic Medical Sciences 520, 540, 564, 534, M42, M43, 536, 530 Biosciences All in 520-599 not listed above Psychology 600-699 Social Sciences 700-799 except 730 Arts and Humanities 800, 730, 810-830, 840, 888, 889 Education 900-999 Professional and Other 850-880, 899, unknown Internal Medicine M10-M19 Other Clinical Medicine M01-M06, L21-L50, M20-M94, except M42, M43 Allied Medical Sciences L01-L15, L60-L90, 510-519 283

284 APPENDIX B: COMPILATIONS OF DATA When even coarser groupings were indicated, the following designations were used: Engineering, Mathematical and Physical Sciences (EMP): Mathematics, Astronomy, Physics, Chemistry, Earth Sciences, Engineering Biological Sciences: Agricultural Sciences, Basic Medical Sciences, Biosciences Medical Sciences: Internal Medicine, Other Clinical Medicine, Allied Medical Sciences Other Fields: Psychology, Social-Sciences, Arts and Humanities, Education, Profes- sional and Other Because these groupings are somewhat arbitrary we have included Table B-l, displaying information on postdoctorals by fine field. In only the following cases have we combined two or more three-digit codes: Pathology: 534, M42, M43 Education: Administration: 930, 933, 935 Educational Psychology: 630, 635, 910, 915 Guidance and Counseling: 940, 945 Measurement: 920, 925 Methods: 970-996 Philosophy: 900, 903, 905 Special Education: 950-958 General Dentistry: 516, L60 Optometry: 515, L71 Pharmacy: 511, L80 Public Health: 512, L14 Veterinary Medicine: 513, L90 Medical Sciences, Other: 510, 514, 518, 519, L01

285 FINE FIELD DISTRIBUTION OF POSTDOCTORALS SPECIALTIES LIST Fields Not Elsewhere Classified 899—Sci. General; Sci., Ocher; Other General Field Engineering 400—Aeronautical & Astronautical 410—Agricultural 420—Civil 430—Chemical 435—Ceramic 440—Electrical 445—Electronics 450—Industrial 460—Engineering Mechanics 465—Engineering Physics 470—Mechanical 475—Metallurgy & Physical Met. Engin. 480—Sanitary 485—Textile 498—Engineering, General 499—Engineering, Other Agricultural Sciences 500—Agronomy 502—Animal Husbandry 504—Fish & Wildlife 505—Forestry 506—Horticulture 508—Agriculture, General 509—Agriculture, Other Medical Sciences (For Medical Specialties, see reverse side.) Mathematics 000—Algebra 010—Analysis 020—Geometry 030—Logic 040—Number Theory 050—Probability, Math Stat. (see also 544, 670. 725, 920) 060—Topology 080—Computing Theory 4 Practice 085—Applied Mathematics 098—Mathematics, General 099—Mathematics, Other (note also 984: Math Educ.) Physics and Astronomy (Note: Theoretical scientists mark "T" on questionnaire following code No.) 100—Astronomy 110—Atomic & Molec. Physics 120—Elect romagnetism 130—Mechanics 132—Acoustics 134—Fluids 136—Optics 138—Thermal Physics 140—Elementary Particles 150—Nuclear Structure 160—Solid State 198—Physics, General 199—Physics, Other Chemistry 200—Analytical 210—Inorganic 220—Organic 230—Nuclear 240—Physical 250—Theoretical 260—Agricultural & Food 270—Pharmaceutical 298—Chemistry, General 299—Chemistry, Other (see also Biochemistry, 540) Earth Sciences 300—Mineralogy, Petrology, Geochemistry 310—Stratig.; Sedimentacion 320—Paleontology 330—Structural Geology 340—Solid Earth Geophysics 350—Geomorph., Glacial Geology 360—Hydrology 370—Oceanography 380—Meteorology 390—Applied Geol.: Geol Engr.; Econ. Geol.; Petroleum Geol. 398—Earth Sciences, General 399—Earth Sciences, Other 510—Medicine & Surgery 511—Pharmacy 512—Public Health 513—Veterinary Medicine 514—Hospital Administration 518—Medical Sciences, General 519—Medical Sciences, Other Biological Sciences 520—Anatomy 522—Cytology 524—Embryology 530—Physiology, Animal 532—Physiology, Plant 534—Pathology 536—Pharmacology 540—Biochemistry 542—Biophysics 544—Biometrics, Biostatistics (iee also 050. 670. 725, 920) 550—Botany 552—Phytopathology 560—Ecology 582—Encomology 570—Genetics 562—Hydrobiology 564-Microbiology 580—Zoology 598—Bio-Science, General 599—Bio-Science, Other

286 APPENDIX B: COMPILATIONS OF DATA SPECIALTIES LIST (CONTINUED) 830—Music 840—Philosophy 815—Speech S< Dramatic Arts 888—Arts & Humanities, General 889—Arts it Humanities, Other Prof. Fields Not Listed Above 850—Business Administration 855—Home Economics 860—Journalism 865—Law, Jurisprudence 870—Library & Archival Science 880—Religion & Theology Education Psychology 600—Clinical 610—Counseling & Guidance 620—Developmental & Gerontological 630— Educational 641—Experimental 642—Comparative 643—Physiological 650—Industrial & Personnel 660—Personality 670—Psychomecrics (see also 050. 544. 920) 635—School Psychology 680—Social 698—Psychology, General 699—Psychology, Other Social Sciences 700—Anthropology 705—Archeology 745—Area Studies (specify area) 720—Economics 725—Econometrics (see also 050, 544, 670, 920) 727—Statistics 730— History 740—Geography 755—International Relations 750—Political Science, Public Admin. 760—Social Work 710—Sociology 798—Social Sciences, General 799—Social Sciences. Other Arts & Humanities 800—Art, Fine & Applied (incl. hist. & crit.) f 810—Eng. & Amer. 820—Modern Foreign, unspcc. 821—German 822—Classical (specify) 823—French 824—Spanish 4 Portuguese 825—Linguistics 826—Italian 827—Russian 828—Other Slavic 829—All other modern lang. 810-829 Lang.. and Lit. Noce: For fields 900-947 and 960-967 final digit indicates level: 0—unspeci- fied; 1—preschool; 2—elem.; 3—secon- dary; 4—teacher training; 5—higher educ.; 6—adult educ.; 7—other. 900—Foundations: Social, Philosoph. 908—Elem Educ., General 909—Secondary Educ., General 910—Educational Psychology 920— Educ. Meas. & Slat. 930—Educ. Admin. 4 Superv. 940—Guid., Couns, Student Pers. 950-959—Special Education 950—Field Unspecified 952—Gifted 954—Speech 956—Phys. Handicapped 958— Emot. & Ment. Handicapped 960—Audio-Visual Media Note: For fields 970-997, and 952-959 even number is for secondary level; next odd number indicates other than secondary level. 970—Agric. 988—Phys. Ed., Health 972—Art & Recreation 974—Business 990—Science Educ. 976—English 992—Social Sci. Educ. 978—Foreign L. 994—Vocational Educ. 980—Home EC. 996—Other Special 982—Ind. Arts Field 984—Math 998—Educ., General 986—Music 999—Educ., Other

287 FINE FIELD DISTRIBUTION OF POSTDOCTORALS MEDICAL SPECIALTIES LIST (For use with the Postdoctoral Survey) M01 Administrative medicine M02 M03 M04 M05 Anesthesiology Chemotherapy Dermatology General Practice M06 Geriatrics M10 Mil M12 Internal medicine, general Allergy Cardiovascular disease M13 MI4 M15 M16 Endocrinology Gastroenterology Immunology Infectious disease M17 Metabolism M18 M19 Nephrology Pulmonary diseases M20 Rheumatology M31 Obstetrics M32 M35 M36 Gynccology Oncology Ophthalmology Otolaryngology M37 M42 M43 M44 Anatomic pathology Clinical pathology Hematology M90 M9I M92 M93 M94 Pediatrics, general Pediatric allergy Pediatric cardiology Pediatric hematology Pediatric neurology LOl Physical medicine & rehabilitation LI 2 Aerospace medicine L13 Occupational medicine L14 Public health LI5 General preventive medicine L21 Psychiatry L22 Neurology L24 Nuclear medicine L25 Radiobiology L26 Clinical radioisotopes L30 Radiology L32 Radiological physics L40 General surgery 1,41 Cardiovascular surgery L42 Colon & rectal surgery L43 Neurological surgery L44 Orthopedic surgery L45 Plastic surgery 1.46 Thoracic surgery L47 Urology L50 Tropical medicine L60 Dentistry, general 1.61 Dental public health L62 Endodontics L63 Operative dentistry 1.64 Oral pathology 1.65 Oral surgery 1.66 Orthodontics 1.67 Pedodontics L68 Periodontics L69 Prosthodontics L71 Optometry L72 Osteopathy L73 Podiatry (Chiropody) L80 Pharmacy L90 Veterinary medicine

u X 'v, CD •*• M » r'- o* m p- <M o ^ o* i>\ <vj rs, -^ <n rg --i t f OG THEORY C PO OCSt GECERAL flCSt OTHER , MOLEC. PHYS MAGCETOSM »RY PARTICLES 1 X o 1STROCOHY STRUCTURE _| o oe e 1 c 4 Ui new 'HYSOCS <J 3 t HEORY c ^ r : >- in 1 UJ o t X :s < I 0 < >- < - : : — ^ 1 - 5 < < < z *l ~ z S * ^ « n r LU _> . ; i S LOGIC £ TOPOL X X 1 — = a S Z — I _> _J _| UJ 2 OPTOC QC >rt t E e 2 5 E z c o </) o >- 0 UJ > 3 C x x : 288

(M — rg tft •+ rv r* r- .* G IM fi »o «•> f" ,-. „, « —« <-4 m 00.00 mr~ a:* m — * — - m «* — <M«Di\Jff-"rM« ~ * -' •* ry r~ IN » <v in ^ IM N -i .O — «^r-»***^« — IM M — « fyir — au-j^^ Ml — ^ 2 —. f l*( _< ,~i 'r «t tfl - -4 tf> »O •« 7 n M ^ w ^ .H Qk o «H IM r' en M r- gQ .^ -< O — rg iM M« 7 m O — f si I s | s r __ a ! iuiAU2ac«/)< M «- « o ™ «— aj iuiuz o MU g -J XbOZluo M M « O -J O-JIU>- — >1-Za e X O o * HJ-WXXOUOUJK- «« • IL acae>d« o a a, c — ^ XM <rt aj UJ a U O O (^ J » U < U>- .«3 ^ XXt-»« laj UlQOtf) UJX X < tu »- ui tn • -j d o»iuiu » Xa. £ _J O O C — ti.' _J O >-> >-O OX _l -J 4 «- _l XUO >• - I O I 4 Z1-<« (AOd * or _j t- 4»»iuztu» o >- a. rj r- a _i tuaeooae -< o Z Z ^ ^ >* >> H- O —• LU ^i •- a a. o — lo—.OaG"'n:iTOOLLF'~'»-1- -J OZaeac«»O9 »-3»-<«-*<uuj«/>V>.jw)UJae_ja:«az «P _j ai i D — o a: ac uj IAJ i- z ~t voo>-ouvzx-t«— « « — cccaezcc»-oi.i<ii.Zox — -IH-1-ZZIO<J — — -)ae-i<aeiflorit-i-Jtaeiuiu<aj~<Dr-»-t-zo — <l"-oo— "-SI* ajOZz5aiXIIIr-ata.iv>-aj"".»'<C»-»-<<»-OUJOiaJi — jjzzziuua

S£ "o,. I X Q — M — « « if, C " •* <M m O « CO » <n M -• * O rv •* —' rt 9 -« « « 0 « -• « 2 --,«-, j, ~ - - IT ™ -i (»! - « * — 1n o(M -«<«i* -«» - « -» - " " i - - * 2 s -2 r- in 9- 0 4* — 5 2 , = = 3 " i it rv (X i..:•» ir• •* u" .C M «M -. -i — m rv » - f 8 . - - o £ ^ in r• • ^ --**« ^^mx^pg* — m - ^ IM — M M m rg m iM -. a. - pg — IN. " w £ - <NJ « ft 2 " (M J- .— J" C .0 7 ^- (Ni •« IM — rvi o „ o M </> 1 4 O oe 2 > z O 41- , n Z 0 < « £ I § - c ^ UU i at oe x LU m O O t- 1- 0 OJ D 0 LU ,~i.-r< y. -- 3 a) o I •— x a o. O UJ V) — * UJ — < II, O ~ • £ «- oe z X 2 5 > i > 2 5 x a o »-O c>wiwikflt« > i i it, . £ z J - 2 ~> t> ° " J 3 8 S f Z 0 s 2 I S z S g 5 acc — <Co1-<>-C««CC 5 > 3 f 5 O - z 290

s - s -i a -» * KI (M —l M - 2 § - rTi 0" rg — 0" C | ae at t/> X ^ 1 § I O lal 1- Z 0 Uj 4 O iu 1/> 01 a. t- S at at wi z I y I 5 5 •• «- 0 — -1 « 11, M tft *- I 0 ". ~ a _l ^ < O 3 0 LU |5| S S i 5 - £ ~ • « o < -i - 5 S > tn M 8 3 S t- . > o — O ii. C > - O u< S g 5 u 5 m * 1 £ « S 3 S 1. Ic 1/> ; o — « a g 3 3 S £ S S 5 j tu - g s - 8 — — _< o o « -> « C 2 = i ? < _J O O •« i * ; C S I a. a g S S £ > ~j I1 J S 2 I v> O f T - s s I 2 J ^ g g 2 S S S S 5 S 291

_i X QC a- Lij < 4 .U D i £ s ac. n 0. lij r T GEKMA o 0 2 o Z •J Z S a 5 1 I LO Z i 1 O s _j Z oc C D Of a. o D D - c 1 1 I 1 1 ^ z ~ ' u. § 3 z o o X 3 i 4 a X I . C a. . . . J C L/-• U - •J — 558 C a < z — c o • < t- Z < s s g -- r — Is u o *- z § s 292

P- O (M * « -0 4 TRATOOC 12 2 ce 15 &L SCIECCE 2 >/> i 1 C LL.. z o c u. 4 >- > > -J O J 0 0 > _l < ^ M « .~J « * * ^ c z c ;•) < Z Z o C 0 ri <U M Z O 0 I UJ I/> X i/> > < i/l O § i 0 « uu iu ° - * O u, * 1 X Of. n a < 2 § =5 J > 8 ^ 0 O U) i 5 S UJ _j u. X UJ 0 * " O X z § O l/) z o </> — U z £ d£ t^ at 3 i 5 -COCOJ-0 X 0 > s — X s — QC t- I/I 1- UJ Iu < -i a - L 2 » o I O >- >- oe c »-OOO1-ZOC « * b « QC 1- *- H- * " Z 1 5 » UJ * * " U "- S £ 5 ° ^|g||w*| gc -1 — 0 K U X ', s 0 0 o s ? ^ C O •« « z z z o 2W & 293

X 5 z Ul Q_ Q- 294

-. — •* i- « r- iri •* if> X X */^ 1/X —,i•--- V ^ ^ — S S — » -« rg *T. o or- •» — c -• .*) .•»i -• -i o -i a "' ^ S * IT 1n 7 * o' — o — ' r\. .C C C 3- C s a* — -« « rM O * .r s - s 1 r- -. * | 1 <M rg oc .C ^r fv. r-- r~- •* ] i« m 1 1 - - * « ; * 5 c 1 i 1 i S onnaire. ~ m J ^ s Ol (V i i I c S 1 employer a f ~ s 1 i » — o * 0 1 i 0 W a 5 3 e a £ A — i fi ^O 9 ,T 0" ! 0 I e £ •* r- r- - « .n |* 1 s C i 111 ;C»L SCOECCES MEDOCOCE UJ X .1ED MEDOCAL orals did not rep ctorals did not r< itorals did not re| , Office of Scieni 0 Q • MEDOCOCE ' MEDOCONE o i M i C; HI 1 O * ^ z I ! I V 0 Of > - X < tr z C i & 1 Z a. « a £ 4 Z J* j -a ^ .a 3 *" ! C D O X > ~J -U "- •- J c i a « 1 g Lii 8 a S X at a a. 2 LU LLI >" LLI ? 8 1 X 0 i.l o 4 295

296 APPENDIX B: COMPILATIONS OF DATA B-2 Distribution of Postdoctorals among Universities For the purpose of presenting data on the distribution of postdoctoral s among universities, we have grouped the universities on the basis of reputation. This grouping was determined in part by using Cartter's ranking of departments1 and in part by the productivity of the institutions with regard to doctorates, especially in the sciences. In particular, following a compilation2 by H. W. Magoun of Cartter's data, the top ten ranking institutions in each of six major field categories were grouped together and labeled "ten leading universities." The next twenty institutions in each of the major fields were grouped and labeled "twenty other major universities." Below these categories, the further use of Cartter's rankings seemed to us to be much less valid and another means was used to categorize institutions. Using data from the Survey of Earned Doctorates3 the remaining universities were divided into two groups depending on their production of doctorates. If an institution had produced 200 doc- torates in the physical or biological sciences between 1920 and 1961, or if it had produced 400 doctorates in all fields between 1950 and 1961, or if it had been included among either the first ten or the next twenty in any field, it was included in a group entitled "established universities." All other doctoral insti- tutions were grouped together and labeled "developing universities." After this process was completed, the lists were scanned and several institutions that had been created in the 1950's or 1960's and that had rapidly developed into established institutions were shifted into the "established" category from the "developing" one, e.g., University of California at La Jolla. Separate rankings were compiled for the physical sciences (including mathematics), engineering, the basic medical sciences, the plant and animal sciences, the social sciences, and the humanities. Since Cartter's ratings did not include the medical and other professional fields, a number of institutions were not rated. Also unrated were some new institutions and non-members of the Council of Graduate Schools in the United States. Table B-2 gives the number of postdoctoral s at each institution reporting postdoctorals and the number in the major fields by which institutions were rated. Associated with each institution and field is the rating used in this study (1 - ten leading, 2 - twenty other major, 3 - established, 4 — developing, 1Allan M. Carttei,An Assessment of Quality in Graduate Education, American Council on Education, 1966. 2H. W. Magoun, The Cartter Report on Quality in Graduate Education, Journal of Higher Education, Vol. XXXVII, No. 9, December 1966. 3NAS-NRC, DoctorateProduction in UnitedStates Universities, 1920-1962, Publ. 1142, Washington, D. C., 1963.

297 DISTRIBUTION OF POSTDOCTORALS AMONG UNIVERSITIES U - unrated). Also included in the table are the numbers in clinical medicine and the numbers in all other fields. Two points should be stressed. The numbers given represent the numbers of those postdoctoral s who responded to our Census questionnaire in the spring of 1967. There probably were more postdoctoral s at these institutions and, in some cases, substantially more. The other point is that the ratings by reputation are somewhat arbitrary and were determined by dated information. Disagreements with how a particular institution was rated are not only pos- sible but even valid. The information is provided here to allow institutions to interpret where their university is represented in the tables and in the text.

Z«ZOOOZ« ae<«1- — — * (C <B >- 298

o £ 2 -. o — — a r* tfi CM -' rv rv - — * (M •JT m » .« IM o JOHCSTOWC J a. NIVERS1TV i TECHCOL e e IX Z o a > 0 >• a. I 5 M i O o 5 wi Z z >- o m > tu X o > 1 JS COLLEG UJ z % UCOVER lU •Si i z § I . UCIVERS DOVERSOTY 4E UCOVER 4 UCOVERS SHOCGTOC > O o o o o o z 3 O Z 0 Z >- $ 3 > > 0 5 " . 0 D O 3 X •• 3 jj Z o a. i X U M COCCOC CLARE" « CLARKS ZOO -J O O COCCEC CORCEL DARTHOO a e O Z x DUKE U DUOUES | FLOROD FLOROD FORDHA IU GEORGO GEORGI X IU -i « a O 4 O O c c a," o O at 0 Q 0 0 « 299

.£ § V ob ui m o z LU a. I LU ^ ii si s ° X u. o J X — 0 o 0 C u > t- S O U. o ., s «/> z — — I/I t- U. a: — - I !• U S I — O at £ J; > >- — u. c- X a: 0 3 HI I/I OC _ o UJ >/> t. > ae f •> ~ 3 | ~ z LL, VI z > u 3 ~ C . at iu _l ] > X 0 z > D u. Z - 03 Z 0 Id UJ X laj Z § 0 >- D 1- Z 3 > > U > 3 = § SOOCO1 n «OHV i/i Z o * s >- >- •RECCE Z _J 3 _ 3 Z o z « M 1 1 i X t- 10 < at z 8 >- >Sl l/t 300

-I -« tf* *fi .f * .e — — ,0 _. — g* « — I« .O (7- Z « o .j, c >- — X O U >- -J t- C -* Li — > O « tu > O LU « — 4- .yi OO>_>— ujIu.1-D— > — 2 ' ~ -j C —* > • - «-"-5"-ZO — «-u. • a .- 1 _j >-• a - O C — - a. a 3 U XOt-OOXD C O oj UJ ° D 2t_i«— -T.I X XiU X X X X X X C X* * 301

302

•o if # ~t t K O i i -. 0- — .0 ^ in -. — -• — «n -t ^ rf\ o- r- i i i i i •8 1 CD m O« -• Z > — »- < u ID Ouji/l — D kO X C>0 ^ It« Qt- — Z -< D — _i »OOO« o ci. -J J- Z>D S^ U>-^cO»- — — UOOul< — ZIvoo"z< u. uo aeZZZZZZ >>ot<£:zac=;> I 1 .2 303

304 APPENDIX B: COMPILATIONS OF DATA B-3 Distribution of Foreign Postdoctorals by Country In this study data on the foreign postdoctoral were presented for the most part by gathering the home countries into four groups determined by the per capita gross national product. The rationale was that degree of educational development is more likely to be a function of national wealth than geographi- cal location. As with any categorization, there are flaws, and countries like Kuwait will be ranked as a rich country although its educational development does not match its wealth. (There is not much distortion in this case, however, as we detected no postdoctorals from Kuwait.) The classification of countries by per capita GNP is based on World Bank figures,4 and the nomenclature we used is: Classification Per Capita Gross National Product High income More than $750 Medium income $250-$749 Low income $100-$249 Very low income Less than $100 In Table B-3 we provide data on postdoctorals, listing each country sepa- rately. The per capita GNP classification is given with the code: High — 1, Medium — 2, Low - 3, Very Low - 4. Escott Reid, The Future of the World Bank, International Bank for Reconstruction and Development, September 1965.

305

5 3 CO CD CD 2 X Q Z (M i«i IM .a •r tr rv -s -t -• — « m m — m — «n iM UN in —i — f"lQr\J--.CI<MrV*i-~—' <M O tf> •*• -• m — •* — m — z Q irt 4 4 4 « Z s w» at a ae i £ § • 4 irt _J (/I 4 ft. ; i ; S i u i 5 s i a z o _j 3r- >- -a .a uj 'U a. ° Z a !ii|i§!|ii * ^ 4 Z Z « ~i in <i •a u S 3 0 « 306

ss — -« rv — PJ <f, O -• — — r\I O f, z • s Z Q < — a. «i S 8 3 OOO OOOO.XXZZD Q.aae^irt^3D>- 3 -si C UJ 307

31 s ec — 3 Q z O u. D O 308

APPENDIX iibJtography American Medical Association, The Graduate Education of Physicians, The Report of the Citizens' Commission on Graduate Medical Education, 1966. Berelson, Bernard, Graduate Education in the United States, McGraw-Hill, 1960. Postdoctoral Work in American Universities, Journal of Higher Education, March 1962. Bush, Vannevar, Science, the Endless Frontier, a Report to the President, Washington, D.C., U.S. Government Printing Office, 1945. (Also reprinted by the National Science Foundation, 1960.) Cain, Arthur S., Jr., and Lois G. Bowen, The Role of Postdoctoral Fellowships in Aca- demic Medicine, The Journal of Medical Education, Vol. 36, No. 10, October 1961. Cartter, Allan M.,An Assessment of Quality in Graduate Education, American Council on Education, 1966. Ingraham, Mark H., The Outer Fringe: Faculty Benefits Other Than Annuities and Insur- ance, University of Wisconsin Press, 1965. Journal of the American Medical Association, annual education numbers. (Contain de- tailed information each year on the postdoctoral population in medical schools.) Manpower Resources for Science and Technology, The Brain Drain, Her Majesty's Sta- tionery Office, London, 1967. Miller, John Perry, Under the Tower, the Postdoctoral Fellow, Ventures (magazine of Yale Graduate School), Vol. 5, No. 2, Fall 1965. The Modern Language Association of America, Recommendations Concerning the PhD in English, PMLA, Vol. 82, No. 4, September 1967. National Academy of Sciences, Doctorate Recipients from United States Universities, 1958-1966, Publ. 1489, Washington, D.C., 1967. National Academy of Sciences-National Research Council, Doctorate Production in United States Universities, 1920-1962, Publ. 1142, Washington, D.C., 1963. 309

310 APPENDIX C: BIBLIOGRAPHY National Academy of Sciences-National Research Council, Profiles of PhD.s in the Sciences, Publ. 1293, Washington, D.C., 1965. (Summary report on follow-up of doctorate cohorts, 1935-1960.) National Academy of Sciences-National Research Council, Chemistry: Opportunities and Needs, Publ. 1292, Washington, D.C., 1965. (A report on basic research in United States chemistry by the Committee for the Survey of Chemistry.) National Academy of Sciences-National Research Council, Physics: Survey and Outlook, Publ. 1295, Washington, D.C., 1966. (A report of the present state of U.S. physics and its requirements for future growth.) National Academy of Sciences-National Research Council, The Plant Sciences: Now and in the Coming Decade, Publ. 1405, Washington, D.C., 1966. (A report on the status, trends, and requirements of plant sciences in the United States.) National Research Council, Office of Scientific Personnel, Proceedings of the Conference on Postdoctoral Fellowships and Research Associateships in the Sciences and Engi- neering, Williamstown, Mass., Sept. 10-12, 1967. National Science Foundation, Graduate Student Support and Manpower Resources in Graduate Science Education, Washington, D.C., 1968 (NSF68-13). Perkins, Dexter, John L. Snell, and the Committee on Graduate Education of the Ameri- can Historical Association, lite Education of Historians in the United States, McGraw- Hill, 1962. Research Policy Program, Brain Drain and Brain Gain, Lund, Sweden, 1967.

LB 2371 .154 1969 The Invisible university: postdoctoral education ir 71 .154 1969 The Invisib1e university: postdoctoral education in