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 Abstract The academic labor market for behavioral science Ph.D.s continues to expand at a moderate rate of growth. Academic employment of Ph.D.s and post- doctoral training increased in 1983 and are higher than they were in the mid-1970s. Most of the growth has occurred in the field of clinical psychology. Behavioral Ph.D. production rebounded in 1983 from the decline in 1982, but the number of new doctorate recipients is only a slight fraction above the level recorded in the late 1970s. Concurrent with the continued expansion of the academic labor market are continuing declines in R and D funding, undergraduate enrollments, and first-year graduate enrollments. Recently, total graduate enrollments in the behavioral sciences have begun to fall. However, there appears to be a substantial amount of behavioral science courses being taught to graduate students at professional schools--public health, law, medicine, and business--which tends to increase the demand for behavioral scientists. The shift toward clinical psychology that began in the mid-1970s is continuing. In 1977, clinical psychology became the dominant behavioral science Ph.D. employment field and now accounts for about 45 percent of the behavioral Ph.D. labor force, up from 38 percent in 1975. INTRODUCTION AND OVERVIEW Within the behavioral science fields, it is evident that a strong movement into clinical psychology has occurred in recent years. There has been an increase, both absolute and relative to other areas, in the number of Ph.D.s who identify themselves as clinical psychologists and in the number of new Ph.D. degrees being granted in that field. From the standpoint of the federal government's research program in the behavioral sciences, this trend is cause for some concern since most clinical psychologists work outside the academic sector and many do not contribute to the research effort. Even within the academic sector, clinical psychologists are more likely than other psychologists to be involved in the provision of services and other 79
80 non-research activities. One of the principal purposes of the government's research training programs is to provide an adequate supply of scientists with current knowledge of their fields who can be relied on to carry out the agencies' research agenda competently and productively. Practically all of this research is done by Ph.D. scientists in colleges and universities, so we are especially concerned in this study about the career prospects and opportunities in the academic sector. Academic employment of behavioral science Ph.D.s has increased without interruption since the early 1960s (which is as far back as our data go). However, this growth has not been uniform throughout the behavioral sciences. Academic positions in nonclinical fields of psychology have decreased since 1981. Since the bulk of psychological research has been carried out in the academic sector by nonclinical psychologists, this trend is of some possible significance for future research. In this chapter we will examine these trends more closely and make some projections of academic demand and training needs through:1990. As a result of suggestions made at public hearings and in private communications, the data on behavioral scientists are disaggregated In this report to a greater degree than has been done in the past. In the early work of this committee, the behavioral sciences--psychology, sociology, anthropology, and speech and-hearing sciences--were treated as a single group. In 1978 the behavioral sciences were divided into clinical and nonclinical fields. This disaggregation proved helpful because it enabled the identification of divergent market trends within the behavioral sciences. In the current report we carry the disaggregation one step further and divide the nonclinical fields into nonclinical psychology and other behavioral sciences. This yields three behavioral science subdivisions: --clinical psychology; --nonclinical psychology; and --other behavioral science fields (sociology, anthropology, and speech pathology/audiologyj.~ CURRENT SUPPLY/DEMAND INDICATORS With this additional level of disaggregation, substantial differences among disciplines and education levels {graduate and undergraduate) begin to emerge. For example, undergraduate enrollments in psychology have not been subject to the decline experienced by the other behavioral fields. Here are some other highlights: iThe nomenclature used in the NRC surveys was changed in 1983. Speech pathology/audiology replaced speech and hearing sciences. This field is more clinically oriented than either sociology or anthropology but it is not part of clinical psychology, and is too small (113 Ph.D.s awarded in 1983) to be considered separately.
81 Graduate enrollments in the behavioral sciences have been less susceptible to attrition than have undergraduate enrollments. Enrollments at the graduate level rose every year from 1978 through 1981 and only in 1983 did enrollments fall by more than a fraction of a percentage point. In 1983 graduate enrollments in psychology declined less than those in other behavioral science fields. R and D funding for behavioral science research decreased in 1983. Research funding for psychology did not decline, however. It remained at its 1982 level which (after adjusting for inflation) was about equal to the 1975 funding levels. The decline in R and D funding occurred in behavioral fields outside of psychology. Employment in the behavioral sciences continued to grow along the same lines observed in earlier reports. The self-employment and business sectors were the most rapidly expanding areas. These patterns were general across all C! 1 phi; `, i c i mn c: of Ph ~ huh AN; or a 1 .~ i annex . Academic employment . continued to grow in 1983, but the increases were . concentrated in the fields outside of psychology: sociology, anthropology, and speech pathology/audiology. The number of behavioral scientists on postdoctoral appointments rose in 1983. Clinical psychology had the greatest increase followed by large increases for the combined category of sociology, anthropology and speech pathology/audiology. Nonclinical psychology had fewer postdoctoral appointments in 1983 than in 1981. The number of nonclinical psychologists with postdoctoral appointments fell to its lowest level since the committee began monitoring these data. While some of these shifts may be due to sampling fluctuations, 2 the change could signal an imporant decline in research potential. Clinical psychologists on postdoctoral appointments are often being trained in clinical skills; nonclinical psychologists on postdoctoral appointments are more likely to augment the pool of researchers in the behavioral sciences. In the following sections the supply and demand outlook for behavioral scientists is examined in greater detail. 2The data on the PheDe labor force came from a 16 percent sample of the doctorate population. See NRC, 1985 for a discussion of sampling error.
82 Enrolknents(Tables4.~-4.3 and Figured) Total enrollments in the behavioral sciences (graduate and undergraduate) declined steadily from 1976 through 1980 (Table 4.1, line 4a). 3 In 1981, the downward trend was halted and behavioral science enrollments rose for the first time since the early 1970s. The long-term decline in enrollments is due exclusively to trends for undergraduate majors. Graduate enrollments in the behavioral sciences increased during the late 1970s and have remained stable (near 64,000) since 1978. Only in 1983 was there a non-tr ivial decline in behavioral science graduate enrollments. 800 700 600 500 400 300 Z 200 100 o I I I I 1 1 64 66 68 70 72 74 FISCAL YEAR / Total Behavioral Psychol onyx Other Behavi oral ~ ~` 76 78 80 82 FIGURE 4.1 Behavioral science undergraduate and graduate enrollments in colleges and universities, 1964-81. See Appendix Tables C1, C4-C8. 3The enrollment data in Tables 4.1 through 4.3 come from two sources. Undergraduate enrollments are estimated from U.S. Department of Education Higher Education General Information Surveys (HEGIS). Graduate enrollments are from National Science Foundation Surveys of Graduate Students and Postdoctorals.
83 P" ·C~ - .= o so U. o en ._ I ._ V _ _ 4.} ~ TIC C C ° ~ C' U) 3 E ~ oo o~ _ oo _ oo o' _ o°` _ _ o~o _ _ _ _ U) ~ _ ~ o · . . · · . ~ _ ~ ~ ~ U) 1 1 _ o _ · · . _ ~ U) U) 1 oo ~ ~ 2 ~ o · ~ oo °e ~ ~ X _ U) ~ · ~ ~o ~ _ ~o ~ o~ ~ . ~ U) _ ~o ~ · ~ U) _ _ X °°\ ~ C ~ _ o ~ · ~ U) ~ _ X o o o~ ~ ~ ~o ~ · ~ _ ~ -c ~= 5 _ C Z ~ o ~E ~ ~ ~'-~-'C ;:~) ~ ~ ' ' h ~ ~ ~ . ~ C a U) ~ ~ _ 00 ~ 0' ~ ~ . . . . . . . . . _ _ 1 _ _ _ O ~ 0\ ~ X ~ _ _ ~ ~ . . . . . . . . . · . ~ U) U) ~ _ ~ ~ ~0 ~ \0 _ 1 - _ _ _ ~o 0 ~ ~ u~ _ _ x ~ O 00 ~ ~ ~ ~ _ U) ~ ~ 2 2c 2 2 2 c ~ _ e~ · ~ C o - ~o o _ - _ o _ 2 c 2 ~ c 2 2c 2 ~ - - .' o~ o o . _ o o - X ~o · - , ~ o o . _ o _ X ~ ~ ~ X o~ 0 x u~ ~c 0 u; et ~ X ~ ~ _ _ o~ ~o _ ~ ~ ~ U) c 2 2c 2 ~ 2 2 2 ~ ~ ~ _ U) ~ U) ~ ~ X oo ~ ~ ~ X ~ _ U) U) ~ ~ U) U) _ ~ U) _ 2 2cc 2 2c 2 2c 2 r~ ~ _ x ~ o~ o ~ ~ ~o ~ ~ X ~ ~ ~ ~ ~ _ ~ X X ~ _ ~ ~ _ ~ _ ·. Vl _ V~ ~ · ~o _ 6 ~ X ~ V~ v, O e' _ C ~ · C~ o,c ~ ~ ~ o _ ,~ Y ~ ~ A _ e _ E _ E E E, E E c - s E A O ~ ~ ~ = e ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~o ~ ~ OD C · 0 1 ._ 8 + ~ o ~- o ~ $ + o~ ) ° ~ 6 E 3 _ E =, u~ ~ _ _ 0 ~ 0 ~ 8 g ~ ~ . x ~ ~ ~ ~ _ ~ r~ ~ ~ 0 r~ ~ ._ os ~ ~ ._ ~ x ~ ~ ~ ~ ~ ~ ~ ~ ~ '6 .u~, ~- ~ s~ ° O ~ ox 5_ e °` +-~° X ~ S _ = ~° ·- g 0 _ 0 ,e 0 e ~ ' _ _ ~ ~ _ 6~ ,, ~' ~ e 0 o~ ~ ~ 0 ~:ai ~ a e ._ i~ !~-~^ ~ ~ ~ v ~ ~ ~ ~ ~ ~ u)~E, ~ ° tc e 0 g° x x s {,, s e 11 ~ '~ ~ ~ u~ ~ ~ s °t ~ j- _ ~ ~ ~ ~ ~ ~ e ~ 4,l 5 ·£ s O ~; o 11 ~ C ! 5 u ~ ~ 9 ~ E E ~ ~ s ~ ' E O ~ -5 E I E q ~ ~ ~ ~q ~ ~ ~ ~o en ~c c ~ _ ·. ~ iE; O ~ ~ _ ~ _ as 0 ° 0 Z ~5 ~ ~ 0\ 0 · . · ~ O O 1 1 X ~ `0 ~ · . . . _ ~ _ ~ 1 1 1 ~ C~ X O 2 2 ° ~ ~ o~ C' ~ _ ~o 2 2 ~ ~ et 0\ ~o - C~ ~o ~ oc L~ en o~ en C 0` L,. . C V o - - C~ + - U) + - U) + - 11 - - - oo 1 - o~ - ~r x l ~o o~ - - - - r 0\ U) 0\ - - X - - X ~r - - C o - as C~ o C - . C~ . - - - U) X U) 0\ - e' 00 1 - - U: z - 1 - V) l 00 U) - - z .. U: - o U,
84 TABLE 4.2 Current Trends in Supply/Demand Indicators for Clinical and Nonclinical Psychology Ph.D.s Fiscal Year Growth Rate Latest from 1975 to Annual 1975 1976 1977 1978 1979 1980 1981 1982 1983 Latest Year Change 1. SUPPLY INDICATORS (New Entrants): Nonclinical Psychology: a. Ph.D. production 1,607 1,590 1,637 1,591 1,582 1,517 1,615 1,477 1,545 - 0.5% 4.6% b. % of Ph.D.s without specific employment prospects at graduation 13.1% 16.3% 15.5% 16.3% 14.6% 13.1% 12.2% 15.6% 16.3% 2.8% 4.5% c. Postdoctoral appointments 398 n/a 394 n/a 527 n/a 511 n/a 302 - 4.4% - 23.1% Clinical Psychology: d. Ph.D. production 1,144 1,293 1,353 1,464 1,509 1,581 1,743 1,681 1,762 5.5% 4.8% e. % of Ph.D.s without specific employment prospects at graduation 14.7% 14.8% 16.0% 17.6% 16.3% 14.6% 14.8% 15.7% 16.9% 1.8% 7.6% f. Postdoctoral appointments 156 n/a 357 n/a 302 n/a 262 n/a 466 14.7% 33.4% 2. DEMAND INDICATOR: a. Psychology R&D at colleges and universities (1972 $, milt) 63.1 58.2 60.1 59.0 60.7 62.3 65.9 63.7 63.6 0.1% -0.2% 3. LABOR FORCED Ph.D.s employed in nonclinical psychology fields: a. Total 15,387 n/a 16,102 n/a 16,688 n/a 18,791 n/a 19,431 3.0% 1.7% b. Academic (excl. postdocs.) 10,863 n/a 10,905 n/a 11,538 n/a 12,586 n/a 12,404 1.7% - 0.7% c. Business 1,218 n/a 1,344 n/a 1,355 n/a 1,827 n/a 2,258 8.0% 11.2% d. Governments 1,170 n/a 1,404 n/a 1,164 n/a 1,235 n/a 1,320 1.5% 3.4% e. Hospitals/clinics 470 n/a 447 n/a 401 n/a 905 n/a 1,328 13.9% 21. 1% f. Nonprofit 560 n/a 519 n/a 574 n/a 507 n/a 629 1.5% 11.4% g. Self-employed 401 n/a 443 nla 321 n/a 631 n/a 451 1.5% - 15.5% h. Other (incl. postdocs.) 527 n/a 649 n/a 961 n/a 806 n/a 635 2.4% - 11.2% i. Unemployed and seeking 178 n/a 391 n/a 374 n/a 294 n/a 406 10.9% 17.5% Ph.D.s em~ployed in clinical psychology fields: j. Total 14,846 n/a 17,578 n/a 21,268 n/a 23,775 n/a 26,285 7.4% 5.1% k. Academic (excl. postdocs.) 5,140 n/a 5,438 n/a 5,790 n/a 6,172 n/a 6,370 2.7% 1.6% 1. Business 165 n/a 409 n/a 417 n/a 880 n/a 1,004 25.3% 6.8% m. Government. 1,252 n/a 1,216 n/a 1,671 n/a 1,653 n/a 1,854 5.0% 5.9% n. Hospitals/clinics 4,425 n/a 5,102 n/a 5,702 n/a 5,937 n/a 5,737 3.3% - 1.7% o. Nonprofit 363 n/a 662 n/a 1,093 n/a 1,032 n/a 1,165 15.7% 6.2% p. Self-employed 2,292 n/a 3,201 n/a 4,785 n/a 6,264 n/a 7,999 16.9% 13.0% q. Other (incl. postdocs.) 1,151 n/a 1,468 n/a 1,674 n/a 1,629 n/a 1,995 7.1% 10.7% r. Unemployed and seeking 58 n/a 82 n/a 136 n/a 208 nta 161 13.6% - 12.0% 4. PSYCHOLOGY ENROLLMENTS: a. Total undergraduate and graduate 425,000 434,000 419,000 422,000 408,000 415,000 426,000 n/a n/a 0.04% 2.7% b. Est. undergraduates 392,000 399,000 384,000 383,000 369,000 375,000 385,000 n/a n/a 0.3% 2.7% c. Total graduate 32,794 35,318 35,363 38,628 39,207 39,786 40,636 40,691 40,098 2.5% - 1.5% d. Est. nonclinical graduate 17,954 18,393 18,099 18,915 18,856 18,608 18,985 19,039 18,761 0.6% -1.5% e. Est. clinical graduate 14,840 16,925 17,264 19,713 20,351 21,178 21,651 21,652 21,337 4.6% - 1.5% a Since labor force data are not available for 1982, latest annual change represents average annual growth rate from 1981-83. b Also includes FFRDC laboratories. c Estimate~d by the formula U; = (Aj+2/Bj+21C;, where U; = psychology undergraduate enrollment in year i; Ai+2 = psychology B.A. degrees awarded in year i+2; Bi+2 = total B.A. degrees awarded in year i + 2; C; = total undergraduate enrollment in year i. The FY 1981 figure is a preliminary estimate. SOURCES: NRC (1958-85, 1973-84); NSF (1973-85a, 1975-85); U.S. Department of Education (1948-81, 1948-84, 1959-79, 1961-84a, 1961-84b, 1973-82, 197~83).
85 - o o o ;^ o o o Q - U, q en ·C) - ·~ o em IS lo o U. o en · ~ e rot U. . CQ I-, o ~ O V O o As, r.1 _ ~ ~ Cut ~ ~ ~1 {. ° a. ~ a: U) sat A: Go _ o' _ so _ 0 so o' C' Cat ._ _ _ V) _ Go U) ~ ~ · · 0 _ ~ ~ . 1 ~ o to ~ ~ _ · · oo ~ _ 1 i` . ~- u) 1 u) r~ mo ~ ~ ~ - · - - ~ - ~ ) (~ ~ t - - ~ oo - · - ~ - - ~ o ~ u) - . ~ - - ~ ) - · ~ - u - ~ ~ . oo o u) Yo w u) o' - · - - u - ~ ~ o~ ~ ~ ~ - · - ~ ~ - ~ ~ ~ ~ ~ o~ ~ o~ ~ ~ r~ ~ oo ~ ~ · . . . . · . . . ) ~ l ~ ~ ~ ~ ' - 1 ~ o~ ~ ~ ~ ~ ~ ~ o~ u) · · · l u) ~ v) ~ 2~2~ r~ u, 0 U) ~o ~ U) oo _ _ es oo ~ ~ ~ ~ ~ ~ o~ U) U) V) _ ~ ~ _ _ ~ _ e5 ~ :5 e~ a~ ~ e5 es 2 2 2 ~ 2 2 2 2 ~ V] ~ ~ o _ ~ ~o _ o ~ ~ _ ~ o oo - 22222~2 ~ b~O ~ O ~ _ _ O _ 00 U) U) oo C,- V) - , · . o ~ ~ _ _ ~ _ ~ ~ U) ~o ~ ~ _ _ _ ~ ~ ~ ~ _ ~ e. l: ~ C) Z ~ 8 ~ C~ _ . ~ E" ~o .~ ~ ~ ~ Z ~ ~ ~ ~ ~ V) ~ ~ C~ - o~ o~ · · 1 1 1 ~ U) ~ · . ~ ~ U) o 1 1 ~2 - o o 2 2 ~ CJ _ ·C~ ._ _ o ·> ~o ~ s _ O X ~ g ~ ^= . a 0 E E 'a E S e S o = 3 ~ ~ ~ ~ a Y ~ a~ ~ GL, c~ ~ ^ c.~ ~ ~ c_ Oti,= ~ · z ~ ~ OD o,,~, a 0 ~ ~ ~ O O Z eQ L. a: e~ ~0 · 4) .E . ._ ~ ~ _ ;-, t a5 o3 ~ ~ . s o~ ' a5 o~ o 8 ~ ,, o o ~ ~ +~;: ~ · ~ _ :t ~ o~ E ;-, ~ ~ ~ X ~ ~ . e' :5 ~ · o~ ~ ~ . - o- o u) - o ^3 = - ~ ~ 0 e .e o o ~ ~ ~ E 8 8 _ c ~ o ° . C C~ e¢ C , ~ C~ o C ~ _ c .~ 0` e ~ ~, ~ ~. O 0 oo ~ ._ _ ~ g g ~ ~a ~ 0 ~ ~ _ _ _ ~ _ - ~ ~ ~ _ , ~ cq ~ ~ ~ ~ ~ 11 j g ° ~ e ° ~ Ui ~ ~ ~ ~ :- ~ ~ _ ~V ~ - ~ ~ ;^ ~ ~ _.C ~ X 0\ ^ ~ ~ _ ~ ~ _ O + ~ >m Z ::.,+4V ,' 5~= ~ ~ ~ o ° 2 ' 2 + ·a ~ ~- ~ O q ~ ~ .5 ~
86 Psychology Enrollments Most of the decrease in behavioral science enrollments is found in disciplines other than psychology. Psychology enrollments have been very stable since 1975 while enrollments in the other behavioral sciences have declined (Figure 4.1~. Total psychology enrollment (graduate and undergraduate combined) was approximately 425,000 in 1975 and approximately 426,000 in 1981 (Table 4.2, line 4a). Undergraduate psychology enrollments declined in 1977, 1978, and 1979 and rose again in 1980 and 1981. At the same time that undergraduate enrollments were declining, graduate enrollments in psychology rose. The number of psychology graduate students grew rapidly from 1975 through 1978 and smaller gains in enrollments continued through 1982. Only in 1983 did the graduate enrollments in psychology decline. almost all of the growth in psychology graduate enrollments was due to students in clinical specialties. The clinical fields grew each year from 1975 through 1982. During this period the average growth rate for clinical psychology enrollments was 5.6 percent (Table 4.2, line 4e). In more recent years the clinical growth rate has slowed. Nonclinical psychology graduate enrollments have been extremely stable, rising from 17,954 in 1975 to 19,039 in 1982 (Table 4.2, line 4d). In 1983 both clinical and nonclinical psychology graduate enrollments fell by 1.5 percent. Other Behavioral Science Enrollments Total graduate and undergraduate enrollments in behavioral science disciplines other than psychology have declined steadily since 1975. The rate of decline in recent years is only slightly smaller than the average decline over the entire period (Table 4.3, line 4a). Most of the decline is due to trends in undergraduate enrollments which fell from 276,000 in 1975 to 197~000 in 1981. Graduate enrollments in these behavioral science fields rose at the end of the 1970s and have begun to fall in 1982 and 1983. Over the entire 1975 to 1983 period, the average growth rate for other behavioral science graduate enrollments has been 0~2 percent (Table 4.3, line 4c). In summary, the trends in behavioral science enrollments vary by field and by educational level. Undergraduate enrollments have fallen. This decline is less pronounced in psychology and more concentrated in the other behavioral science fields. Graduate enrollments increased in the late 1970s and have not declined appreciably in recent years. Most of the expansion in graduate enrollments has been due to the growing number of graduate students in clinical psychology.
87 Reliability of Behavioral Science Undergraduate Enrollment Estimates In the preceding discussion, the estimates of undergraduate enroll- ments were based on earned baccalaureate degrees and thus tend to measure the number of students majoring in behavioral sciences. Some observers have suggested that the role of the behavioral sciences in undergraduate education may be underestimated by relying on number of majors rather than on number of course enrollments. The argument is based on the view that behavioral science courses are frequently taken as requirements or electives by non-majors. Thus, the proportion of students in behavioral science courses may exceed the proportion of behavioral science majors and it is possible that the committee's estimates of enrollments in these fields underestimate the true teaching loads. We have undertaken an examination of that possibility. Course enrollment data are not collected in a systematic and comprehensive manner by federal agencies. However, a 1982 survey by the American Council on Education (ACE) collected course enrollment data in science, engineering, and the humanities for the fall of 1980 (Atelsek and Anderson, 1982~. The ACE data were obtained from a stratified sample of 698 institutions, of which 498 (71 percent) provided usable responses. The hypothesis that the number of behavioral science majors is not a good estimator of teaching loads in these fields can be tested by comparing course enrollments with the number of B.A. degrees granted in each field. If the service load is heavy, course enrollments in a field--expressed as a percentage of total course enrollments--should be greater than B.A. degrees in the field--expressed as a percentage of total B.A. degrees. In other words, one would expect that for fields with heavy service loads: where: CEi ~ BAi GET BAT CEi = course enrollments in field i CET = total course enrollments in all fields BAi = B.A. degrees awarded in field i BAT = total B.A. degrees awarded. The relevant data are shown in Table 4.4. Course enrollment data are shown for fall 1980 and B.A. degrees are shown for 1982. Note that in math and English, fields known for high service loads, the shares of undergraduate course enrollments are much higher than the shares of bachelor's degrees. In the behavioral fields that is not the case.
88 OD .~ · · C5 ~ 3 V, ·_ A ._ en C) ._ C) I CO q) . C I - o V X = U2 U) _ CQ as of I c: ~ o :~q ;^ Z Z ~ Ma o ~ Z To Z I{~ so ~ ~ ~ ~ o ~ lo ~ ~ e, ~ ~ ~ ~ U) _ ~ ~ U) ~ ~ O ~ ~ ~ ~ ~ O 00 _ _ _ ~ _ 0 us us ~ he ~ ~ _ ~ ~ 0 oo oo u) ~ ~ _ ~ X ~ 0 U) et O ~ 0 ~ ~ u~ o~ 0 u~ ~ ~ r~ oo o ~o ~ ~ ~ ~o o 00 ~ ~ \0 1~ ~t U) ~ ~ o o~ X L ~ ~ - I~ U) oo ~ ~ U) V) ~ ~ ~ ~ U) U) ~ e ~ · · · · e r~ v~ _ ~ ~ ~ 0 0 0 ~ u~ _ _ _ ~ r~ o ~ ~ t oo ~ ~ ~o ~ ~ ~ ~ o U) ~ ~ ~ X ~ ~ ~ ~ ~ o o o o r~ u~ ~ ~ _ ~ ~ oo ~ ~ oo ~ _ oo o ~ ~ m I~ ~ o~ ~ ~ ~ o o~ oo ~ .~ o ~ o ~ ··········. · ···~ ~ U) ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ oo ~ U) ~ o ~ ~ ~ ~ ~ ~ ~ ~ ~ oo ~ o~ U) o oo ~ ~ o U) ~ ~ ~ V) ~o ~ ~ ~ U) ~ ~ ~ ~ ~ U) X ~ ~ ~ ~ o V) ~ Io o oo ~ oo ~ ~o oo ~ o o V) ~ ~ ~ · e e e e e e ~ ~ - - 0 U) ~ ~ ~ ~ 0\ ~ ~ 00 ~ ~ ~ ~ ~ ~ r~ 0 ~ _ u, u~ ~ ~ u~ o ~ ~o ~o ~ ~ V) ~ ~ V) ~ U) ~ ~ ~ 00 \0 ~ _ _ ~ ~ ~ ~ ~ ~ ~ ~ _ ~ _ 09 ~ ~ ~ `0 ~ ~ U) ~ ~r _ ·~ U, o ~1~ e ·" ·~ 0 ~ C~ _ C,} ·_ e c~q g ca e ~ `; E = C ~:~ ~ E. ~^ Y ~. ,,,o. ~o U, IU) ~ oo o~ ~o ~o ~ oo ~ ~ ~ U) ~ oo o ~ ~ ~ _ 1 ~ ~ 1 =1 ~ :' 1 I ~ ~ ;^ =1~ ~ ~: ·_| ' ~ ' o~ o CJ s" eo, ·" U' o~ o o C~ 5 O _ ~ 1 ~ ~ 1 ~ ~ 1 4) ~ 1 ._ _ 1.= 1 1 1 1 - _ ~ o _ 1 ~ ~ C" S" 0 4) ;- _ ~ ~: - ~ c~ ~ o ~ CQ
89 In psychology, CEi = 8.3% and BAi = 10.6% CET B. ~ In the social sciences, 4 CEi = 16.0% compared to BAi = 22.2% CET BAT These data do not seem to support the hypothesis that the service load in behavioral fields when measured by course enrollments is greater than when measured by majors. However, more conclusive evidence may be needed. The data cover only a single year and perhaps the relationships have been changing over time. At the graduate level, the situation may be different. There appears to be a substantial number of behavioral sciences courses being offered to graduate students in professional schools such as public health, law, medicine, and business. In schools of public health, for example, the accreditation rules require behavioral science courses to be included in the core curriculum. In business schools, courses are offered in marketing and industrial psychology, and sociology of organizations. Law students may take courses in research on jury selection, and medical students are offered courses in psychopathology, death and dying, and public health. Ph.D.Production(Tables4.~-4.3 and Figure4.2) In 1983, 4,318 Ph.D.s were granted in the behavioral sciences (Table 4.1, line la). This represents an increase of 3.1 percent over the 1982 level. All of the growth in behavioral science Ph.D. production is due to the rising number of psychology Ph.D.s (Table 4.2, lines la and ld). The number of new Ph.D.s in clinical psychology rose by 4.8 percent in 1983; the rate of increase for nonclinical psychology Ph.D.s was only slightly lower (4.6 percent). In both cases the 1983 gains reversed the drop in Ph.D. production observed in 1982. It is only when long-term growth--e.g., 1975 to present--is considered that a difference between clinical and nonclinical psychology Ph.D. production is found (Figure 4.2~. The number of new doctorates in clinical psychology had a growth rate of 5.5 percent, while in nonclinical areas of psychology the rate of growth was -0.5 percent. In the behavioral science fields other than psychology there was a decline in Ph.D. production in 1983. The number of new Ph.D.s fell from 1,030 in 1982 to 1,011 in 1983 (Table 4.3, line la). This decline of 1.8 percent is roughly equivalent to the annual rate of change from 1975 through 1983. 4The following fields were included in the social sciences in the ACE survey: agricultural economics, anthropology, archeology, economics, geography, history of science, linguistics, political science, and sociology.
so s ~ ooo c, 4, 000 cow cam ~ 3, 000 L'J LL] <3 2 000 Lo c' ~ 1, 000 cam cam Cal Behavioral Noncl inical Psychology - +~ me; ogy Soothe; Behavi oral 61 63 65 67 69 71 73 75 77 79 ~ 81 83 F I SEAL Y EAR FIGURE 4.2 Ph.D. degrees awarded in behavioral science fields, 1961-83. See Appendix Table ClO. The estimation of the supply of behavioral researchers is further complicated by the rise of autonomous (so called "professional") schools of psychology. As research training is not generally a focus of the curricula of these schools, it seems reasonable to assume that they contribute to the number of individuals seeking employment in psychological service, and add little or nothing to the supply of research scientists in psychology. A report published in 1982 counted 43 schools of professional psychology. Nineteen offer the Ph.D. degree, 21 offer the Psy.D. (Doctor of Psychology), and 3 offer both degrees (McNett, 1982~. Only Ph.D. recipients are included in the present discussion of trends in the behavioral sciences. As a result, our figures will not be the same as those which include Psy.D. degree holders. The exclusion of Psy.D. degree holders does not completely eliminate the graduates of these schools of professonal psychology from our tabulation of psychologists. All schools granting the Ph.D. degree are included in our figures. Four large professional schools offering the Ph.D. degree produced 148 clinical psychology Ph.D.s in 1984 and 6 nonclinical Ph.D.s. The 148 clinical psychology Ph.D.s represent 8.6 percent of all Ph.D.s granted during the year. Other unidentified schools of professional psychology also add to the supply of clinical Ph.D.s. Thus, the number of researchers in psychology included in our data on the Ph.D. labor force may be overestimated due to the presence of Ph.D. holders from schools whose curriculum has little research emphasis.
91 Postdoctoral Appointments (Tables 4.~-4.3 and Figure 4.3) The number of rose from 1981 to period. In 1981, there were 972 behavioral science postdoctoral appointments. By 1983 this number reached 1,039, an increase of 3.4 percent (Table 4.1, line lc). The growth rate for postdoctoral appointments during the 1975 to 1983 period was 5 percent, over four times the rate of growth for behavioral science Ph.D.s. As was the case with graduate enrollments and Ph.D. production, the greatest growth in postdoctoral appointments is found in clinical psychology (Figure 4.31. The number of clinical psychologists on postdoctoral appointments rose from 262 in 1981 to 466 in 1983 (Table 4.2, line if). This rapid rate of postdoctoral growth (33.4 percent) outstripped the growth rate for new Ph.D.s. in clinical psychology. These new Ph.D.s, it may be assumed, contribute mainly to the supply of clinical service providers and add less to the supply of research scientists. behavioral scientists on postdoctoral appointments 1983, reversing the pattern of the previous two-year 1, 200 1, 000 as ~ 800 to cam 600 to ~ 400 to ca to cam 200 Total Behavioral / / ~ ,Noncl i n i Cal Psychol ogy ~ ~ xx+~+x - - - xx ~3 XX X C1 inica1 Psychology in',' ~^ mother Behavi oral 0. +~.~.x XXX50°'g9 ~~.. 62 64 66 68 70 72 74 76 78 80 82 84 FI SCAL YEAR FIGURE 4.3 Postdoctoral appointments in behavioral science fields, 1962-83. See Appendix Table C3.
92 The number of nonclinical psychologists on postdoctoral appoint- ments fell from 511 in 1981 to 302 in 1983 (Table 4.2, line lo). The decline in postdoctoral appointments was far greater than the decline in Ph.D. production for the same time period. In 1975 the majority of the psychologists on postdoctoral appointments were from nonclinical fields. By 1983, there were more clinical than nonclinical psycholo- gists holding postdoctoral appointments, a reversal of the 1975 pattern. In the other behavioral sciences, the number of postdoctoral appointments rose sharply during the 1981-83 period while the number of new Ph.D.s declined. The growth rate for postdoctoral appointments was 16.7 percent while the number of Ph.D.s produced dropped by 2 percent. R and D Expenditures (Tables 4.~-4.3 and Figure 4.4) Behavioral science research and development expenditures at colleges and universities declined in 1983 by 2.7 percent from the 1982 level. This represents a smaller decline than the one recorded in the previous year (Table 4.1, line 2a). Behavioral science R and D expenditures declined in the late 1970s (from $117.2 million in 1975 to $103.0 millions in 1978), rose in 1979, 1980, and 1981, before dropping again in 1982 and 1983 (Figure 4.4~. The 1981 funding level ($99.3 million) is below the 1975 funding level after adjusting for inflation. Research and development expenditures for psychology show little change between 1982 and 1983 (Table 4.2, line 2a). The $63.6 million (in constant 1972 dollars) that universities spent on psychology R and D in 1983 is just $.5 million above the 1975 level. While there were some fluctuations during the intervening years, the growth rate from 1975 to 1983 was 0.1 percent. R and D expenditures in sociology have declined during the same period. Funding declined from $38.3 million in 1981 to $35.-6 million in 1983, a loss of 7.1 percent. The 1983 R and D figure is well below the $54.0 million reported in 1975 (Table 4.3, line 2a). Sociology R and D expenditures declined in 1976 and 1977, rose in 1979 and 1980, and then declined again (1981 through 1983~. Over the entire period the average annual decline in funding was 5.1 percent. Employment Sectors(Tables4.~-4.3) For the total behavioral science population (regardless of field), the number of Ph.D.s increased during 1983 in every employment sector except nonprofit corporations (Table 4.1, lines 3a-i). The largest gains over 1981 were in self-employment (up 14.9 percent) and business sThese and all other R and D expenditure figures refer to constant 1972 dollars.
n .oa an 60 An 20 / 93 / Total Behavioral Psychol ogy A-------- --- ·. Sociology 64 66 68 70 72 74 76 F I SCAL YEAR 78 80 82 84 FIGURE 4.4 Behavioral science R and D expenditures in colleges and universities, 1964-83 (1972 $, millions). See Appendix Tables C18-C21. employment (up 10.7 percent). Unemployment also grew in 1983 (but here the percentage change figures are misleading because of the small numbers of unemployed persons). Academic employment of behavioral scientists rose 2.7 percent in 1983. These changes are continuations of longer trends observable since 1975. The only major shift in behavioral science employment patterns is the decreased level of employment in nonprofit corporations. This reverses a longer term pattern of expanding employment in this sector. Some important variations were found when the behavioral science Ph.D. population is disaggregated. Nonclinical psychologists had increased employment in hospitals and businesses in 1983, continuing trends that have been present since the 1970s (Table 4.2, lines 3a-3i). Employment in nonprofit corporations increased in 1983. Academic employment of nonclinical psychologists showed a small decrease from the 1981 level. Since 1975, the growth rate for academic employment has been 1.7 percent. Unemployment among nonclinical psychologists also rose in 1983. For clinical psychologists, the largest gains in 1983 were in the self-employed category. The number of clinical psychologists reporting self-employment rose 13.0 percent above the 1981 level (Table 4.2, lines 3j-3r). The next largest gains (10.7 percent) were for "other n employment, the category that includes postdoctoral appointments. Gains in self-employment and "other n employment followed long-term employment trends for clinical psychologists. In 1983 there was a reduction in the number of clinical psychologists unemployed and seeking work. Employment in businesses showed smaller growth in 1983 (6.8 percent) than had been the case in the period from 1975 to 1981.
94 In the other behavioral sciences, employment in hospitals, businesses, and self-employment continued to grow in 1983, but at a much higher rate than before (Table 4.3, lines 3a-3i). Academic employment also rose for this group at a level slightly greater than the average growth rate for the 1975 to 1983 period. Fewer persons were employed by government and nonprofit corporations and (like nonclinical psychologists) unemployment rose in 1983 for other behavioral scientists. Similar trends are reported by Huber (1985), who examined the employment patterns of sociologist separately from those of other behavioral scientists. Since 1973 there has been only slight growth in the academic employment of behavioral scientists. Areas of rapid growth have been in the nonacademic sectors. The changing employment distribution of behavioral scientists (favoring clinical and other nonacademic settings) may have important implications for the future of behavioral science research. Psychologists in human service settings devote a much smaller portion of their time to research than do psychologists in academic settings (Pion and Lipsey, 1984~. The shifting employment picture has had a subsequent effect on academic departments. Pion and Lipsey report that the percentage of "academic" programs in psychology has decreased while the percentage of clinical and counseling programs has increased. The Behavioral Ph.D. Faculty/Student Ratio In the biomedical and clinical fields, faculty/student ratios have had a strong positive correlation with R and D expenditures, but no such relationship can be found for the behavioral sciences--faculty/ student ratios have continued to rise while R and D expenditures and enrollments have declined (Figure 4.5~. In its 1983 report, the committee speculated on the reasons for these apparently conflicting trends. Several possibilities were considered--enrichment of faculty by Ph.D.s, graduate enrollment trends, increasing part-time employment, and inaccurate enrollment estimates. Evidence has accumulated on each of these considerations, and we are now in a better position to assess the significance of each of these potential factors. Part-time Employment Part-time academic employment of behavioral science Ph.D.s has been increasing much more rapidly than full-time employment since 1973. However, it still constitutes only a very minor portion of total employment and does not seem to be an important factor in the growth of Ph.D. faculty.
95 Academically Employed Behavioral Science Ph.D.s FY Total Full-time Part-Time ~ Part-Time 1973 19,928 19,220 708 3.6 1983 29,776 28,091 1,685 5.7 SOURCE: Appendix Table C17. 0.046 0.042 0.038 lo - ~ 0.034 z Lo ~ 0.030 - 0.026 0.022 0.018 _ , l _ O. 000-L 1 1 1 Psychology ~< Tota ~ / Behavioral i/ / f f - f if Behavioral 64 66 68 70 72 74 FISCAL YEAR 76 78 80 82 FIGURE 4.5 Behavioral science Ph.D. faculty/student ratio, 1964-81. See Appendix Table C16.
96 Accuracy of Enrollment Estimates The argument that the committee's data on behavioral science majors underestimates the behavioral science undergraduate course enrollment was discussed earlier. From the examination of enrollment data for 1984, it appears that the use of data on majors as an approximation for enrollments does not underestimate the share of the undergraduate teaching load attributed to the behavioral sciences. Enrichment As the number of behavioral scientists employed by colleges and universities has expanded over the past 20 years, an increasing proportion of them have been Ph.D.s. This process is known as enrichment. It occurs as Ph.D.s replace non-Ph.D.s on the faculty, or as non-Ph.D. faculty members receive Ph.D. degrees. Much of the increase of Ph.D.s on behavioral science faculties indecent years can be explained by this process. In 1966 only 57 percent of the academically employed behavioral scientists had Ph.D.s; by 1983, the figure had risen to 79 percent (Table 4.5~. More importantly, the Ph.D. component has continued to increase since 1977 even though total academic employment of behavioral scientists has declined. Enrichment is likely to have had its greatest impact at colleges and universities without doctoral programs. Doctorate-granting universities have had a high percentage of Ph.D.s on their faculty for several decades. However, some of the increase in the proportion of Ph.D. faculty could be due to changes in the timing of entry into the academic labor market. During the period of rapidly rising behavioral science enrollments, young scholars frequently took faculty positions without having completed their dissertation. Data from the Doctorate Records File indicate that 24.1 percent of the new behavioral science Ph.D.s in 1972 had been employed on college or university faculties during the year prior to receipt of their doctorate degree. As they finished their degrees, the number of Ph.D. faculty members rose. With the passing of the period of high enrollments and the onset of a much more competitive market, the employment of scholars without completed degrees became steadily less common. Only 10.2 percent of the 1982 behavioral science Ph.D. recipients reported faculty employment in the year prior to the receipt of their doctorate. This would also serve to increase the proportion of faculty members with Ph.D.s. Graduate Enrollment Trends Behavioral science graduate enrollments have only recently begun to decline after growing steadily throughout the 1970s. The table below shows that the graduate component has almost doubled in size relative to total behavioral science enrollments since 1970.
97 Behavioral Science Enrollments . Total Graduate Behavioral Behavioral FY Enrollments Enrollments # % of Total 1970 637 ~ 000 36 t 500 5 e 7 1975 723~000 55~400 7~7 1980 645~000 64~200 lO.Q 1981 648 ~ 000 65 ,200 10 el TABLE 4.5 Behavioral Scientists Employed in Colleges and Universities, 1961-83 Totala Ph.D.sb Non-Ph.D.sC F~scal Year N % N % N % 1961 13,700 100.0 1962 n/a 5,339 1963 n/a n/a 1964 n/a 8,143 1965 15,691 100.0 n/a 1966 17~304d 100.0 9,783 56.5 7,521 43.5 1967 18,916 100.0 n/a ~Ja 1968 21,574d 100.0 12,915 59.9 8,659 40.1 1969 24,231 100.0 n/a n/a 1970 26,180d 100.0 16,175 61.8 10,005 38.2 1971 28~129 100.0 n/a n/a 1972 29,744d 100.0 n/a n/a 1973 31,359 100.0 19,928 63.1 11,572 36.9 1974 32,980 100.0 n/a n/a 1975 35,883 100.0 23,624 65.9 12,252 34.1 1976 38,121 100.0 n/a n/a 1977 39,237 100.0 25,582 65.2 13,655 34.8 1978 39,159 100.0 n/a n/a 1979 38,458d 100.0 26,896 69.9 11,562 30.1 1980 37,758 100.0 n/a n/a 1981 38,074 100.0 28,235 74.5 9,839 25.5 1982 38,335 100.0 n/a n/a 1983 37,875 100.0 29,776 78.6 8,099 21.4 a Includes psychologists and sociologists only. b Includes psychologists, sociologists, anthropologists, and speech pathologists/audiologists. c Obtained by subtracting number of Ph.D.s from total behavioral scientists employed in colleges and universities. d Interpolated. SOURCES: NRC (197~84); NSF (1965 84).
98 Since graduate education traditionally puts greater demands on Ph.D. faculty members than does undergraduate education, it seems reasonable to conclude that the relative growth of graduate enrollments has helped to maintain the growth of behavioral science Ph.D. faculty. The best explanation for the steady growth in behavioral science Ph.D.s employed at colleges and universities is that it is due mainly to the enrichment process. The growing importance of graduate education in the behavioral sciences, and its more intensive demands on faculty time, can also be seen as a contributing factor. THE MARKET OUTLOOK Projections of Academic Demand for Behavioral Ph.D.s The number of behavioral science Ph.D.s employed in the academic sector from 1962 to 1983 forms a typical growth pattern--rapidly increasing in the early stages and slowly increasing in the later ones {Figure 4.6~. To obtain projections of academic demand, a Gompertz- type curve has been fitted to the time series from 1962 to 1983, and extrapolated to 1990 as shown in Figure 4.6. This provides estimates of the size of the Ph.D. faculty in 1990 from which we can estimate the average annual demand due to expansion of faculty. The 95 percent confidence limits are used as the upper and lower bounds on this estimate. 35 30 Ed or n3 Cal i) :~ o Cat 4~ 20 15 10 5 Am/ /~' 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 F I SCAL YEAR FIGURE 4.6 Behavioral science Ph.D.s employed in colleges and universities, 1962-83. Solid line represents a growth curve of the form: Y = (K-C)exp(-ea~bt) + C fitted to the data for 1962-83. Parameters derived from these 11 observations are: K= 35,000; C = 2,500; a = 1.73314; b = 0.12466; R2 = 0.997. Curve has an inflection point at 1968. Broken lines represent 9570 confidence limits on the estimated curve. See Appendix Table C12.
99 Additional demand is generated by attrition due to death, retire- ment, field-switching, and job changes. Estimates for these attrition rates are derived from the National Research Council's biennial Survey of Doctorate Recipients and are shown in Table 4.6. Because of expected increases in the age distribution of academically employed behavioral Ph.D.s in the next few years (Appendix Table C24), we expect faculty attrition rates to increase. Previously we have assumed a 1 percent per year attrition rate due to death and retirement through 1988. For projections to 1990, we use an attrition rate of 1.5 percent per year for death and retirement, and 3.5 percent per year (+0.5 percent) for other causes as suggested by the data in Table 4.6. The calculations are shown in Table 4.7. TABLE 4.6 Inflows and Outflows from Academic Employment for Behavioral Science Ph.D.s, 1981-83 I. Average Annual Attrition from Academic Employment in the Behavioral Sciences 1981-83 1. Total behavioral science Ph.D.s employed in academia in 1981: 28,235 2. Leaving academic employment in the behavioral sciences each year to: % of Academic N Employment a. nonacademic sectors 875 3.1 b. postdoctoral appointments 53 0.2 c. death and retirement 338 1.2 d. unemployed 205 0.7 e. total attrition 1,471 5.2 II. Average Annual Accessions to Academic Employment in the Behavioral Sciences 1981-83 1. Total behavioral science Ph.D.s employed in academia in 1983: 29,776 2. Entering academic employment in the behavioral sciences each year from: % of Total N Accessions a. nonacademic sectors 601 26.8 b. postdoctoral appointments 107 4.8 c. unemployed 151 6.7 d. Ph.D. recipients 1981-82a 1,164 51.9 e. other fieldsb 219 9.8 f. total annual accessions 2,242 100.0 III. Balancing: 1981 academic employment - attrition + accessions = 1983 academic employment 28,235 - 2(1,471) + 2(2,242) = 29,777c a Based on postdoctoral plans of Ph.D. recipients, it is estimated that 15% of these new Ph.D. cohorts took a postdoctoral appointment before taking an academic position. b These individuals were all academically employed in 1981 and 1983. The number shown represents the net number switching from nonbehavioral to behavioral fields. c Does not agree with line II.1 because of rounding. SOURCES: National Research Council (1958-85, 197~84).
100 TABLE 4.7 Projected Growth in Behavioral Science Ph.D. Faculty, 1989 High Estimate Middle Estimate Expected size of behavioral science Ph.D. faculty (F) in 1990 Annual growth rate in F from 1983 to 1990 Average annual increment due to faculty expansion Annual replacement needs due to: death and retirements other attritions Expected number of academic positions to become available annually for behavioral science Ph.D.s III Low Estimate 32,930 1.45% 450 470 1,250 2,170 32,740 1.36% 420 470 1,090 1,980 32,530 1.26% 390 470 930 1,790 a Faculty in this table is defined as all academically employed Ph.D.s, excluding postdoctoral appointees, regardless of tenure status. These projections are based on the following relationship: (F)' = (32,500)exp(-e~ 733~0 0~2466~) + 2,500, where F = size of behavioral faculty in year t. See Appendix Table C12. b Based on an estimated annual replacement rate of 1.5% due to death and retirement. c Based on high, middle, and low attrition rates of 4%, 3.5%, and 3%, respectively. The best estimate of behavioral science Ph.D. faculty size in 1990 is 32,740, an increase of 420 positions per year from the 1983 level of 29,780. Attrition due to death and retirement would add 470 · . 1,090. Total annual ~ .. _~ __ positions. An upper bound 2,170 is derived from using attrition rates of 4.0 percent per year for "other causes." A lower bound of 1,790 is computed using attrition rate of 3.0 percent per year for "other causes. n positions, and other attrition would add another demand expected under thrum ~c:~mnt-;^nc: ; c: 1 Clan ESTIMATING PREDOCTORAL AND POSTDOCTORAL SUPPORT LEVELS UNDER NRSA PROGRAMS The next 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. 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. an
101 PostdoctomlTraining Levels The features of the postdoctoral training system which must be considered in addition to the projections of faculty growth and attrition are shown in Table 4.S and are described as follows: 1) The number of accessions to academic positions who have (or should have) postdoctoral research training, (line 2~. We estimate that 20 percent of all vacancies will be filled by former postdoctoral trainees. In the best-guess case, this number is estimated to be about 400. TABLE 4.8 Estimated Number of Behavioral Science Postdoctoral Trainees Needed to Meet Expected Academic Demand Through 1990 Under Various Conditions Projected 1983 90 Annual High Middle Low Average Estimate Estimate Estimate 1981-83 1. Academic demand for behavioral science Ph.D.s annual average: a. due to expansion of faculty b. due to death and retirements c. due to other attritions 2. Total vacancies filled by individuals with postdoctoral research training annual average:C 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: a. 60% b. 70% 4. Annual number of behavioral science postdoctoral trainees to be supported under NRSA programs: a. if 40% of pool is supported under NRSA b. if 50% of pool is supported under NRSA c. if 60% of pool is supported under NRSA 49~570 620-720 740-860 a Assumes annual attrition rate due to death and retirement of 1.5%. b Assumes high, middle, and low annual attrition rates due to other causes of 4%, 3.5%, and 3%, respectively. c Assumes that 20% of all vacancies will be filled by individuals with postdoctoral research training in the behavioral sciences. d Assumes that 15% of the 1981-82 Ph.D. cohorts took a postdoctoral appointment before taking an academic position. See Table 4.6. 2,170 450 470 1,250 4 1,980 420 470 1,090 400 1,790 390 470 930 360 1,430 1,330 1,200 1,230 1,140 1,030 460-530 570-670 680~00 2,241 770 338 1,133 1 10 - 285d 1,005 365 (1981-82) 410-480 520-600 620-720 SOURCES: Tables 4.6 and 4.7.
102 2) The appropriate length of the postdoctoral research training period and the proportion of trainees who aspire to research careers (line 3~. If the appropriate length is 2 years, then the pool size needed to produce 400 trained scientists each year would be 800. If only 60 percent of the trainees seek academic appointments after completing their training, then the necessary pool size must be 1,330. 3) The proportion of support of the total pool of behavioral science postdoctoral appointments that should be provided by the federal government (line 4~. We are assuming a range between 40 and 60 percent. The resulting range of NRSA postdoctoral trainees is between 410 under the lowest set of assumptions, and 860 under the highest set. The best- guess assumptions yield a range of 460-800 postdoctoral trainees in the behavioral sciences. Predoctoral Training Levels This analysis of the training system may also be extended to graduate education in the behavioral sciences and the level of predoctoral support under NRSA programs. The size of the postdoctoral pool needed to satisfy academic demand under specified conditions was computed in Table 4.8 (line 3) to be between 1,140 and 1,330 in the best-guess case. This becomes the basis for estimating NRSA predoctoral support levels as shown in Table 4.9. If the training system requires a postdoctoral appointment of two years duration, then between 570 and 665 postdoctoral trainees would be expected to leave the pool each year (Table 4.9, line 2~. To maintain a stable system, the number of Ph.D.s entering the postdoctoral pool each year would have to equal the attrition. And if the number of Ph.D.s who seek postdoctoral appointments is between 14 and 16 percent of each cohort, then the annual Ph.D. production rate must be between 3,560 and 4,750 (line 3~. The ratio of Ph.D.s granted to graduate enrollments in behavioral fields has varied in a narrow range between 6 and ~ percent since 1960. If this ratio holds for the next few years, graduate enrollments would have to be between 44,500 and 79,200 (line 4~. The percentage of graduate enrollments that receives predoctoral support from NRSA programs is small--currently around 1 percent. To maintain the system at this level, 450 to 790 predoctoral trainees in the behavioral sciences would be needed each year (line 5~.
103 TABLE 4.9 Estimated Number of Behavioral Science Predoctoral Trainees to be Supported Under NRSA Programs Projected 198~90 Actual 1983 1. Estimated number of postdoctoral trainees needed to satisfy demand under the committee's most likely estimate (from Table 4.8) 1,14~1,330 1,039 2. Annual attrition from postdoctoral pool if average length of appointment is 2 years 57~665 520 3. Number of Ph.D.s needed each year to maintain postdoctoral pool level if percentage of Ph.D.s seeking a postdoctoral appointment is: a. 14% - 4,07~4,750 4,318 b. 16 To 3,56~4,160 4. Average graduate enrollment needed to produce the required number of Ph.D.s if annual completion rate is:a a. 6% 59,30~79,200 63,500 b. 8% 44,504}59,400 5. Annual number of NRSA predoctoral traineeships needed if 1% of graduate students are supported under NRSA programs 45~790 516 (1982) _ a The completion rate is defined here as the ratio of Ph.D.s awarded to graduate enrollments in a given year. This ratio has varied in a narrow range generally between 0.06 and 0.08 since 1960. See Appendix Tables C1 and C10. SOURCES: Table 4.7, Appendix Tables C1 and C10. SUMMARY which there has been a small Apart from nonclinical psychology, in reduction, behavioral science Ph.D. faculty size has continued to increase throughout the 1970s and early 1980s. This is not due to a general increase in faculty size--total behavioral science faculty (including non-Ph.D.s) leveled off around 1977--but to an increase in the Ph.D. portion at the expense of those without doctorate degrees. The percentage of behavioral science faculty with doctorate degrees increased from 57 percent in 1966 to 79 percent in 1983. Projections to 1990 indicate that this growth in Ph.D. faculty positions will continue, but at a slower pace. Judging from the age distribution of the faculty, attrition due to death, retirement, and other reasons is expected to accelerate toward the end of this decade. A portion of the vacancies created by expansion and replacement should be filled by behavioral scientists with some postdoctoral training experience. The fraction of faculty accessions
iO4 with such experience is currently small, but the committee has recommended that it be increased. If we assume that 20 percent of newly hired behavioral science faculty members should have post- doctoral training, then the appropriate level of NRSA postdoctoral training in the behavioral sciences for the 1988-90 period should be in the range of 460-800 trainees and fellows annually. The committee's last recommendations for postdoctoral training in the behavioral sciences were for 440 trainees in 1985, rising to 540 in 1987. to maintain the current structure of the system, a number of predoctoral awards should be provided. currently in the behavioral science fields about 1 percent of graduate students receive support from NRSA programs. If the system is maintained at this level, 450-790 predoctoral training awards should be made available annually during the period 1988-1990. Similarly,
