Click for next page ( 53


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 52
3. BASIC BIOMEDICAL SCIENCES During the past year, the Committee and its advisory Panel on Basic Biomedical Sciences have reconsidered each of its most recent recom- mendations in light of The present status of those factors that are relevant to the training of biomedical scientists and their employment. In particular, attention teas been given to (1) updating the model projecting supply and demand for biomedical researab personnel, (2) relationships between graduate student enrollment and doctoral degrees awarded, (3) changing employment patterns for young investigators, and (4) the effect of training grant support on the quality of the training program. Assessments of these subjects as they relate to personnel needs in the biomedical sciences are discussed in this chapter. It is important at this point to emphasize that the term Faculty, as used in this chapter, id ludes all biomedical scientists holding doctoral degrees, except those having appointments for postdoctoral training, who are employed by academic institutions. Thus, it refers to individuals with regular academic faculty appointments as well as those occupying research positions. When reference is made to those having full academic responsibilities, including that of serving as principal investigator under research grants and contracts, the term n regular faculty. is used. Projections of demand for biomedical science Ph.D. 's in academia are based on needs for all faculty-- regular and research {nonfaculty doctoral research staff--positions. To properly assess the data and recommendations presented in this report, it is essential to understand how so ientif ic training is obtained in the basic biomedical sciences and what roles are played by various agencies and elements in the scientif ic community in this process. We shall give a general description of the career path of the average scientist and the characteristics of a typical graduate training program, but there is great variability f rom individual to individual and f rom program to program. THE TRAINING OF BASIC BlOME:DICAL SCIENTISTS It seems to be generally true that many students entering college have already made decisions about the broad field in which they will pursue their education. Those who have the potential for a career in the biomedical sc fences bave taken a college preparatory course in high school that includes chemistry, physics, and a substantial amount of mathematics where these have been available. While the requirement for ~ strong mathematical preparation is somewhat less important for students wishing to obtain an undergraduate degree in biology, as compared to chemistry and physics, such preparation is essential for students who wish to enter a graduate program and obtain an advanced degree in one of the basic biomedical sciences. It is with this group that we are concerned. The importance of the underg raduate program and its prerequisites lies in the limits that these rigorous preparatory courses place on the numbers of students qualified to proceed into graduate study and who ultimately supply the national need for highly trained personnel in the biolog ical sc fences. 52

OCR for page 52
During the past 10 years the number of students opting to pursue undergraduate degrees in biology has increased steadily. On many campuses, biology is the most popular single major. Simultaneously, the number of students enrolling for undergraduate degrees in chemistry and physics has declined. These changes reflect not only the public ' s awareness of the major advances that have been and are being made in biological knowledge but also the social relevance of the study of biology. In addition, the desire of large numbers of college students to qualify for medical school has led to increased enrollments In under- graduate biology programs. Indeed, a majority of biology majors are aiming for careers in the practice of medic ine or one of the other health prof essions. Nonetheless, a small but vital percentage have a commitment to a career in research in the basic biomedical or clinical sciences. Graduate training programs in the basic biomedical sciences pri- marily recruit from the pool of undergraduate students in biology and chemistry. Muab smaller numbers come from undergraduate backgrounds in physics, mathematics, or other science disciplines. By far the largest number of students entering graduate training in the basic biomedical sciences were undergraduate majors in biology. For most students the training program requires 5 to 6 years of study and research effort. Students are required to formulate and 'pursue a research program that constitutes a significant contribution to knowledge in some specialized field and to demonstrate a mastery of research skills. Although the period is one of training, it must'a'lso be viewed as a period in which the student's ability to function creatively as a thinker and as a doer is developed and thoroughly evaluated. Most students take advanced courses in their disciplines and related fields during the first two years and devote the principal portion of the final 2 or 3 years in the program to research. Research is carried out with the day-to-day guidance of a regular faculty member who serves as the mentor, and it involves periodic evaluation by a committee of scientists in the discipline and related fields. Completion of the Ph.D. often requires that the work submitted in the thesis be accepted for publication in a refereed journal, and it always requires that the student present and defend his or her thesis in a public forum. ' An effective environment for graduate education in the sciences requires that many resources be brought into conjunction. They include excellent faculty, highly qualified students, specialized laboratory buildings and complex equipment, good library and computing resources, and a tradition of open and free inquiry. Many U.S. universities have created such environments. It is tempting~to try to estimate the cost of such systems for education and research, and attempts have been made to do so. There is no question that costs for quality graduate education are high and must include the costs of faculty, facilities, equipment, energy, shared resources, and so forth. These costs are 'met f ram a multitude of income sources including tuition, endowment, state appropriations, research grants from federal and other agencies, and gifts from various sources. Support for trainees (in the font of stipends) is a vital but relatively small part of the total cost. In most graduate programs in the basic biomedical sciences, students are provided with stipends or tuition remission, or both, drawn f ram a variety of sources. Many students support themselves as teaching assistants and research assistants. Stipends for such positions are approximately $5, 000 per year. They are intended to cover basic living expenses for the student with the understanding that the main objective 53

OCR for page 52
of the student is to pursue the educational program required for the Ph.D. degree. Another large group of students support themselves by loans, personal finances, and/or employment not related to academic pursuits. The latter tends to compete for attention and detract from graduate studies and to extend the period required for students to complete their program. Those who support themselves with jobs outside the academic institution may take 8 or more years. Institutions attempt to provide the more able students with support through traineeships or fellowships or as teaching assistants or research assistants since the experience gained in teaching and research generally is much more appropriate to the career goals of students than is work outside the academic environment. Important elements among the mechanisms for support of graduate students are the training grant and fellowship programs of federal agencies. In the basic biomedical sciences these programs are primarily funded by the National Institutes of Health under NRSA, although a small number of fellowship awards in these f ields are made by the National Science Foundation. Training grants provide institutions with funds to support a number of students by paying a stipend (currently $5, 040 for predoctoral trainees). Students awarded these traineeships do not need to support themselves by working as teaching assistants and research assistants or in outside employment. More importantly, they are relatively free, compared to students supported by research grants/ contracts or other mechanisms, to choose the area in which they will work, the specific research topic they will address, and the approaches they will take. Training grants also pay tuition for trainees and provide funds through the institutional allowances for supplies and special services to facilitate their educational and research programs. Students awarded traineeships generally complete their graduate programs in 4 or 5 years and move rapidly and with vigor into postgraduate training, thus conserving national and institutional resources (NRC, 1976a). The cost to the federal government for stipend (currently $5,040) and tuition is approximately $11,000 per year per predoctoral trainee. In addition, the training grant may provide up to $3,000 per trainee for other items such as research support and seminar speakers for the program. Since university expenses, such as costs for faculty and facilities are very substantial, the cost to the federal government through the training grant represents only a small part of the total cost. Fellowships awarded to individuals--almost always persons who hold the doctorate--represent another important meabanism for facilitating the training of highly qualif fed individuals. Fellowships are awarded on the basis of a proposal f ram the applicant and his or her sponsor which defines specif ic objectives for research training and education. The competition is tough. Most fellowships are awarded for 2 to 3 years of postdoctoral research training and provide a stipend and an allowance for tuition costs and some research expenses. Unlike training g rants, they do not provide funds for the training institution to develop special training prog rams. The average f e llowship cost to the f ederal government is approximately $19,300 per postdoctoral fellow per year. This includes a stipend (approximately $14, 000 average) and up to $5, 000 for tuition, if required, and research support. For most doctoral deg ree rec ipients in the basic biomedical sciences, achievement of f inal career objectives rewires that training continue for a period of 2 to 4 years. In this postdoctoral period young scientists attempt to find a position in the laboratory of an 54

OCR for page 52
established scientist whose work interests them and will extend their expertise and knowledge. The postdoctoral training period may involve a substantial shift in interest for the young scientist or may refine and extend work already begun. Both alternatives have advantages. The trainee who enters a new area contributes fresh insights and benefits by gaining a broader perspective for future studies. There are no formal aspects to most postdoctoral training programs. Training is tailored to individual needs. The trainee has the opportunity for uninterrupted research and so ientif ic development. Most postdoctoral trainees or fellows are highly productive and a large proportion of the nation' s research productivity in the basic biomedical sciences can be traced to the efforts of this group. Postdoctorals are provided with stipends (currently averaging $13, 000 to $14, 000) from research grants, from postdoctoral fellowships granted on the basis of review of individual research proposals, or through training grants awarded by federal agencies, principally the National Institutes of Health. I t is important to note that the training of most basic biomedical scientists occurs in the major research universities. Students are trained bar regular f acuity members who are themselves successf ul and productive scientists. This arrangement ensures that the skills learned and the intellectual attitudes developed are conducive to successful research careers. Those faculty members must generate funds to support the research laboratory in which students can pursue research. Most such support is obtained by competition in the peer-reviewed programs of the National Institutes of Health and the National Science Foundation. Individuals trained in the basic biomedical sciences through predoctoral and postdoctoral prog rams, such as those receiving NRSA training grant and fellowship support f ram the National Institutes of Health, f ind employment in a wide variety of settings. Many, currently a majority (NRC, 1981a), find employment at academic institutions in regular faculty positions that offer an opportunity to continue research as independent investigators and to train graduate and postgraduate students in their own laboratories or in research positions as key members of the larger research teams. Alternatively, those in research positions may seek funds and pursue their own research programs. Some f ind positions in academic institutions with a principal focus on teaching and training underg raduate, prof essional, and g raduate students in the basic biomedical sciences. Still others obtain researab and administrative positions with federal agencies and private researab institutes, or in industrial laboratories, where there is an increasing demand for biomedical so dentists. On the average, graduates move into permanent employment 7 or 8 years af ter they have obtained the baccalaureate degree. The choice with respect to permanent employment may depend in part on salary. Although there is a considerable range, f irst salaries in academic positions at the doctoral level average $20, 000 to $22, 0Q0 per annum ~ 1981), while those in industry are signif icantly higher. In areas of current rapid development such as genetic eng ineering and toxicology, academia and industry compete vigorously and starting academic salaries can be higher. Nevertheless, the f inancial incentives for a career in biomedical research are not great. In addition, those graduate students who have both the personal and academic qualif ications required for admission to medical school must set aside the option of an intrinsically- interesting career in medicine and the financial rewards such careers provide. Further, the current air of uncertainty with respect to research f unding and the current lack of a national commitment to science 55

OCR for page 52
make it difficult for current practitioners to present a rosy picture of the prospects for a scientif ic career to the next generation. Given the long period required to train for a scientific career (7-8 years after the bachelor ~ s deg reel, the f inane ial and other disincentives must be counterbalanced by strong personal motivation to add to knowledge and by assistance in meeting the cost of education. At a minimum, financial support must be provided for the initial stages of training. The federal training grant and fellowship programs are vital in providing this support. THE: MARKET FOR BIOMEDICAL SCIENCE PH.~. 'S In its previous reports, the Committee has evaluated data obtained f tom a number of agencies to make recommendations of the number of trainees needed. A supply and demand model has been used to assess the relationships between the production of Ph.D. scientists, the need for research personnel in universities, and the commitment of funds from various sources to research and development in the basic biomedical sciences. The number of individuals receiving Ph.D. degrees in the biomedical sciences and the number holding postdoctoral appointments were taken as indicators of supply. Demand indicators were undergraduate and graduate enrollments in the basic biomedical sciences, and the availability of funds for research and development (R and D), both of which drive the demand for faculty in these fields. These data, combined with conservative estimates of future trends, have been used to make recommendations about the number of trainees needed. Significant problems exist In making projections from the available data. First, current data are not always available. For example, undergraduate enrollments in particular fields cannot be measured directly but must be estimated from B.A. degrees awarded 2 years later. And since the latest available data on B.A. degrees is for 1979, we can provide estimated undergraduate enrollments in biomedical and behavioral fields only through 1977. Second, changes within specific personnel groups are not measurable--only the number of people in broadly defined categories is known. Third, it had been assumed in the past that turnover in the pool of career scientists in U.S. colleges and universities was principally due to retirement and death. More recent assessments indicate this is not the case--tbere is a substantial amount of switching between academic and nonacademic positions. Finally, life science R and D expenditures have grown faster than expected in recent years. This section presents the most recent data on personnel in training and employed in the basic blamed ical sc fences, and assesses prospects for 5 f uture employment and needs for training . Current Indicators The last complete analysis of the market for biomedical sciences Ph.~.'s was presented in the 1978 report of tbis Committee. The principal finding, based largely on 1977 data, was that the Ph.D. labor force of biomedical scientists appeared to be expanding somewhat faster than the number of positions traditionally held by them. Ph.D.'s entering the labor force who could not f ind permanent positions were 56

OCR for page 52
taking postdoctoral appointments and remaining in these temporary positions for extended periods. While Pb.D. production had remained practically level since 1971, the postdoctoral pool had continued to grow at a rate exceeding 12 percent per year. The Committee was concerned that this continuing growth reflected a softness in the biomedical science job market. Data now available for 1979 show that little deterioration has occurred in the market for biomedical science Ph.D.'s, and in some respects the situation has improved. Between 1977 and 1979 employment of biomedical science Ph.D.'s at colleges and universities expanded by 5.4 percent per year--slightly better than the average of 5.3 percent per year experienced since 1973 (Table 3. 1, line 3b) . Perhaps the main stimulus for demand was provided by the growth in R and D funding, which increased at colleges and universities by 4.4 percent in real terms between 1977 and 1978, signif icantly above the 3 percent annual growth rate of the 1970's (Table 3. 1, line 2b) . NIB research grant expenditures also showed substantial real increases in both 1978 and 1979 ~ line 2c) . The Committee does not foresee continued growth of this magnitude in R and D funds but expects real growth to be about 1 percent per year for the next few years. Biomedical so fence enrollments grew steadily during the f irst half of the 1970 decade but now show signs of stabi lining . The latest year for which we can provide an estimate of undergraduate enrollments in the biomedical science f ields is 1977. These data show that total biomedical science graduate and undergraduate enrollments increased almost 3 percent per year from 1973 to 1977, but declined in 1977 f rom the 1976 levels (line 4d). Total graduate enrollment in biomedical science fields (line 4b) continued to increase by about 3 percent per year through 1979. Another prominent feature of the current market for biomedical science Phi. 's is the continued increase in the postdoctoral pool. The Committee noted in 1978 that this indicator of the health of the market was already at an abnormally high leve l. Yet the pool has grown despite the fact that the rate of Ph. D. production has not increased substantially since 1971 (lines la, lo). The implications of these trends, while not completely understood, are discussed in a subsequent section of this chapter. The Committee would It ke to devote considerable effort to examining this issue in the coming year. Proj ections Through FY 1985 Basic Biomedical Science Faculty Since the education and training of a scientist takes 6 to 8 years postbaccalaureate, the Committee must look beyond the current market situation to anticipate the supply/demand balance that will prevail several years f rom now. In past report s, the Committee has made 5-year projections of demand in universities for doctorate-level scientists. These projections are based on a model that relates the biomedical science Ph. D. faculty/student ratio (_/S) to a lagged function of life science R and D expenditures in colleges and universities (see Figure 3.1).1 The model coefficients were derived from data for the period 1962-1977. 57

OCR for page 52
' E >, I:: _ es ~ ~ ~ ~ Cal ~~ o oo ~ t _ e. ~ _ ~~ U. A: .O U. .O o Co, o Ct ._ 5 .5 U) Cal - Cal P: o ~ - id to At. ~ ~ _ 6 ~ ~ Z \0 ~ d. \,o ~ to A: ~ ~ Z Ha. _ ~ ~ ) ret ret 'c, - d. ko ~ ~ ~ t', - - ~ - ~, ~ o o o ~ ~ ~ ~ ~ ~ ~ ~ ~ oo ~ ~ ~ d. O ~ ~ O ~ 0` _ _ _ 1 ~ ~ ~ ~ ~ oo - c ~ ~ ~o ~ ~ co ~ cr~ v. - o oo o' ~o o o o ~ ~ ~ oo ~ o u) o~ o ~ ~ co ' ~` ~ ~ oo ~ ~ - ~ Yos <~$ - ~ z z z z z z z z o o- o oo ~ o o' ~ ~ ~ Yo ~ o ~ oo ~ - oY o ~ o - ~ oo ~ ~ ~ ^ eg t d _ ~ Z Z Z Z Z Z Z Z 00 ~ ~ ~ O ~ 00 ~ ~ ~ ~ ~ ~ O O o0 \0 0` _ ~ O S: ~ ~ ~ ~ 8 .. o ~3 .5 3 CO ~ _ ~ ~ o~ oo _ _ ~ ~ ~ ~ _ _ d. omm ~ ~ ~ oo ~ _ o ~ ~ oo oo ~ oo _ _ o" oo _ o o~ ~ _ ~ ~ so ~ ~ ~. oo oo _ oo o oo o ~ ~ ~ ~ ~~ - ~ ~ ~ ~ oo ~ o~ ~ ~ ~ u~ ~ ~ ~ o' - - ~o = . x ~ -o ~ Z Y E ~ Y i~ ~ E ~ 3~ 58 V) C~ o. _ oo _ oo ~ so ~ - 4 ~ ~o ~? - .. 5 7 ~ ~b i ' ~ j O d~ d~ ~ ._ ~ t ~ ~ ~ ~ ~ ~ g _ - ^ oo ,= ~ .5 ~ 3 2 ~ ~o - |~~ I- {3 o-` ~ ~ u, aCo ~ ~ tc,, .~ ~ e+ c Z - h4 ~ Z Z Z ~ ~ 3 S E E 7 ~=, ~ ~ ~ _ ~ ~ ~ jI j ~ ~ 4 ~', ~, 7 ~ ~ 3 i ~ ~ ~

OCR for page 52
0.06 o ~ 0.05 C) Z can J 1 C' _ , ~ _ Al O < m O.04 0.0` 0.02 0.01 77 72 74 76 ~ at_ 62 . _ 1 1 1 1 1 1 1 1 1 1 1 1 1 i' 1 200 400 600 800 1,000 1,200 1,400 1,600 LIFE SCIENCE R AND D EXPENDITURES (M) (1972 $, millions) FIGURE 3.1 Ph.D. faculty/student ratio in the biomedical science fields as a lagged function of life science R and D expenditures in colleges and universities, 1962~77. M is a weighted average of the last three years of R and D expenditures, i.e., M = ,/~(Rt ~ 2Rt.1 + Rt.2~. Ph.D. faculty excludes postdoctoral appointees. Solid line represents the estimated growth curve. See note 1 for the mathematical form of the model. The data are shown in Appendix Table A2. To project faculty (see explanation of "faculty" at the tees nning of this chapter} demand, the future behavior of life so fence R and D expenditures and biomedical science enrollments must be estimated. Figure 3.2 shows the behavior of these variables since 1960 and some projections of growth to 1985. Demog raphic data suggest that total enrollments in colleges and universities will peak in the early 1980' s and then begin a 10-year decline, following the pattern set by the U.S. population age 20-24 (Figure 3.3(a) ~ . The effect on particular fields is much harder to predict. Accordingly, biomedical science enrollments in Figure 3.2 (c) have been projected to 1985 using three assumptions about growth rates. The high growth rate of 4 percent per year and the low assumption of no growth represent probable upper and lower boundaries. The middle assumption of 2 percent per year i" considered most likely. Similar assumptions are made for life science R and D expenditures (after adjusting for price changes--see Figure 3.2(a) ~ . These three assumptions about future R and D patterns together with the three assumptions about future enrollment growth produce nine combinations to be considered. The projections of faculty demand derived from these assumptions are shown in Figure 3.2(b) and (d), and in Table 3.2. Under the most likely assumption of a 1 percent per year increase in real ~ and D expenditures (assumption II of Table 3.2), the faculty/student ratio would rise slightly according to the model described earlier. Biomedical science enrollments are also expected to increase by about 2 percent per year through 1985 (assumption B of Table 3. 2) . Under these assumptions, the model projects that academic demand for biomedical science Ph.D. 's generated by growth of f acuity is expected to be about 700 per year through 1985, an annual growth rate of 2.1 percent. Comparing this with the 5.3 percent annual rate of growth since 1973 (Table 3. 1), we project a slower rate of growth in The academic employment of biomed ical so fence Ph.D. 's due to expansion of faculties over the next 5 years, based on these assumptions. 59

OCR for page 52
2.0 `,' 1.5 _ , 1.0 _ ~ m / 0.5 o 3%1yr. ~' __1%/yr. _~~ ~~-1%/yr, 0.07 0.06 0.05 0 0.04 0.03 - Actual _ ___Projected 0.02 ~ . I t l O.O' 1960 65 70 75 80 85 FISCAL YEAR (a) Life Science R and D Expenditures in Colleges and Universities (1972 $) 800 700 400 300 - 200 100 1 1 ' 1 1 _ Actual ____ Projected 1960 65 70 75 80 85 FISCAL YEAR _ _ 2 _ 0.6%/Yr. '.~0,3%1yr. ~ ~-0.5%/yr - Actual _ - __ Projected 85 O 1960 65 70 75 80 FISCAL YEAR (b) Biomedical Ph.D. Facully/Student Ratio 4%/yr. / ,~, 296/yr. ~' 60 ' 0%/yr. {2 .8 50 o Ct LU ax 70 _ 40 30 20 10 _ Actual - -Projected ~ 4.7%lyr. '' __ 2.3%lyr, ^~_0.7%1yr, O 1960 65 70 75 80 85 FISCAL YEAR (c) Total Biomedical Graduate and Undergraduate (d) Biomedical Ph.D. Faculty Enrol Iment FIGURE 3.2 Life science R and D expenditures, academic employment, and biomedical science enroll. meet, 1960~77, with projections to 1985. Projections are stated in terms of expected annual growth rates for high, middle, and low estimates. See Appendix Table A2. 60

OCR for page 52
22,000 ~ 20,000 C) Q 18,000 ~ 16,000 o ce z 14,000 12,000 - ~ v,wu I I I I I I I I I I 1960 65 70 75 YEAR (~) U.S. Populesion Enin~tes of 2~24 Year Old', t960~90. Ses Appendix Table A3. 55,000 50,000 ~ 45,000 O 40,000 ~ 35,000 m 30,000 25,000 '_ 20,000 ~ 15000 1~1 1 1 1 1 ~ I 1 1 1 1 1 1 1 J I I $ 1960 65 70 75 79 YEAR (bt B.A.'s Awarded Annually ;n Biomedical Scionces, 1961 79. See Appendix Table A1. 4,000 3,500 3~000 0 2,500 2,000 1,500 z ~ ~n ,vvv ~ I I I I I ~ ~ I ~ ~ ~ ~ I I ~ I t960 65 70 75 79 YEAR (c) Ph.D.'. Awerded Annually in Biomedical Sciences, 196~79. See Appendix Table A3. Fl(;URE 3.3 Trends in population and biomedical science degrees, 196~79. 61 - ~ _ Actual __ Projected ~ _ I ~ I I I I I I I I I ~ I I I ~ ~ I 80 85 90

OCR for page 52
TABLE 3.2 Projected Grown in Biomedical Science PhD. Faculty, 1978~5, Band on Projections of Enrollment and R and D Expendituresa Assumptions about Real R and D Expenditures (m constant 1972 tollars)b in the Life Sciences in Colleges and Universities I II III Gumptions about Graduate and Undergraduate Enrollments ~ the Bio- medica1 Sciences and Medical and Dental Schools ~11 grow at Will grow at Will decline by 3%/yr. to S2.1 1%/yr. to $1.7 1%/yr. to about binion in 1985 billion in 1985 S1.5 billion in 1985 WDI grow at 496/yr., Expected size of biomedical reaching 744,000 AD. faculty (F) in 1985 43,810 42,260 39,640 decadents by 1985 Annual grown rate in F from 1978 to 1985 4.6% 4.1% 3.3% Average annual increment due to faculty expansion 1,650 1,460 1,140 Annual replacement needs due to: c death and retirement 370 370 350 other attrition 1,120 1,090 1,050 Expected number of academic positions to become arable annually for biomedical Ph.D.'s 3,140 2,920 2,540 B. Will grow at 2%/yr., Expected We of biomedical reaching 637,000 ~D. faculty (F) in 1985 37,510 36,180 33,940 students by 1985 Annual growth rate in F from 1978 to 1985 2.6% 2.19to 1.3~o Average annual increment due to faculty expansion 870 700 420 Annual replacement needs due to: c death and retirement 340 340 320 other attrition 1,020 1,000 970 Expected number of academic positions to become available Dually for biomedical Ph.D.'s 2,230 2,040 1,710 C Will chow essentially Expected size of biomedical no growth from 1977 Ph.D. faculty (F) m 1985 32,010 30,880 28,960 to 1985, leveling off Annual growth rate in F from at S43,000 ~ute~t~ 1978 to 1985 0.6% ~0.1% ~0.7% Average annual increment due to faculty expansion Annual replacement needs due tow death and retirement other attrition Expected number of academic portions to become available annually for biomedical Ph.D.'s 1,430 180 40 310 310 940 920 -200 300 890 1,270 990 aFaculty is defined in this table as all academically employed HLD.'s, excluding postdoctoral appointees. See note 1 to this chapter for a description of me model used to compute the projections. bDeflated by the Implicit GNP Price Deflator, 1972 = 100. CBased on an estimated replacement rate of l.O~o annually due to death and retirement, and 3.0% annually due to other attrition These estimates were derived from NAC (1973-80). 62

OCR for page 52
percent of students receiving federal support has declined, the percent dependent on self-support has increased, especially in the biosciences since 1974 (Table 3.5). Thus, although enrolled as full-time students, many self-supported students necessarily devote a portion of their time, which otherwise would be devoted to studies, to part-time employment to earn subsistence. There are reports also that substantial numbers of students who have financial support that is intended to allow full-time for study and research find it necessary to take leave for periods of 1 year or more to accumulate supplemental funds. These interruptions lead to delay in completion of the degree or, since some do not return, to inc reased attri t ion. The data in Table 3.5 indicate that the percentage of biomedical science students receiving support f rom federal sources has fallen f ram 44 (1967) to 32 (1978). These figures mask a more dramatic change. The percentage receiving federal fellowship or training grant support has decreased since 1970, while the proportion being paid f rom research grants has remained essentially unchanged (NSF 1973-77). Since research grants awarded to regular faculty sponsors are to pursue well-defined research goals, this change limits the range of options available to students for training and research. In a sense, students working as research assistants should be viewed as self-supported since they provide a service for the salary they are paid. In summary, during the last 5 to 7 years the research training system has been subject to a number of significant changes in the enrollment, retention, and graduation of trainees. To better understand its current status and to form a basis for future projections, certain data need to be updated and additional kinds of information need to be obtained. It appears that there has been an increase in the time required to obtain the Ph.D. degree (NRC, 1972-80), a trend that does not serve the national interest. POSTOOC1'ORAL TRAINEES AND ACADEMIC RESEARCH STAFF In the preceding section, "The Market Outlook For Biomedical Science Ph.D.'s," the postdoctoral population was considered in relation to the needs to fill regular faculty and nonfaculty doctoral research staff vacant ies in the colleges and universities of the United States. As noted earlier and in Table 3.1, doctorate-level biomedical scientists in these positions represent about 55 percent of those employed in this sector of the labor force. For most of the disciplines included under the biomedical so fences, postdoctoral training is an essential element in preparation for an academic research career. Also, government agencies, industry, independent research institutes, and the military increasingly recruit f rom the postdoctoral pool. Together they constitute significant elements affecting market demand. The steady increase since 1970 in the number of biomedical science Ph.~. 's holding postdoctoral appointments has been a cause of concern since the inception of this study in 1974. It must be recognized that the data on the number of persons in postdoctoral positions includes foreign nationals. Unfortunately a quantitative assessment of the proportion is not available at this time. The overall growth continued, at a slightly reduced rate, even with steady-state Ph.D. production over the same period. In the past, the growth was perceived as the system's 71

OCR for page 52
or ~ - c 8 Cal _ ._ Cal - C. s lo: .= - . ~ Car en g ~4 U. o U. . cr _ Cal or _ ID or _ or _ en or ~ o . ~ d~` o _ lo. _ or or _ Go ~ _ ~ - o ~ ~ S :] o ~ Cal Us N ~ hi Oi 0 0 _ ~ om ~ o o O Do rid o _ lo. o oo ~ oo om ~ - ~ 1 ~ - ~ ~ 1 o 0 1 o ~ ~ 1 0 ~ ~ ~ o. g ~ ~ ~ _ _ 0 0 ~ ~ ~ 0 0 _ ~ _t 0 U. ~ _ o.~.- o~~t - g oo ~ oo ~ g ~o _ ~ ~ _ _ 0 0 os ~ oo 0 ~ ~ ~ _ o. 8 _ 0 ~o 0 ~ 0 oo 0 oo 0 ~ ~ ~ _ 0 . ~ 0 ~ _ . 0 C~ _ o. o' ~ ~ ~ g ~ ~ ~ cr' 0 ~ ~~ 0 ~ ~o ~ oo 0 ~ ~ _ _ 0 ~ _ ~ ~ 0 0 ~ oo ~ ~ 0 oo o' ~ oo 0 0 ~ ~ _ ~ 0 _ 0 0 00 ~ t 8 ~ ~ _. ~ 0 _ o' ~ ~. O ~ ~ _ _ 0 ~ ~ oo ~ ~ 1 O O ~ ~ ~ ~ 0 _ ~ oo o~ o~~= 0 ~o. ~ 0 oo U. ~ 0 ~m 0 ~ 0 ~ oo 0 ~ oo 0 ~ ~ omooo ~ 8 ~ ~ ~ ~ _ o ~ ~ ~o ~ o _ ~ ~o oo o ~ ~ o. o o ~ ~ _ _ o. ~ ~ _ - , 8 ~ ~ ~ ~ - o o o ~ ~ _ - o. ~ o. o. o o ~ ~ _ _. o. oo ~ ~ ~ o. ~ ~ ~ o. o ~ oo ~ ~ o ~ ~ ~ ~ o ~ ~ _ _ o ~ ~ _ _ ~4 o. o. o. ~ ~ o ~ ~ _ _ o oo ~ o o ~ _ o ~ o ~ ~ ~ _ o. o ~ ~ o. ooo-~= O o0 - ' ~ ~ O _ r~ 0 v, ~ ~ ~ O cr, o ~ ~ _ o ' ~ ~ ~ ~ l, 3~.~ O ~ ~D O oO ~ o ~ _ t_ q omm o~ _ ~ _ o _ o _ o. o ~ _ ~ o _ =~ ~o (~) o~ _ o ~ ~ ~ _ 0 ~ 0 ~ Oi o ~ _ _ o. o ~ o o - o oo ~ ~ oo ~ o ~ ~ ~ u~ o ~ ~ ~ - o. g k - o ~ - ~ c,` o ~ ~ o ~ o o ~ u, o o ~ ~ ~ o o ~ ~ ~ o ~ ~ ~ - - o. o o ~ - o. o' o. ~ oo o u, o ~ ~4 - - o oo o ~ o o - ~ ~ - omo - ~ o - ~= o v, - - o ~ ~ ~ ~ o o ~ ~ ~ o ~ ~ - - o ~ ~ - =` O ~ ~ d. o ~ ~ - - o - mm~, g ~ ~ ~ ~ - o. ~ ~ ~ o. g ~ ~ ~ ~ - ~ .~ o ~ ~S Iai ~ ~ d3 ~ 8 ta U, o. oo ~ ~ V, o ~ _ ~ ~o o ~ C" _ o ~ ~ oo o o ~ o' o ~ ~ _ _ ~ ~ ~o'=~ 72 o ~ ~ ~ ~ ~ o o ~ ~ _ - o o ~ ~ ~o ~ o ~ ~ ~ ~ o ~ ~ _ - o oo oo _ o ~ ~ ~ ~ o ~ ~ _. o ~ o ~ oo ~ o o ~ ~ _ o C~ o ~ ~ ~ o ~ ~ ~ ,,, 0 ~ ~ . _ ~o C ~ 8 ~ F~ ~ o _ C~ C~ ~ o ~o ~ ~ o o ~ _ ~' _ , o. _~m , o ~ ~ mo 1 O O 0 O1 ~ o vo o ~o o _ o o~ ~ ~o o ~ _ o _ O. ~ et ~ ~ o oo ~ ~ o' ~ ~ ~ o oo ~ ~ _ . . ~ o ~ ~ ~ ~ . o ~ ~ ~ o. ~ ~ ~ oo o ~ _ omm _ o ~ C _ t_ o ~- ~ . ~ ~ o ~ ~ ~ ~ o ~ ,~o~ o' - o~ .4 ~: o O CO o .=

OCR for page 52
response to a reduction in faculty employment opportunities for biomedical science Ph.D. 's. This perception was a major element in the Conunittee's decision to recommend, in its 1976 and 1977 reports, a reduction of 30 percent in the number of predoctoral trainees to be supported through federally f unded training grants, but to maintain the number of postdoctoral trainees and fellows. It is too early to know whether the reduction in predoctoral training support frill decrease the number of graduates seeking postdoctoral appointments. It is important, however, to draw attention to other factors that must be considered as contributing to the size of the postdoctoral pool. These include a continued strong demand, fed by the availability of positions funded on research grants, for postdoctorate to do research and the demand by hiring organizations {universities in particular) for more extensive research training for new faculty. In addition, more and more research is being conducted by teams of postdoctorals, working under senior investigators, rather than by graduate students. In some programs experiencing decreases in graduate enrollments, faculty research productivity has suffered until additional postdoctoral associates could be recruited as replacements. While the postdoctoral team operation is most prevalent at the major research universities, industries also use such teams to accomplish their expanding research and development activity. In fact, industrial support of training, which has been common in such fields as chemistry and physics, has very recently been extended to the biological sciences. With the changing patterns of research operations and the evidence of increasing demands for postdoctorals, it is essential to monitor the number of open positions for postdoctoral personnel and to evaluate the chang ing roles of Ph. D. scientists in university and institutional researab programs. At many universities, the nonfaculty doctoral research staff is assuming a more important role and the number holding positions is increasing. As the terminology implies, these researchers are engaged full time in research. Although they do not have tenure, they have a considerable amount of job security and the promise of long-term careers in academic research. Even though the job situation and levels of responsibility of research scientists in this group are quite different from those holding temporary postdoctoral appointments for training, many of the titles are similar. These title similarities may be a source of errors in identifying postdoctoral trainees and may contribute to uncertainty in assessing the size and dynamics of the postdoctoral pool.. The demand for personnel to pursue research as nonfaculty doctoral research staff appears to be increasing, while the training of new Ph.D. 's is decreasing or holding steady. Thus, there will be more nonfaculty positions in academic research laboratories. Depending on classif ication by institutions, this may cause an anomalous change in the faculty/student ratio as the number of faculty members serving only in a research capacity increases. From an ARC study (NRC, 1978a) it was reported that the nonf acuity doctoral research star f in 1977 numbered approximately 4, 200 and represented 3 percent of all Ph.l). scientists and engineers employed in academia. About 37 percent of these were in the biosciences areas. Although this doctoral research staff is small in number, its contributions to the research accomplishments at the institutions are disproportionately greater. The NRC report states that between 1975 and 1977 the number holding these research staff positions increased 2096, about 2 1/2 times the regular faculty growth rate. 73

OCR for page 52
Projections of the number of postdoctoral trainees needed to meet academic demand each year through 1985 (Table 3.3) include those needed to f ill regular faculty and nonf acuity doctoral research staff positions. The Committee will continue to monitor the role and composition of the postdoctoral pool. QUALITY IN GRADUATE TRAINING The Committee and its Panel on the Basic Biomedical Sciences, in considering the role of training in stabilizing the supply of highly qualif fed personnel, have been concerned not only with the quantitative aspects of supply and demand but also with the effects that training grants and fellowships have on the quality of the training environment. Several attempts have been made to obtain quantitative and statistically valid data on such qualitative aspects of training as qualif ications of trainees, numbers of applicants to programs, quality of seminar programs, availability of special courses, f reedom of trainees to select research areas, and so forth. The variability of programs has made comparison of collected data about existing training programs almost meaningless. Each program has been tailored to the needs and qualif ications of its faculty, available local support, local physical resources, arid other institutional individualities. It is almost equally cliff icult to detect the effect of reduction of federal training support on the quality of graduate programs. Academic research departments give top priority to the training of graduate students and postdoctorals. As training grant support has been reduced, every other financial resource has been tapped to replace it so that in many instances the quantitative indicators of quality ini tially have been little affected. However, there are limits to an ~ nstitution' s ability to compensate for suab reductions. It is the Committee' s opinion, based on experiences in its members' institutions, that the limits have been reached and that very real reductions in the quality have occurred in many programs. Some measures are apparent. For example, in the section of this chapter titled Assessing Ph.D. Production in the Basic Biomedical Sciences, ~ the increase in time taken to complete training is discussed. This is a direct response to the changing support pattern and reflects a general decline in the quality of the graduate experience. The Committee is attempting to obtain more prec ise data concerning the career patterns of graduate students in the basic biomedical so fences and the reasons for the extension of time taken to obtain the degree and for leaving predoctoral prog rams. In a 1976 survey of biomedical and behavioral science departments conducted by the Committee, department chairpersons indicated that the loss of training grant support resulted in significant reductions in program activities. Support for course work, student research, student travel, laboratory assistants, and visiting faculty programs were reported as seriously affected {1978 Report, pp. 44, 276~. During 1979-1980 the Panel undertook a study, which included site visits by staff and members of the Panel, of several training programs at each of four universities. The intent of these visits was to evaluate by direct observation the roles that training grants for predoctorals had on the general quality of graduate training and to assess the impacts, both 74

OCR for page 52
short- and long-term, of changes in funding patterns and levels during the past several years. Although many negative effects were ascribed to decreases in training grant support, the Committee concluded that it was not possible to achieve a valid perspective on the effects of funding changes in sp e<:: if ic programs by a site visit assessment of several programs simultaneously on a university-wide level. In an attempt to assess the impact of loss of training grants on the departments incurring the loss, several of the departments inc luded in this study were asked to provide copies of annual reports prepared for intra-institutional purposes. Although these reports attested to the f inancial impact of the loss of training grant support, they were not a rich source of information about prog ram quality. Again, the focus of the reports was too broad, concerning all of the activities of the departments with principal concern for institutional budget reductions and other local matters. Case Studies of the Impact of the Loss of Training Grant Support on Training Programs Despite the problems encountered, the Committee and the Panel recogni ze the importance of obtaining an assessment of the quality of training programs and, in particular, the impact of lost training grant support on program quality. Having concluded that statistically reliable data across institutions cannot be obtained, it was decided to obtain case study data on a small number of specif ic programs. Although such a small sample cannot be considered to ref. lect national numerical trends, the information does offer reliable and documentable examples of specif ic responses. Two programs, one in physiology and one in bioengineering, were selected and studied independently and fairly exhaustively. The site visitors were provided access to faculty, students, administrators, and departmental records. Both departments had lost all (or almost all) of their training grant support during the last 2 years. The loss of training grant support was due almost exclusively to budgetary constraints and not to a decline in the quality of the training programs as ref lea ted by priority scores received in national competition in the NIB peer-review process. Site visits included a thorough review of the status of the program both before and after the loss of training grant support, and interviews with many of the stat f and students. Detai led reports were provided to the Panel, which included data on student admissions, enrollments, sources of support, and numbers of degrees awarded for each department. Training programs in different disciplines and/or universities are not identical in all respects but they do have in common many essential features that are often enhanced when training grant support is available. In attempting to study the effect of loss of training grant support, the Committee has discovered substantial benef its that accrue to having some support. The following signif icant points emerged f rom this study. For these two departments, neither of which has an undergraduate program, the training grant assured that a suff icient number of graduate students would be in training to warrant the presentation of advanced and spec ial courses. In the absence of a critical mass of students, such courses are being elimi- nated. In addition, the ability to innovate in the curriculum is largely lost. In both cases, some concern was expressed rega rding the viability of the g raduate prog ram in the lord term. 75

OCR for page 52
Both departments bad great difficulty in recruiting new graduate student s. In the bioengineering department, most new students accepted for graduate study, when not assured of stipend support, enrolled elsewhere or lef t the f ield. Students already in the program, having lost stipends, found positions in industry to support themselves. Unfortunately, they tended to stay there without complet ing the Ph. D . or extended their training period by several years. Although high salaries in industry are a distraction at all times, the effect is greatest when other sources of support for graduate study are eliminated. In the physiology department, funds were not available to support students during the first 2 years of the program when course work is heavy. Without suab support, recruitment of qualified students was virtually impossible. Students either went to other programs or pursued professional studies. Once in the program students tended to stay and were supported as research assistants in later years. In both departments, loss of the shared responsibility for the training program reduced the cohesiveness of the depart- ment. The effects this may have on joint research and future training cannot be fully assessed, but the effect seems to be real. In both departments, peer discussion and criticism at the student level, important in the training process, lost vitality with decreased enrollments. Research productivity in some laboratories in both depart- ments decreased due to loss of vital contributions by stu- dents. Recovery was dependent on recruiting additional postdoctorals. Seminar programs that are vital for both training and re- search have been reduced. Student travel to scientific meetings and support for student research have been eliminated. Students are now constrained to work on projects funded by competitive grants to regular faculty members who are their mentors. In the bioengineering department, recruiting regular faculty members became dif f icult due to the uncertain status of the g raduate train ing prog ram. Support staff, important for the bioengineering training pro- gram, was reduced with termination of training grant funding. 76

OCR for page 52
TRAINING ENVIRONMENT Universities, charged with the education of future scientists, should have available the instrumentation and facilities needed to ensure the highest- quality training for predoctoral and postdoctoral students. It is generally assumed that the leading research universities will have such resources and that they will be available to the students as they prepare for independent research careers. However, there have been several reports (Abelson, 1978; Blout et al, 1978; Coulter, 1979; Berlowitz et al., 1981) indicating that the quality of research instrumentation and facilities in university laboratories has seriously declined, and there is little indication that this process will be reversed under current policies and findings. Instrumentation With the rapid pace of instrument development, many instruments purchased only a few years ago are now obsolete. In addition, the cost of many new instruments and the expense of maintaining them threatens to make certain high technologies inaccessible to many university researchers. Without up-to-date instruments the capacity of researchers to contribute to new knowledge will be greatly impaired. Furthermore, opportunities to improve the state of the art, develop superior instruments, and expand their uses, all frequent accomplishments in university: laboratories, are being compromised. It is cliff icult, if not impossible, to obtain data on the effects of the deteriorating quality of university instrumentation on the training environment. However, a study by the Association of American Universi ties (AAU, 1980) examines the scientif ic instrumentation needs of research universities and identif ies some of the problems being encountered. The Executive Sununary of the report for that study will be found in Appendix B1. Fac ilities As in the case of instrumentation, training and research programs cannot operate at full capacity if facilities are inadequate. With federal support for construction and renovation of research facilities severely diminished--virtually eliminated, increasing pressure is placed on university funds. Malfunctions in aging facilities may severely impede research efforts. Projects may have to be abandoned when costly renovations are necessary to meet federal regulations. - Estimates of university facilities and special research equipment requirements were gathered in another study by the Association of American Universities (AAU, 1981~. The Executive Summary of the report for that study is provided in Appendix B2. Conc fusion Although innovative resea rch is accomplished in industrial, government, and nonprof it research center laboratories, as well as at universities, it is the university research and educational programs that are critical for training the scientists of the future for all f ields of employment. Training in and of itself, together with the instruments with which it is done and the specialized facilities in which it is done, must be g iven a high national priority. 77

OCR for page 52
In recognition of the cliff iculties being encountered by universities in providing up-to-date facilities and instruments for their research and training programs, the Co~runittee strongly endorses the recommendations presented in the two AAU reports for the revitalization of university research instrumentation and facilities. R;2COMME:NDAT IONS Predoctoral Training Levels In its deliberations, the Committee considered the data discussed in preceding sections of this chapter and drew on the experiences of its own members. It strongly endorses continuation of federal support for the previously recommended number of predoctoral basic biomedical science trainees (Table 3. 6) through the training grant programs of NIH and ADAMHA. The recommended number represents a 30 percent reduction from the level of 6,000 trainees supported in FY 1975 when the Committee began its study, but further reduction does not seem justified at this time. The number of traineeships proposed represents approximately 10 percent of the total graduate enrollment in the biomedical sciences. This recommendation is supported by indications that students supported by traineeships or fellowships have achieved higher Ph.D. attainment rates and have completed their training more rapidly (NRC, 1976a), thus conserving federal and institutional resources. Also, these individuals, following postdoctoral training, frequently as federally supported trainees or fellows, tend to remain in research, and are highly productive. Predoctoral institutional training grants also make a major contribution to the vigor and quality of programs i n the biomedical sciences and the funds should be provided for support of programs as well as trainees. Recommendation. The number of predoctoral trainees and fellows supported annually in the basic biomedical sciences should be maintained at 4,250 in each year from the present through FY 1985. Postdoctoral Training Levels The Committee has previously recommended provision of 3,200 awards (Table 3.6) for support of postdoctoral training through FY 1982 and finds no evidence that adjustment in this annual level for FY 1982-FY 1985 is warranted at this time. These awards do not add to the doctoral labor force but provide for advanced training to increase technical skills and experience so essential to successful independent research careers. In many instances postdoctorate use this training period to gain experience in new fields or in new technologies. At the same time, they contribute significantly to the vigor of research and the advancement of science. The Committee intends to monitor closely the supply/demand balance and will recommend adjustment of the number of awards for later years if necessary. Recommendation. The number of federally f unded postdoctoral awards in the basic biomedical sc fences should be maintained at 3, 200 annually for FY 1982-1985. 78

OCR for page 52
cu. 'A I: - cO it 3 - Cq 3 o - Ct ._ Ct - C~ o o U. I: en - Ct o - C, o _4 o U. o ._ - o .~s Cal .O U. _ C) a: ~ o _ Go ax Go _ Go Go - - oo Cot ooo at - - ~o of US 4u o en ~ ~ 3 e! ~ o o o ~ V) o o o o ~ U. o o o o V) ~ o ooo ooo ~ ~ 0 ~ ~ 0 ~ ~ . ~ ~ ~ ~ 0 0 0 0 0 0 ~ ~ 0 ~ ~ 0 ~ ~ 0 ~o ~ 0 ~ o" 0 oo o' oo _ 0 0 0 kD ~o C~ ~ oo U~ 0 0 0 ~o ~i oo v, ~ 0 0 0 _ ~ ~ 0 0 0 ~ ~ ~ ~ d. a, ~ ~ oo ~ cn o~ _ 0 ox ~ ~ - 3 ', o =- :~& ooo o oo U) ~ o ~ ~ o d. o o o ~ ~ o ~: o ._ g 8 ~ o 79 o ;3 8 U o ~ ~a ~ =- - oo o .= C. 8 ~ o ~ ~ ~ C=D ,_ ~ ~ & - o .= 8 ', o ~ ~ ~ ~q _ I,, ~ "o _~ oo - :^ !4 CL o' - ~

OCR for page 52
Training Grant Awards Studies of the effect of loss of training grant support on training programs show marked decreases in the quality of instruction, reduced access to special programs and instrumentation, and decreases in interdisciplinary efforts when grant support is substantially reduced or terminated. These effects may not be immediately evident and, in different programs, may produce different specific outcomes. In general, however, the end-point is a decrease in the vitality of the program and in the range of options it provides. In response to the Committee's 1976 survey, department chairpersons indicated that reductions in training grant support led to limitations in student research, student travel, visiting faculty programs and laboratory courses. Reductions in numbers of trainees were offset by directing funding for other aspects of the program toward student support. In the Committee's recent case studies, the virtual elimination of training grant support led to substantially decreased enrollments, principally due to loss of funding for new students. These effects were accentuated by concomitant decreases in f unding f rom other sources. While- the case study findings do not constitute proof, they do indicate that loss of training support can lead to critically diminished student numbers with reductions in courses off Bred, and in curricular innovation. The accompanying decrease in communication among students, can substantially decrease the quality of the educational program. While the Committee recognizes that no hard and fast measures can be developed to establish minimum grant size in so diverse a set of training programs as exist in the basic biomedical sciences, it believes that a training grant, in order to be effective, must assure a critical mass of students and serve as a cohesive force for the training program. Recommendation. While training aranEs should fungi nils to be awarded on the basis of national competition, it should also be recognized that a minimum number of g rant- supported trainees, the number of which will vary with the specific training program, is essential to ensure the criti- cal mass necessary for a vital and ef fective program. 80

OCR for page 52
NOTES I. A constrained growth curve of the Gompertz type fits the data for 1962-1977 quite well, resulting in the following mathematical form of the model: [1. 5 - 29363 (M-550, 000) (F/S - 0.037} = (0.023) e -e where: F/S = ratio of biomedical science Ply. faculty to total graduate and undergraduate biomedical science enrollment; M = weighted average of last 3 years of real life science R and D expenditures in colleges and universities; M = 1/4 (Rt + 2Rt1 + Rt+2) These expenditures are adjusted for price changes by means of the GNP price def labor. 2. The data on enrollments and Ph. D. production include non-U. S. citizens, but that fact does not account for the increasing time required to complete the Ph.D. degree. In fact, foreign students, who represent 11.5 percent of all Pb.D. recipients in the biological sciences in 1980, appear to take less time than U.S. citizens to complete the PheDe program. 81