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.
4. Behavioral Sciences- . .. Clinical behavioral science Ph.D.s are faring better than the nonclinical ones, apparently because of the additional opportunity for sel.f-emplogment as . practitioners. But tit is the nonclinical component of the behavioral sciences that conducts most of the research in the field and is the nrin~in=7 In ^' ~ ,lr ~ = ~ _ _ ~ ~ L ~ l_ 1_ ~ ha; ~ ~~_~ men ~ ~ ~ . =.~11= 1C~, A- 111= ~lu~lo~llalca1 component IS employed primarily in academic institutions where only modest growth in Ph.D. employment is expected for the next few years. But attrition from the labor force will begin to increase substantially in the l990s. On the supply side, nonclinical enrollments and Ph.D. production are fal l ing and there is no large reservoir of postdoctoral trainees to call on as there is in the blamed ical f iel ds . INTRODUCTION AND OVERVIEW '! In 1976 this Committee found some disturbing trends developing in the labor market for behavioral scientists. Both the National Science Foundation and the Bureau of Labor Statistics had just published studies warning of a potential oversupply of Ph.D.-level scientists in many fields by 1985 (NSF, 1975b: and U.S. Department of Labor, 1975~. The life sciences and the social se fences ~ inn ~ ',~ i no Hi war! among the fields cited. The Committee's own analyses seemed ~o goof i am Hal ~ ; ^~ ; ^~= ~ .= .r ~ ~ ~~~~ ~ ~ ~ ~ ~ ~ ~ 4 ~ _.__ __ I,__ ~~ ~~~ ~s~= · ^~ ally ~ Although college and university enrollments in the behavioral fields were still growing in the early 1970s, it was evident from demographic considerations that they would soon grow at a slower rate as the "baby boom" bulge completed its passage through the college-age years. That was expected to happen in the early 1980s. After that, enrollments would begin to decline in most fields for the next 10 or 15 years. Since growing enrollments throughout the 1960s had spurred expansion of employment in the academic sector, a reversal of the enrollment 88
89 trend could be expected to cause a corresponding decline in academic demand. Those fields for which the academic sector is the primary employment market would be likely to feel the impact of this shrinking market quite heavily. In the behavioral sciences, about 65 percent of the Ph.D. labor force was academically employed in 1972. So graduates in this field obviously were quite dependent on academic employment and on enrollment trends. Furthermore, by 1975 Ph.D. production in the behavioral fields had reached a level of more than 3,900 per year, twice the number produced in 1968. At that rate of increase in supply, the labor force of behavioral science Ph.D.s was sure to grow rapidly over the next few years. The Committee was concerned that this growth, combined with the expected slowdown in demand, would eventually produce an oversupply of behavioral scientists. Yet it also recognized the need for scientists trained to study important behavioral issues in relation to health problems. The Committee's solution was to call for redirection of the federal training programs from one which emphasized predoctoral training to one which emphasized postdoctoral training. The recommended ratio of federally-supported training positions in the behavioral sciences was ultimately to be 30 percent predoctoral and 70 percent postdoctoral. This was to be achieved gradually by shifting resources within the overall program which was to be maintained at a constant level. Here is the Committee's original statement from its 1976 Report (NRC, 1975-81): The Committee recognizes the need for continued federal support of training of the behavioral scientists who are conducting research relevant to the health needs of the country. Current trends in behavioral science research, however, suggest that a significant reorientation of emphasis in the federally supported research training effort is desirable at this time. Scientific advances in these fields have vastly increased the complexity of research methods and imposed requirements for more intensive training. While the number of Ph.D.-level individuals currently being trained in the behavioral sciences appears~to meet market demands in the conventional disciplines, there is a growing need for more specialized behavioral science research training to deal with these increasingly complex research questions in the area of behavior and health. The Committee therefore recommends an orderly taper ing down of predoctoral support with a concomitant emphasis on providing for research specialization through postdoctoral training, thus assur ing the active application of advanced research training to meet the health needs of the country e The Committee recommends that the current apportionment of about 10 percent postdoctorate and 90 percent predoctorals trained in the behavioral
so sciences through this program should be modified so that ultimately 70 percent of the individuals supported by NRSA funds are postdoctoral students and 30 percent are;predoctoral students. In this way the Committee believes sufficient opportunity for training in the behavioral sciences at the postdoctoral level will be assured, while an adequate number of awards for basic research training at the predoctoral level also will be maintained. However, it is recommended that this change should be implemented gradually and at essentially a constant level of federal funding in FY 1976, FY 1977, and FY 1978, in order to minimize the dislocations that could otherwise occur for both programs and personnel. Because of the greater cost of postdoctoral training, this shift will mean significant reduction in the number of behavioral science investigators trained with federal funds during the 3-year period; however, the change is expected to enhance the quality of both the programs and the trainees.... The Committee will monitor closely the result of this change and will discuss in future reports whether modification of this recommendation is warranted. . The Committee's recommendation was not, and still is not, readily accepted by some influential members of the behavioral science community. The American Psychological Association (APA) in particular has on several occasions objected both to the Committee's assessment of the market outlook and to its recommended shift in training support. On August 25, 1982, the APA held a symposium to discuss the role . of postdoctoral training in the behavioral sciences. Several directors of postdoctoral training programs described their experi- ences in initiating and conducting their programs. In some cases, the programs seemed to be working well and providing valuable opportunities for sharpening research skills--in psychotherapy and in developmental psychology, for example. But while some program directors seemed quite receptive to the notion of expanding the role of postdoctoral training, they also~were uncomfortable with the idea of relinquishing any support for predoctoral trainees. The Committee has given these developments a great deal of consideration. We have heard statements both supporting and opposing our position from participants at the public meetings we convened to discuss our reports. We have continued to monitor the system, reviewed past data and collected new data, disaggregated the data to the extent we felt it feasible, reconsidered the recommendations and the evidence on which they were based, and discussed at length with agency represen- tatives the difficulties involved in implementing the recommendations. Based on testimony at Committee's public meeting on June 2, 1982. See Pallak (1982~.
91 It should be noted here that the constant level of funding for research training in the behavioral sciences recommended by the Committee has not occurred. The research training budget of ADAMHA dropped from $19.7 million in 1975 to $16.9 million in 1978 (Appendix Table Did. It rose somewhat in the interim but by 1982 it was down to $17.2 million--still below the 1975 level. Furthermore, in the last few years there has been a rather obvious shift away from traditional research fields into the more clinical fields of behavioral sciences. This has occurred both at the employment level and at the training level. Graduate enrollments and Ph.D. production in nonclinical fields are beginning to decline while those in the clinical fields continue to increase. So after carefully considering the arguments on both sides, and after reviewing the latest data and analyses presented in this chapter, the Committee has concluded that circumstances have changed enough since 1975 to warrant a revision to previous recommendations. Whereas the short-term outlook in earlier years appeared to support the Committee's earlier projections, current circumstances, longer-term considerations, and the difficulty of implementing the recommendations in a period of reduced budgets convince the Committee that some modification is now in order. The recommendations for behavioral science fields stated in Chapter 1 reflect the Committee's belief that predoctoral training should remain at the 1981 level (about 650 awards) during the mid-1980s while post- doctoral support should increase modestly from the 1981 level of-about 350 awards to about 540 awards by 1987. THE TRAINING SYSTEM IN BEHAVIORAL SCIENCE FIELDS Training and career patterns for Ph.D. behavioral scientists differ in some important respects from those that-are found in the biomedical sciences generally. The major differences arise from the fact that psychologists, who constitute the bulk of the behavioral science doctoral output, have open to them the possibility of independent practice in the broad areas of clinical and counseling psychology. Practice of this kind is regulated in the U.S. through the mechanism of licensing, with accompanying requirements as to the specific content of the training of the license holder. Doctorates in clinical and counseling psychology are usually eligible for licenses to practice provided that they have had a certain amount of supervised Clinical experience and upon passing an examination. Doctorates in other fields of psychology may make themselves eligible by undertaking additional training at a postdoctoral level, either through the formal mechanisms of postdoctoral fellowships designed for the purpose or through informal arrangements in clinical settings. Doctorates in psychology may be divided into several identifiable types. First, there are those who are trained in the nonclinical areas and are aiming at a research career similar in all essentials to the career patterns found in the biological sciences. Second, there are those who are formally trained in the clinical areas of psychology and whose career patterns show resemblances to those of physicians,
92 namely a pattern of research, of clinical service, or some mixture of the two. For the first group, postdoctoral training in research is an established part of normal career development--again much as is the case in the biological sciences generally. This is particularly true of those areas of nonclinical psychology that overlap directly into the biological sciences, such as in psychopharmacology, behavior genetics, neuropsychology and the like. It is less true of the non-biological nonclinical areas of psychology such as social psychology, personality and developmental psychology. Postdoctoral research training is relatively rare for doctorates in clinical and counseling psychology. ~ Because there is a channel for entry into a career of clinical service, the pool of research trained psychologists is subject to some drain-away from research activity into service and practice. When employment in research and academic posts becomes difficult to find, the pressures to seek a license and enter a clinical service career - become strong. More widespread coverage of clinical services by health insurance plans adds to the attraction of a clinical practice career. Hence, estimates of the magnitude of the pool of research scientists in the field of psychology must be moderated by recognition of the rate of shift into clinical and self-employed categories of people trained originally for research careers. As later portions of this~chapter indicate, such a shift has now been under way for some time and is increasing rapidly. Figure 4.1 provides an illustration of the "training system" in the behavioral sciences and may be helpful in describing the stages through which promising young students pass along their way to careers in research. The numerical estimates in this figure represent the average numbers of individuals each year who have followed particular career paths during the 8-year period from 1973 to 1981. Entering at the far left are 5,300 students (path A), who having earned baccalaure- ate degrees in psychology and other social science disciplines, decided to pursue doctoral study in the behavioral sciences. This group constitutes less than 5 percent of all individuals receiving B.A.s in the social sciences. Another 2,000 individuals enrolled in Ph.D. programs after completing their undergraduate training in fields outside the social sciences (path B}. At the right in Figure 4.1 is the active Ph.D. labor force--estimated to include 52,900 behavioral scientists (excluding postdoctorals) in FY 1981. Between 1973 and 1981 the loss from the labor force due to death and retirement (path I) has averaged only 300 individuals (approxi- mately 0.5 percent) annually. During this same period the total number of Ph.D.s entering the behavioral science work force for the first time (paths D, G. and H) is estimated to have been 4,200 scientists. The majority of these individuals have entered directly after completing their doctoral training (path D), while the number finishing postdoctoral training (path H) has been quite small-- especially in contrast with the biomedical sciences. As a result of the sizeable stream of new Ph.D.s and the very modest attrition out of the work force, the total pool of doctoral scientists employed in the behavioral fields has been growing at a rate of 3,900 individuals a year. Much of this expansion has occurred outside the academic sector and involved employment in the clinical specialties.
B . A . ~ DEGREES ~ .> IN 5,300 SOC I AL SCIENCES (est. size in 1981:141,500) rim l 2,000~4 I B . A . I >~ I acre c c c I ~ PH.D. PROGRAMS IN BEHAV I ORA L \500 DEGREES 1/ I I N OTHER ~ . I L F I ELDS J / 93 or- I EMPLOYMENT OR ~ I POSTDOCTORAL I ~ I N OTHER I 400 ~ I F I E LDS ~ '> I ( NONBEHAV I ORAL ) it 3,200 SC IENCES 1~ ~ ~ NEW PH . D . S I FROM OTHER ;~ 1 1 FIELDS I \ ~ _ 1 POSTDOCTORAL POPULAT ION IN BEHAV IORAL SC I ENCES 500 Hi_ 10~(e t. size in 19&1: 1 000) / 500 ~ TOTA L ACAV EM I C AND NIJNACA:DEM I C PH . D . WORKFORC E EXC LUD I NG POSTDOCTORALS IN BEHAV I ORAL SC I ENCES ~ (est. size in 1981: bi,900) 300 ~ FRET IRE- ) FIGURE 4.1 Doctoral training system in the behavioral sciences. Estimates represent the average annual number of individuals following particular pathways during the 1973-81 period. No estimates have been made for immigration, emigration, or re-entry into the labor force. A more detailed analysis of the training system in the behavioral sciences is shown in Figure 4.2. In this figure the behavioral sciences are divided into clinical and nonclinical components.2 Numbers for BA degrees and graduate school enrollments (used to calculate paths A and B) are not available separately and therefore are not presented in the figure. The average number of Ph.D.s awarded over the 1973-81 period was 1,400 for the clinical areas and twice that number for the nonclinical areas. Switching to other fields has been much more prevalent for nonclinical Ph.D.s (path C), many of whom switch into the clinical fields. In addition, more Ph.D.s are entering clinical fields from other areas than are entering nonclinical fields (paths F and G). 2 Clinical fields are counseling and guidance, clinical psychology, and school psychology. Nonclinical fields are sociology, anthropology, and nonclinical psychology fields.
94 B.A. DEGREES IN SOCIAL SCIENCES -}I ~ , ,~ B.A- I Art' 1 I IN OTHER I L FIELDS ~ DEGREES ~ PH.D. PROGRAMS IN CL AND NCL BEHAVIORAL SCIENCES ~ . 1/ I NEW PH.D.S r I FROM OTHER r l I FIELDS L ___ r - - - - - - _ _ _, I EMPLOYMENT OR I 1 1 CL: 100 ~ POSTDOCTORAL I ~ , IN OTHER FIELDS /NCL: 600 CL: 1,100 ~3 NCL. l,900 \CL: 200 CL: 300\,. POSTDOCTORAL POPULATION IN CL AND NCL BEHAVIORAL SCIENCES CL: 200 it_ NCL: 300 CL:l~(e t. size in 1981 C : 300 £) NCL: 700 /NCL: 50 CL: 500 v NCL: 200 TOTAL ACADEMIC AND NONACADEMIC PH.D. WORKFORCE EXCLUDING POSTDOCTORALS IN CL AND NCL BEHAVIORAL SCIENCES , l (est. size in 1981 CL: 23,500 NCL: 29 ,400 ) CL: 100/ DEATH ANT) CL: 200\: FIGURE 4.2 Doctoral training system in the clinical (CL3 and nonclinical (NCL) behavioral sciences. Estimates represent the average annual number of individuals following particular pathways during the 1973-81 penod. No estimates have been made for immigration, emigration, or reentry into the labor force. Movement between clinical and nonclinical specialties within the behavioral sciences is treated as field switching in the above figure but it was not in Figure 4.1. Therefore, the sum of coca and nonc~n~cai scientists following path (: in Figure 4.2 does not equal those taking path C in Figure 4.1. The same is true for paths D, E, F. and G. ~ 1 _ _ _ _ 1 _ _ 1 _ _ _ _ ~ 1 . · , ~ ~ 1 · . ~ i_ · ~ . , ~ ~ THE MARKET OUTLOOK . The latest data on the indicators of the current state of the labor market for behavioral scientists are shown in Tables 4.1 and 4.2. The complete set of data pertaining to the behavioral sciences is presented in Appendix Tables Cl-C12.
95 . - ~¢ o · - e o Id . "C5 - ~ lo it: - ;^ en A: . - c) As m at: En r" Cd ~ 0> tt ~ em CD cat CC IS: <o ° C) = - a ~ :^ 3 o o ~ I_ Ct C: ~ ~ '' ' ' . .~.. ., _ Go _ Go _ Cal _ Go _ Cat _ ED _ CC AL 0` Cal ~ ID Go c-l o ~ ~ ~ O. o Vat ~ Cal o ( o oo o o ~ ~ ~ ~ oo ~ o _ I ~ ~ I I I ~ ~ O ~ ~ ~ O ~ ~ oo ~ 0~ oo d- oo _ ~ ~ j ~ ~ ~ ~ ~ ~ ° _ ~ I t . ~ . _ o ~ ox ~ oo Ce d- Ce _ ~ ~ _ ~ ~ ~ ~ ~ ~ ~ _ · O ~ d- ~ cr~ ~ ~ ~ ~ ~ r~ ~ _ ~ _ _ ~ X ~o V) C~ C~ ~D cr~ ~ ~ ~ U~ · _ _ ~, cd X, r~ ~ - oo ~ , ~ ~ · O ~D , _ _ _ X V3 - ~r o _ _ o ~ _( o ° _ o ~ Ct Ct Ct Ct Ct Ct C~ Ct C: ~ ~ c: c: =: ~ _ oo ~ o o ~ oo ~ ~ u, 00 X ~ ~ ~ ~ ~ ~D o~ _ ~ ~ _ o o~ o o ~o o o ~ o oo X ~) ~ O O 0 00 0 O t~ O O O ~ ~ O ~ O O ~ O O t O O ~ e, ~ 0` ~D ~ ~ CD _ =_ .. ~ ~ ~ Z ~ ~ ~ ~ ~ ~ ~ ~ 2 ° ~ ~ ~ ~ ~ =; ~ 2 c~ ~~ ~ ~ c, .~~ ~ 5 .= ~ ~ ~ ~ oo s . m . . . . _ ~ ~ ~ . v, c' a~ . _ ~ <: U3 s ~ ~ _ ce ce 0 s ~ 11 .5 .+ ~ m ~ . _ .+ ce v'- ~D + _ ~ ~ . + c3 o C~ o ~ C~ X + 04 Ct C~ t) ~D C~ _ _4 ·- 11 _ _ ~ cn .c~ _ 3 , ,~ ,,= s .o C~ CO S ~D o 3 ~ 11 ~ ~ C'\ ~ ,, ~ ~ ~ . _ o ._ ,) t s 3 Ct C~ o ~ ~ > e,C ·cO~ - O _1 C: 1: ~ ~ O ~ ~ u, O ~ ~ ~9.1i 'I ~ C ~ 8 2 ~ ~a ~ ,: ~ ~ ~ ~ ~ e _ D :^ ~ ~ E o- ·C) ~ ~ a ~ C,, ;^ ~t = = D ~ ~ = U, ¢ `;Q 3 Ct x - - X - - x - - - cr - - oo X - _ ._ o ~_ C~ C~ . _ - oo u. - - oo 0N - - C~ z . _ - oo 0\ - oo X - - z . . C~ ~: o V,
96 v) C C, 6 Hi ° ~ C ~ ~ 6 ~ ,~ o ~ E ~ o C _ o or or Do or or _ _ or ,_ or ~ ~ or ~ 0 or 0 0 ~ fir ~ ~ ~ O4 Do ~ _ Do ~ _ C ~ O O X _ ~ O 1 ~ ~ ~~ _ ~~} I~) _ ·~) OD ~ o r ~ ~ , ~ ~ ~ ~ ~ ~ ~ 0~ ~ 00 _ ~ O4 0\ ~ ~ ~ ~ O I ~ _ I ~ ~ ~ ~ ° ~ ~ I ~ ~ ~ o ~ I ~ _ o _ ~ ~ ~ a~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ _ ~ ~ _ ~ o o ~ ~ ~ ~ ~ ~ ~ . . . . . . . . . . . . . . . T -~ ~ cx ~r-- - os ~ _ C ~ ~ CO o ~ o ~ ~ ~ _ > ~ ~ oo °~ ~ ~ o~ . oc ~ oo t_ t_ t_ . ~ _ _ oo ~ U~ 0 0 _ ~ r ~ . ~ . _ C ~ _ _ _ _ _ ~4 ~o ~, ~ o~ 0 ~ C · ~ U) ~ ~ . ~ C ~ ~C _ _ 0 ~ oo oo 0 ~ ~ ~ ~ O ~ ~ ~ ~ 0` ~ ~ C o~ a' oo . ~o ~ . ~ ~ _ ~ ~ V~ oo - V^, ; ~o - - - - r" ~ ce ~ ~ ~ c~ ~ a, °` O. ~ °` ~ e c c c ~ ~ c c ~ c ,,, - ~r ~ ~ ~ ~ ~ ~ ~ ~ C~ ~ C ~ C C ~ C ~ ~ X ~ ~ _ ~ C ~ ~ ~ ~ ~ ~ ~ ~ 0 0 ~ ~ ~4 ~ 0 ~ _ =` _ _ _ 00 ~ e, 1_ ~ O O0 ~ 00 ~ ~ `0 0 ~ ~ ~ `0 \0 _ _ ~ c~) OC, _ _ ~ ~ ~ ~ ~ _ . . U, ~ ~ ~ .. .. R ~ 3e ~ e . ,e ~ ~F 3= 3 ; = i U, Z ~ D t~ (i ~ C, _ 1 CL ~ D ~ ~ C, ~ U. ~ ._ ~ ._.y _ C C O ~ [J _ ,,%~ s ~ . - ~ C ~ ~ ~ E · U} ~ o C ~ ° a' _ ~ ~ ~ ._= _ C V =, ._ (,, _ <, E s c, ~ ~ s C C ~ ~ Z 5: ~ ~ Ct .+-_. o s ~ C~ ~ . ~ . o& E t' C ° ~ o Ct . ~ _ _ q) .O ~, _ to _ =, · O D ;^ ~ ' ~ o C.) · .0 _ _ C ~ O ~ t, ~ 3 D ~ ~ ~ - ; ·C o o ~ Ct, _ {_ ~ C} 0 = =-,,e ~ 0 ~ ~8 3 ao ~ Q ~ C: C ~ ,t, C ~ 3 ~ o E O e O oG 8.el ° 8 ~;a D U' E e, ,, y O ~ ~ _ Y, ,,a~ a D y3~ " '' s s ~ ~ -E, E D c, c, 3 - x o~ - o~ - ~1 x - ~o - o~ r~ o~ - r~ x oo - - ._ - :L1 o c E ;~ . ~ :t x r~ o' - - C~ z - oo o~ - x x o~ = ~ . - X z cn ;= ;~ o C~
97 The total labor force of Ph.D.s employed in the behavioral science fields continued to grow in 1981 at a steady rate of about 5 percent per year (Table 4.1, line 3a). Employment in clinical behavioral fields--defined as clinical and school psychology, counseling and guidance--is growing much faster than in the nonclinical fields-- defined as anthropology, sociology, nonclinical psychology, and speech and hearing sciences. Growth in academic employment of both clinical and nonclinical behavioral science Ph.D.s continues at a steady pace of about 3 percent per year (Table 4.2 and Figure 4.3~. Some other highlights of these data follow. EnroUments Total behavioral science graduate and estimated undergraduate enrollment have continued a decline that started in 1973. Most of the decline has come at the undergraduate level. Graduate behavioral enrollments in both the clinical and nonclinical fields have increased since 1977--up by a strong 6.5 percent per year in the clinical fields and by a more moderate 1.8 percent per year in the nonclinical fields, although the latter declined somewhat in 1980 and 1981 and appears to be headed down. Ph.D. Production The trend in Ph.D. production reflects the same general pattern that prevails in most other parts of the behavioral science area--up in the clinical fields and down in the nonclinical ones. In terms of total behavioral Ph.D. production, these trends tend to balance out. Total behavioral Ph.D. production has been essentially level since 1970 after a long period of growth, although both clinical and non- clinical Ph.D. production rose in 1981 (Figure 4.4~. Clinical Ph.D. production in 1981 rose 10 percent over the 1980 level, continuing the rising trend that has been evident since 1975. Nonclinical Ph.D. production showed an increase of 4.5 percent over 1980, but the long-term trend since 1975 has been one of gradual decline. The peak production was in 1976 (2,897} and no subsequent year has returned to that level. As 1981 is the only year since then in which there has been an increase in output over the preceding year, and even so is below the 1979 level, it appears to be minor upward fluctuation in a basically declining curve. Postdoctoral Appointments There is one item that reverses the general pattern--postdoctoral appointments in the clinical as well as the nonclinical fields dropped in 1981 (Figure 4.4~. However, the growth rate in postdoctoral appointments since 1975 is almost twice as high in the clinical fields--8.5 percent per year compared to only 4.4 percent per year in the nonclinical ones {Table 4.2, lines lo and if).
98 30 _ 25 u' ~ 20 s - c, 15 o ,,, 10 z O6 8 ~n ~5 c~ ~ 2 ~: ~ o z All School s Publ ic ~~ School s - ~_ L 62 64 66 68 70 72 74 FI SCAL YEAR Private _~ School s I ~ 1 ~ 76 78 80 82 (a) Ph.D.s employed in all behavioral science fields at colleges and universities. See Appendix Table C7. c~ ° 10 c~ ~ 5 m z o Pri VAtP : ~ 62 64 66 68 70 72 74 76 78 80 82 FI SCAL YEAR (b) Ph.D.s employed in clinical behavioral science fields at colleges and universities. See Appendix Table C8. 62 64 66 68 70 72 74 - All School s Publ i c __ Schools Pri vate __~ School s 76 78 80 82 FISCAL YEAR ( c ) Ph . D . s empl oyed i n noncl i ni cal behavioral sci ence f i el ds at colleges and universities. See Appendix Table Cg. FIGURE 4.3 Ph.D.s employed in the behavioral sciences at colleges and universities, by control of institution, 1962-81.
99 5,000 4,000 CY 3,000 ' LL a, : an 2,000 1,000 2,000 1, 500 em, 1,000 an 500 3,000 2,500 2,000 1, 500 1, 000 500 O All ___ School s ~ . Postdoctoral Appoi ntments ~ Ph. D. s Awarded i Publ ic / he_ School s Pri vate a, ~ School s 60 62 64 66 68 70 72 74 76 78 80 82 PI SCAL YEAR (a) Postdoctoral appointments and Ph.D. degrees awarded in all behavioral science fields. Ph . D . s Awarded _ Postdoctoral Appointment ' 60 62 64 66 - 68 70 72 74 76 78 80 82 FI SCAL YEAR (b) Postdoctoral appointments and Ph.D. degrees awarded in clinical behavioral science fields. All School s Publ i c Schoo ~ s Pri vate School s Al' Publ i c Schools Ph. D. s Awarded/ ostdoctora ~ AppO1 ntments ~. ,,~~- 60 62 64 66 68 70 72 74 76 78 80 82 FISCAL YEAR (c) Postdoctoral appointments and Ph.D. degrees awarded in nonclinical behavioral science fields. FIGURE 4.4 Behavioral science postdoctoral appointments and awarded Ph.D. degrees, 1960-81. See Appendix Tables C3 and C4.
100 R and D Expenditure Behavioral science R and D expenditures in colleges and universities increased by almost 7 percent above inflation in 1981 from the previous year. This is the second straight year of increases after a long series of declines dating back to 1974 (Figure 4.5 and Table 4.1, line 2a). 150 125 100 75 50 All School s Public Schools Private Schools 60 62 64 66 68 70 72 74 76 78 80 PI SCAL YEAR FIGURE 4.5 Behavioral science R & D expenditures in colleges and universities, by control of institution, 1 960-80 (1 972 $, millions). See Appendix Table C11. Employment Sectors In 1981 there was a large jump in the number of behavioral Ph.D.s who were self-employed--up almost 40 percent since 1979 {Table 4.13. Self-employment of clinical behavioral Ph.D.s has been growing sharply for some time. There are now more self-employed clinical behavioral Ph.D.s than are employed in any other sector including academia (Table 4.2, line 3p). Even many nonclinical Ph.D.s--for whom self-employment has not been a very significant factor--shifted into self-employment in 1981. For this nonclinical group, self-employment more than doubled in 1981 from the 1979 level. Employment in the business sector also showed large increases in 1981 for both types of behavioral scientists.
101 Short-Term Projections of Academic Demand Having discussed the current market situation and given consideration to the issues surrounding predoctoral and postdoctoral training in the behavioral sciences, we turn now to an assessment of the near-term and long-term outlook for scientists in this area. Because academic employment of behavioral science Ph.D.s has been increasing while enrollments have been declining, the Ph.D. faculty/ student ratio (F/S) has moved sharply upward since 1972 (Appendix Table Con. The explanation for the strong and continued growth in the F/S ratio has proven to be doggedly elusive. In the biomedical and clinical fields we have been able to construct models of academic demand based on the concept that such demand is dependent on three main academic activities: teaching, research, and, in the case of clinical faculty, patient care. The faculty/student ratios in these fields show a strong positive correlation with R and D expenditures and professional service income. But no such relationship can be demonstrated in the behavioral fields. The F/S ratio for behavioral Ph.D.s has continued to increase substantially during the 1970s, while R and D expenditures in the behavioral sciences at colleges and universities have been generally declining in real terms. A plausible explanation is that Ph.D.s have been steadily replacing non-Ph.D.s on behavioral science faculties. This phenomenon--known as n enrichmentn--has apparently been taking place at least since 1966. At that time, Ph.D.s constituted 56 percent of total behavioral science academic employment. Since then that percentage has steadily increased--by 1981 it was almost 75 percent (Table 4.3~. So either the growth in academic employment of behavioral science Ph.D.s has been at the expense of non-Ph.D.s, or many non-Ph.D.s have attained their doctorate degree while academically employed during the last 15 years. Another small portion of the growth may be due to a shift to part-time rather than full-time employment. These factors go part, but not all the way, toward a full accounting of the growth in academic employment of behavioral Ph.D.s over the last decade. After all, we still must confront the fact that total behavioral science academic employment--including Ph.D.s and non-Ph.D.s--although declining slightly in recent years, has remained at relatively high levels while estimated total behavioral enrollment has fallen rather sharply. Another possible explanation is that our enrollment estimates are faulty. Data on undergraduate enrollment by fine field are not available from any source. The U.S. Office of Education collects data on total undergraduate enrollments and we have used these to estimate - undergraduate enrollments in biomedical and behavioral fields as described in the footnote of Table 4.1. The resulting estimate approximates the number of students majoring in a field--it does not measure the number of course enrollments, which could be quite different. It is possible that in the behavioral fields, course enrollments (and hence the demand for teaching staff) have been increasing even though the number of students majoring in these fields has been decreasing. This possibility needs further study.
102 TABLE 4.3 Behavioral Scientists Employed in Colleges and Universities, 1961-81 Toters Ph.D.sb Non-Ph.D.s Fiscal Year N To N To N To 1961 1962 1963 1964 1965 1 3,700 n/a n/a n/a 15,691 100.0 100.0 5,339 n/a 8,143 n/a 1966 17,304C 100.0 9,783 56.5 7,521 43.5 1967 18,916 100.0 n/a n/a 1968 21,574C 100.0 12,915 59.9 8,659 40.1 1969 24,231 100.0 n/a n/a 1970 26,180C 100.0 16,175 61.8 10,005 38.2 1971 28,129 100.0 n/a n/a 1972 29,744c 100.0 18,602 62.5 11,142 37.5 1973 31,359 100.0 19,787 63.1 1 1,572 36.9 1974 32,980 100.0 22,324 67.7 10,656 32.3 1975 35,883 100.0 23,63 1 65.9 12,252 34.1 1976 38,121 100.0 n/a n/a 1977 39,240 100.0 25,588 65.2 1 3,652 34.8 1978 39,159 100.0 n/a n/a 1979 38,458C 100.0 26,894 69.9 1 1,564 30.1 1980 37,757 100.0 27,586C 73.1 l0,l7lc 26.9 1981 37,976 100.0 28,277 74.5 9,699 25.5 aIncludes psychologists and sociologists only. bIncludes psychologists, sociologists, and anthropologists. CInterpolated. SOURCES: NRC (1973-82), NSF (1965, 1978, 1982). There seems to be some correlation between graduate enrollment and academic employment of Ph.D.s in the behavioral sciences but it is difficult to interpret. Although graduate enrollment is only 10 percent of the total behavioral science enrollment, its growth pattern over the past 10 years has been roughly parallel to the growth in academic employment. It is possible that increasing graduate enrollments have created accompanying academic vacancies for people to teach them. Given the time lag in budgetary responses to teaching demands and the predominant role played by undergraduate enrollments in affecting the creation of teaching positions, this possibility seems, prLma facie, to be of low probability. Alternatively, it is possible that growth of faculty positions, determined by a mix of
103 other factors, creates a situation wherein more applicants for admission to graduate study are accepted and in this way the size of the academic staff determines the number of graduate students admitted. Neither the direction of causality nor even a conclusion of causality can be determined from the available information. In summary, it appears that the continued growth in academic employment of behavioral science Ph.D.s is at least partially due or related to four factors: (1) enrichment or upgrading, (2) somewhat greater reliance on part-time employment, (3) underestimated course enrollments, and (4) growth in graduate enrollment. There is good evidence that enrichment has been an important factor, and that part-time employment has increased {see Appendix Table C13~. The connection between graduate enrollment and academic demand is purely speculative and needs to be verified by further study, along with our estimate of undergraduate enrollments. The Mode! of Academic Demand for Behavioral Ph.D.s This brings us to the issue of how to make projections of academic demand for behavioral science Ph.D.s. For this report we will continue our past practice which has been simply to make independent assumptions about future patterns of enrollments and the F/S ratio. Then these projected values are combined to produce projections of academic employment levels, usually for 5 years ahead of the date of publication of the report. This procedure yields estimates of academic levels due to expansion (or contraction) of faculty. To this we must add demand created by attrition, which we estimate from recent actual experience. Recently, we have obtained more detailed information about the inflows and outflows in the academic labor market for behavioral scientists. The data from the 1979 and 1981 Surveys of Doctorate Recipients show that a fairly substantial amount of field switching out of behavioral fields has been occurring which must be considered in estimating attrition rates. There is, of course, also a substantial amount of switching into behavioral fields, and this will be accounted for when we consider how the total number of vacancies occurring each year are filled. To implement this projection procedure, therefore, we must formulate high, middle, and low assumptions about enrollment patterns and future F/S levels. For analysis purposes, the faculty/student ratio will be constructed as F/WS defined as follows: F = Ph.D.s employed by academic institutions in behavioral science fields (excluding postdoctoral trainees) WS = 4-yr weighted average of enrollments, i.e. WSt = 1/6 (St + 2St-1 + 2St-2 + St-3) where S = behavioral science undergraduate and graduate enrollments.
104 Assumptions The Committee's best guess is that total behavioral science enrollments will decline slightly from their 1980 level to about 625,000 by 1988. The upper and lower limits on this estimate are 700,000 and 550,000, respectively (Figure 4.6~. As of 1~ O--the latest year we can compute it--the F/WS value in the behavioral fields was 0.044 (Figure 4.7~. We expect it to stay at about this level for the next 5 years within a range of 0.035 to 0.055. coo 700 600 - ~n ~ 500 o 4 _ D As - o no 400 300 200 100 Actual --~~ Projected / Estimate (700 200. / ,~ - Middle Est. (625,000) All Schools / Public Schools`` Lo''~~~~~~~~- / Private School s O 1 ~ ~ 1 1 ' 1 ~ 1 1 1 1 1 1 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 FISCAL YEAR FIGURE 4.6 Total behavioral science undergraduate and graduate enrollments in colleges and uni- versities, by control of institution, 1960-80, with projections to 1988. See Appendix Table C1. Projections of Academic Demand to ~ 988 The combination of three estimates of enrollment growth together with three estimates of the 1988 F/WS level provides nine estimates of academic employment of behavioral science Ph.D.S in 1988 as shown in Table 4.4. Under the high assumptions (I-A of Table 4.4), enrollments would grow to 700,000 and the F/WS ratio would be 0.055, yielding an estimated faculty size of 37,700 behavioral science Ph.D.s in 1988 {Figure 4.8~. This would represent a growth of about 1,270 per year from the 1980 level of 27,600. Deaths and retirements would generate 330 vacancies, and other attrition from academic employment and field switching would add 1,240 vacancies, for a total annual demand of 2,840 positions.
105 0.06 O.0$ 0.04 lo ~ ~ . 03 3 0.02 0.01 Actual Projected A, <~0 ~5,5-~-~~ __--'~ Middle Est. (0.045) ~_______________________ (O. 03S; -' 1 1 1 1 1 1 1 1 62 64 66 68 70 72 74 76 78 80 82 84 86 88 FISCAL YEAR FIGURE 4.7 Behavioral science Ph.D. faculty/student ratio (F/WS), 1964-80, with pro- jections to 1988. Faculty (F) is all Ph.D.s employed in the behavioral sciences at colleges and universities. Students (WS) are defined as a 4-year weighted average of enrollments in the behavioral sciences: (WS)t = 1/6(St + 2St 1 + 2St 2 + St 3) where St = total behavioral graduate and undergraduate enrollments in year t. See Appendix Table C10. 40 35 _ `,, 30 In 4> LLJ A A: 15 _ cr: Lid m me 25 20 10 o ' _ 1 Actual -----a--- Projected Al 1 School s_ ~ ,Publ i c School s Private School ~ t --'` MlOUle tST. t`O,lUUJ 744?~;~` - 62 64 66 68 70 72 74 76 78 80 82 84 86 88 FI SCAL YEAR FIGURE 4.8 Ph.D.s employed in all behavioral science fields at colleges and universities, by control of institution, 1962-81, with projections to 1988. See Appendix Table C7.
106 TABLE 4.4 Projected Growth in Behavioral Science Ph.D. Faculty, 1980-88, Based on Projections of Enrollment and Faculty/Student Ratiosa Assumptions about the Faculty/Student Ratio for Behavioral Ph.D.s in Colleges and Universities (0.044 in 1980) I II III Assumptions about Behavioral Science Will continue to Increases slightly Declines to Undergraduate and Graduate Enrollment grow, reaching to 0.045 by 0.035 by (estimated at 627,300 in 1980) 0.55 by 1988 1988 1988 A. Will grow to 700,000 students Expected size of behavioral science faculty (F) in 1988 37,700 30,900 24,000 by 1988 Annual growth rate in F from 1980 to 1988 4.0% 1.4% -1.7% Average annual increment due to faculty expansion 1,270 410 - 50 Annual replacement needs due to:b death and retirement 330 290 260 other attrition 1,240 1,110 980 Expected number of academic positions to become available annually for behavioral Ph.D.s 2,840 1,810 790 B. Will remain at about the 1980 Expected size of behavioral science faculty (F) in 1988 34,400 28,100 21,900 level (625,000) through 1988 Annual growth rate in F from 1980 to 1988 2.8% 0.2% -2.8% Average annual increment due lo faculty expansion 850 70 - 710 Annual replacement needs due to:b death and retirement 310 280 250 other attrition 1,180 1,060 940 Expected number of academic positions to become available annually for behavioral Ph.D.s 2,340 1,410 480 C. Will decline to 550,000 students Expected size of behavioral science faculty (F) in 1988 31,000 25,400 19.700 by 1988 Annual growth rate in F from 1980 to 1988 1.5% - 1.0% - .1% Average annual increment due to faculty expansion 430 - 280 -980 Annual replacement needs due to:b death and retirement 290 260 240 other attrition 1,110 1,010 900 Expected number of academic positions to become available annually for behavioral Ph.D.s 1,830 990 160 aFaculty is defined in this table as all academically employed Ph.D.s, excluding postdoctoral appointees. The denominator of the faculty/student ratio is a weighted average of the last 4 years of enrollments: (WS)t = 1/6(St + 2St 1 + 2St 2 + St 3), where S = total graduate and undergraduate enrollments in the behavioral sciences. See Appendix Tables C 1 and C 10. bBased on an estimated replacement rate of 1.0% annually due to death and retirement, and 3.8% annually due to other attrition from academic positions in the behavioral sciences. These estimates were derived from the National Research Council (1973-82).
107 In the best-guess case (II-B of Table 4.4), enrollments would remain at about 625,000 and the F/WS ratio would be 0.045 by 1988. Academic employment would expand by 70 Ph.D.s per year, with another 1,340 added by attrition from all sources, for a total annual demand of 1,410. Under the lowest combination of assumptions (III-C of Table 4.4), enrollments would drop to 550,000 and the F/WS ratio would decline to 0.035 by 1988. There would be a drop of about 980 academic positions per year due to faculty contraction, but attrition would add about 1,140 vacancies per year, leaving a net annual demand of 160 positions to be filled. Estimating Postdoctoral Support Levels Under NRSA Programs The final step in our quantitative analysis of the market is to attempt to translate the projections of academic demand into recommended levels of postdoctoral training under NRSA programs, as shown in Table 4.5. This step requires certain additional assumptions about how the system has functioned in recent years with regard to postdoctoral training and its sources of support. TABLE 4.5 Estimated Number of Behavioral Science Postdoctoral Trainees Needed to Meet Expected Academic Demand Through 1988 Under Various Conditions Projected Through 1988 Annual High Middle Low Average Estimate Estimate Estimate 1979-81 1. Academic demand for behavioral science Ph.D.s- annual average: 2,840 1,410 160 1,919 a. due to expansion of faculty 1,270 70 - 980 691 b. due to death and retirementa 330 280 240 238 c. due to otherattritionb 1,240 1,060 900 990 2. Total accessions with postdoctoral research training- annual averages 710 350 40 364d 3. Size of behavioral science postdoctoral pool- annual average Size needed to meet academic demand assuming a 2-yr. training period and portion of trainees seeking academic positions is: 1,080 a. 60% 2,370 1,170 130 b. 7 0°70 2,030 1,000 110 Annual number of behavioral science postdoctoral trainees to be supported under NRSA programs: 349 a. if 40°,/0 of pool is supported under NRSA 810-950 400470 40-50 b. if 50~7O of pool is supported under NRSA 1,010-1,190 500-580 50-70 c. if 60% of pool is supported under NRSA 1,220-1,420 600-700 70-80 aAssumes an attrition rate due to death and retirement of 1.0% per year. bAssumes an attrition rate due to other causes of 3.8% per year. CAssumes that 25% of all vacancies, including those created by field switching, will be filled by individuals with postdoctoral research training in the behavioral sciences. The remaining 75% will be filled from nonacademic sectors (1 870), unemployed (2%), and new Ph.D. recipients (55%). dAssumes that 15% of the 1979-81 Ph.D. cohorts took a postdoctoral appointment before taking an academic position. This estimate is based on data from the National Research Council (1958-82, 1973-82). SOURCE: Table 4.4.
108 The features of the system which must be considered in addition to the projections of faculty growth are as follows: 1. the number of accessions to academic positions who have (or should have) research training 2 e the appropriate length of the postdoctoral research training period 3. the proportion of individuals in the research training pipeline who aspire to academic careers 4. the proportion of support of the total pool of behavioral science research trainees that should be provided by the federal government. With the aid of the data on inflows and outflows from academic employment of behavioral science Ph.D.s during 1979-81, shown in Table 4.6, we can make reasonable assumptions about these features--first presented in the Committee's 1981 Report--in order to provide a quantitative basis for the recommendations. Using the projections of academic demand derived in Table 4.4, we calculate in Table 4.5 the range of behavioral science postdoctoral trainees that should be supported by NRSA programs under the specified conditions. Line 1 of Table 4.5 is a summary of the projections of academic demand for the extreme cases and the best-guess estimate derived In Table 4.4. Line 2 shows the number of academic positions to be filled by individuals with postdoctoral research training experience assuming that 2S percent of all vacancies will be filled by former postdoctoral trainees. In the best-guess case, this number is estimated to be about 350. Line 3 indicates the size of the behavioral science postdoctoral pool required to supply the necessary number of individuals with postdoctoral training under certain assumptions about the length of the postdoctoral training period and the proportion of the pool seeking academic employment. If the appropriate length of postdoctoral training is 2 years, then the pool size needed to produce 350 trained scientists per year would be 700. If only 60 percent of the trainees seek academic appointments after completing their training, then the necessary pool size must be 1,170. Line 4 shows the estimated number of behavioral science post- doctoral trainees that should be supported annually by NRSA programs under different assumptions about the proportion of total support provided by that source. The resulting range is between 40 under the lowest set of assumptions, and 1,420 under the highest set. The best-guess assumptions yield a range of 400-700 postdoctoral trainees in the behavioral sciences. This range results from different judgments about the share of postdoctoral training support in the behavioral sciences that should be provided by the federal government through NRSA programs. Based on the contributions of these programs to the quality of postdoctoral training, the Committee believes that 50 percent is the appropriate federal share, yielding 500-580 NRSA awards per year.
109 TABLE 4.6 Inflows and Outflows from Academic Employment for Behavioral Science Ph.D.s, 1979-81 I. . Average A'Z!7Ual Attritio'' fro,'7 Acade'''ic k,'~plo~ '''e,?t i'' tare Bengal ioral Scie,~c`,s Total behavioral Ph.D.s employed in academia in 1979: 26,894 2. Leaving academic employment between 1979 and 1981 in the behavioral sciences to: N To of Academic Employment a. nonacademic employment 664 2.5 b. postdoctoral appointment 40 0.1 c. death and retirement 238 1.1 d. other ficldsa 184 0.7 e. unemployed 102 0.4 i'. total attrition 1,228 4.8 11. Average Annual Accessions to Academic Emplc'~,ment in tale BeI'a~'ioral Sciences 1. Total behavioral Ph.D.s employed in academia in 1981: 28.277 2. Entering academic employment between 1979 and 1981 in the behavioral seines from: N '%,ot'Tota1 Acecssio ns a. nonacademic sectors 352 18.3 b. postdoctoral appointments 160 8.3 it. unemployed 45 2. 3 d. Ph.D. recipients 1979-81 b 1.362 71.0 e. total annual accessions 1.919 100.() 111. Balancings: 1979 academic employmentattrition + accessions = 1981 academic employment 26,894 - 2( 1 ,228) + 2( 1 ,9 1 9) = 28,276 aThese individuals were ;~11 academically employed in 1979 and 1981. The number shown represents the net number switching t'rom behavioral to nonbeh;~viora1 t'ields. bit is estimated that 15','; fit these new Ph.D. cohorts took ~ postdoctoral appointment before taking an academic posit i`,n. tDoe~; not aYree with line 11.1 because ot rounding. SOU RUES: National Research Council ( 1 958-8 9. 197 3-82).
110 Summary In the above analysis, we have attempted to project short- term trends in academic employment of behavioral science Ph.D.s and to translate them into postdoctoral training levels needed to satisfy academic demand under certain specified conditions. The data on trends in academic employment are carefully collected and are probably quite reliable--especially since 1972--but they are difficult to interpret. The growth in academic employment in the behavioral fields during the 1970s has occurred despite an apparent drop in total behavioral enrollments and does not seem to be related to research expenditures. Enrichment--the process of enlarging the proportion of Ph.D.s in academic positions--has obviously been taking place and can account for much of the growth. Graduate enrollment in the behavioral fields has also continued to grow, possibly providing some stimulus for faculty expansion. The best-guess projection yields a modest increase in academic employment of behavioral science Ph.D.s by 1988. Converting the projections of academic demand into recommended postdoctoral training levels requires certain assumptions about how the system has operated in the past and how it can be expected to operate in the next few years. The actual data on dynamic movements into and out of the academic labor market during 1979-81 give new insight into the recent operation of this market. If we base the estimates on these recent data, we find that the level of postdoctoral training that should be provided under NRSA programs in the best-guess case is somewhat higher than current numbers actually being funded by the federal agencies. Demand Outside the Academic Sector Of particular interest to the Committee and its Panel on Behavioral Sciences is the steady increase in employment outside the academic sector. During the 1973-81 span the number of Ph.D. scientists in behavioral fields who were employed in nonacademic settings grew at an average annual rate of more than 10 percent, while the number employed by academic institutions rose at a rate of only 4 percent (see Appendix Table Cod. Within the nonclinical areas in the behavioral sciences, expansion outside the academic sector has been especially great in the last 2 years. Between 1979 and 1981 the numbers of nonclinical scientists holding positions in business, government, and other nonacademic settings grew at a yearly rate of nearly 10 percent. The implications of this trend with regard to need for federal training support in the behavioral sciences rest, in part, on whether or not the movement toward nonacademic employment represents an appreciable decline in the fraction of behavioral scientists involved in research. To address this issue a detailed analysis has been made of 1973-81 trends in the employment situations of behavioral science Ph.D. recipients entering the labor force. It is presumed that these new entrants would have been most vulnerable to significant changes in the availability of career opportunities.
111 Findings from this analysis are summarized in Table 4.7. As can be seen from the data presented, there has been an appreciable decline in recent years in the percentages of behavioral science Ph.D. recipients taking faculty positions. This decline has been particularly noticeable in the nonclinical areas of the behavioral sciences. In these disciplines the fraction of new entrants employed in faculty positions dropped from approximately two-thirds in 1973 to only one-third by 1% 1. This drop has been offset, in part, by increases in the percentages taking postdoctorate and other nonfaculty staff positions in universities and colleges. By 1% l more than one- fourth of the new Ph.D. cohort who had entered from the nonclinical fields held academic positions not considered to be faculty appoint- ments. During this 8-year span there also have been substantial increases in the numbers of both clinical and nonclinical Ph.D. recipients who took positions outside the academic sector--and many of those in the nonclinical fields were engaged in research. Thus, although the findings indicate that an appreciably smaller fraction of behavioral science grade tes have moved on to university faculty appointments, there has been a modest increase in the fraction involved in research outside the academic setting. Although the unemployment rates for behavioral science Ph.D.s have never exceeded 4 percent, it may be noted that the rates for Ph.D. recipients in nonclinical fields have been generally higher than the rates for clinical gradm test These findings are consistent with the committee's impression that the employment prospects for those trained in clinical areas have been much better than the prospects for other behavioral science graduates. Further evidence of this may be discerned from an examination of the field-switching patterns of recent graduates. As shown in column (2) of Table 4.8, the percentage of Ph.D. recipients in nonclinical areas of the behavioral sciences who left these fields within 2 years after receipt of their doctorates has increased from 22 percent for the FY 1971-72 cohort to as much as 37 percent for the FY 1979-80 cohort. During this 8-year span the percentage leaving the clinical fields has never exceeded 14 percent. A detailed analysis (not presented here) of field-switching patterns reveals that more than half of the recent graduates leaving the nonclinical areas of the behavioral sciences were either psycholo- gists moving into clinical specialties and the life sciences or anthro- pologists and sociologists switching into other areas of the social sciences . One useful barometer of the changing demand for Ph.D. recipients in a field is the ratio of the number of recent graduates employed in that field to the number trained in the field. As shown in column (5) of Table 4.8, this ratio has dropped significantly during the past decade in the nonclinical fields. By the late 1970s the number of new graduates employed in these fields represented only about two-thirds of the number of individuals who had received their doctorates in the nonclinical fields. In contrast, throughout this period there have been significantly more young graduates working in the clinical fields of the behavioral sciences than had been trained in these fields. In view of these findings it is not surprising to find that an increasing- ly larger fraction of the graduate students enrolled in the behavioral sciences have earned their doctorates in the clinical fields. _ _ _
112 TABLE 4.7 Employment Situations of Behavioral Ph.D. Recipients Who Had Received Their Doctorates Within the Previous 2 Years, 1973-81 - Fiscal Year 1973 1975 1977 1979 1981 Employment Situation N ~ N ~ N 56 N ~ N . . .. PH.D.S IN CLINICAL FIELDS Total Entrants 1,793 100.0 2,051 100.0 2,427 100.0 2,621 100.0 2,936 100.0 Academic Sector 547 30.5 604 29.4 752 31.0 585 22.3 750 25.5 Faculty Positions 441 24.6 477 23.3 590 24.3 229 8.7 519 17.7 Postdoctorals 9 0.5 26 1.3 15 0.6 24 0.9 12 0.4 Other Staffa 97 5.4 101 4.9 147 6.1 332 12.7 219 7.5 Nonacademic Sectors 1,211 67.5 1,437 70.1 1,634 67.3 1,956 74.6 2,186 74.5 Research Positionsb n/a n/a 268 13.1 237 9.8 217 8.3 155 5.3 Other Staff Positions n/a n/a 1,169 57.0 1,397 57.6 1,739 66.3 2,031 69.2 Unemployed and Seeking Position 35 2.0 10 0.5 41 1.7 80 3.1 0 0.0 PH.D.S IN NONCLINICAL FIELDS Total Entrants 3,569 100.0 4,236 100.0 4,479 100.0 4,786 100.0 4,627 100.0 Academic Sector 2,727 76.4 3,059 72.2 3,004 67.1 3,212 67.1 2,757 59.6 Faculty Positions 2,387 66.9 2,425 57.2 2,146 47.9 2,212 46.2 1,578 34.1 Postdoctorals 97 2.7 284 6.7 286 6.4 385 8.0 433 9.4 Other Staffa 243 6.8 350 8.3 572 12.8 615 12.8 746 16.1 Nonacademic Sectors 786 22.0 1,106 26.1 1,294 28.9 1,408 29.4 1,761 38.1 Research Positionsb n/a n/a 702 16.6 695 15.5 519 10.8 976 21.1 Other Staff Positions n/a n/a 404 9.5 599 13.4 889 18.6 785 17.0 Unemployed and Seeking Position 56 1.6 71 1.7 181 4.0 166 3.5 109 2.4 aIncludes individuals employed in academic positions that were not considered faculty (tenure track) or postdoctoral appointments. bIncludes individuals employed in nonacademic sectors who indicated that they devoted at least one-fifth of their time to R and D activities. SOURCE: National Research Council (1973-82).
113 TABLE 4.8 Field Switching into and out of Clinical and Nonclinical Areas of the Behavioral Sciences by FY 1971-80 Ph.D. Recipientsa (1 ) (2) (3) (4) (5) Total Ph.D.s Leaving Switching Total Empl. Ratio in Field Field into Field in Field (4)/(1) Ph.D. Cohort N N To N N FY 1 9 71- 72 Clinical Fields 1,758 23n 13.4 333 1,855 1.06 Nonclinical Fields 3,513 791 22.5 309 3,031 0.86 FY 1973-74 Clinical Fields 2,041 202 9.9 547 2,386 1.17 Nonclinical Fields 4,165 1,037 24.9 269 3,397 0.82 FY1975-76 Clinical Fields 2,386 261 10.9 617 2,742 1.15 Nonclinical Fields 4,298 i426 33.2 298 3,170 0.74 FY1977-78 Clinical Fields 2,541 350 13.8 721 2,912 1.15 Nonclinical Fields 4,620 1,782 38.6 323 3,161 0.68 FY 1979-80 Clinical Fields 2,936 ~ 1899' 6.4 472 ~ 3,il9` 1.10 Nonclinical Fields 4,518 1,678 37.1 191 3,031 0.67 aData reflect field of employment within 2 years of the time an individual received the doctorate. SOURCE: National Research Council (1973-82). Long-Term Considerations In considering the implications of market trends in the behavioral sciences, we must note that there is, on average, a 7-year lag from the time a student enters graduate school to the time he or she completes requirements for the doctorate. Consequently, undergraduate students now contemplating research careers in the behavioral sciences are not likely to be looking for jobs until the early l990s. Although we are reasonably confident about our short-term projections of academic demand, the long-term outlook is much more difficult to quantify. But certain trends are developing now which, if continued to the l990s, could have great impact on the market for behavioral scientists. During the past decade, for instance, we have witnessed an expansionary trend that few could have predicted in the employment of behavioral scientists outside the academic sector. Much of this growth has been in the non-research areas of the behavioral sciences and, as such, may not affect the demand for research personnel. The
114 factors underlying this trend are not fully known, and it is difficult to foresee whether this trend will continue during the next 15 years. Far more predictable, on the other hand, are the numbers of individuals who will leave the Ph.D. labor force during this period. Estimates of attrition due to deaths and retirements based on the age distribution of the FY 1981 labor force are illustrated In Figure 4.9 .3 Between 1982 and 2001 the number of clinical and nonclinical 1, 500 1_ z _4 <a At: to I_ s m Lo to 1 ,000 CO 500 an - f a i/ / / / / / f ~ _ ~ V J may of J 1 l I ~ I I I I ~ ~ I 1 1 1 1983 85 87 89 Noncl i ni Cal C] inical 91 93 95 97 99 2001 FISCAL YEAR FIGURE 4.9 Annual number of behavioral scientists in the FY 1981 Ph.D. labor force expected to reach the age of 65 during the FY 1983-2001 penod. See Appendix Table C14. 3 The attrition rate used in the short-term projections through 1988 was based on a fixed percentage of the academic labor force. Long-term attrition estimates must consider the age distribution of the total Ph.D. labor force of behavioral scientists.
115 behavioral scientists who will reach the age of 65 Is expected to increase appreciably. By the end of this period the estimated annual attrition will be nearly three times the 1983 level. Between 1992 and 2001, a total of 11,000 behavioral Ph.D.s are likely to be lost through attrition due to death and retirement, compared with 6,900 in the preceding 10-year span. This substantial rise in attrition is likely to occur during a period when a diminishing number of individuals will be receiving doctorates in the behavioral sciences. Annual Ph.D. production in the behavioral sciences is projected to decline from 4,469 in FY 1981 to approximately 3,700 awards by FY 1988 (Figure 4.103. In the nonclinical areas, a drop in doctoral awards of 9 ,ooo 8,000 7,000 6,000 cow LL m Actual Prod ected 8, 017 Fi rst-Year, Ful 1~ a__ Graduate Enroll Tents in Ph.D.- Granting Departments in the Behavioral 6 68 Sciences (advanced seven years) , 1 5,000 4,000 3, 000 2,000 1 ,000 o7 Ph . D . Awa rds i n the Beha vi o ral Sci ences Ph.D. Awards in Noncl inical Subfields 3,726, 2.012` ~;,7~40 1,714' Ph.D. Awards in Cl inical Subfields 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 F. I SCAL YEAR FIGURE 4-10 Comparison of first-year, full-time graduate enrollments and Ph.D. awards in the behavioral sciences, 1973-88. See Appendix Tables C2 and C4.
116 more than 25 percent is expected, while in the clinical areas the projected decline is very slights In making projections for the subfields it has further been assumed, on the basis of recent trends, that an increasingly larger fraction of the behavioral science Ph.D.s will be awarded in clinical areas. During the last 10 years the percentage of behavioral science doctorates in clinical fields has steadily increased from 30 percent to its current level of 40 percent. These projections are tentative and should be interpreted cautiously. Whether an overall decline of 18 percent in Ph.D. awards in the behavioral sciences occurs will depend on a variety of factors such as the availability of financial support for graduate study, the average time it takes students to finish their doctoral programs, and other factors affecting the Completion rate" for first-year graduate students in the behavioral sciences. If, for example, past trends continue and a decreasing fraction of the graduate students currently enrolled complete work for the Ph.D., the decline in doctoral awards may be even greater than projected. Furthermore, it should be recognized that even if the annual number of doctoral graduates is greatly reduced, this number will still exceed the total labor force attrition (due to death and retirement) which is expected to average fewer than 700 scientists per year between now and 1988. Thus, continued expansion in the behavioral science labor force may be anticipated, although at a rate somewhat below that experienced during the past decade. But if recent trends continue, Ph.D. awards in the nonclinical specialties may be expected to drop far more sharply than doctoral awards in the clinical specialties. In fact, it is highly probable that by the mid-199Os more individuals will earn Ph.D. degrees in the clinical fields than in the other behavioral sciences. This prospect is of particular concern since the largest share of the research in the behavioral sciences is conducted by individuals with training in the nonclinical areas. On the basis of these long-term considerations, the Committee concludes that there is a continuing need for federal training support at the predoctoral level in the behavioral sciences. 4 It has been assumed in these projections that there is a 7-year lag between the time a student initially enrolls in grade te school and the date of completion of the doctoral program. It has also been assumed that approximately 56 percent of the students enrolling will successfully complete their doctoral programs. The latter assumption is based on the ratio of Ph.D. awards in FY 1~ 1 and FY 1% 2 to first-year full-time graduate enrollments 7 years earlier.
