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 17
2. ESTIMATION OF MANPOwER NEEDS
DESCRIPTION OF THE MARKETS
How the System Works
Fundamental to the task undertaken by the NAS under the NRSA
Act is an. understanding and ar.a] ysis of the system by which
biomedical and behavioral research scientists are trained
and absorbed into the labor f orce . The mechanisms that
affect career chop ces of students must be identified and
quantif fed a s f ar as po s sible . On the demand side, the
f orces that determine the number of positions available to
these scientists simil arly must be identified and measured.
In shore, a clear concept of how the education-trair~ing
system works must be developed. In this chapter the
Committee Or e sent s it s concept of the s ystem. Ir, the
following chapters the measurements that have been made on
it to date are summarized.
The population of research scientists involved consists
of ~ he following groups of individuals:
1.
Academic doctorate recipients (Ph. D., D.. Sc., etc.
in the areas and fields that the Committee has
Refined as biomedical and behavioral sciences for
purpose s of this study ~ see the section on taxonomy
in this chapter for a list of these areas and
fields) .
2. Academy c doctorate recipients in other areas who
are employed in biomedical or behavioral f ields .
3. Profess tonal doctorate recipients (M. D., D. D. S ,
D.V.M., etc. ~ whose primary interests and training
are in research.
4 . Other profe ss ionals, usually nurses with
baccalaureate degrees, whose primary interests and
training are in the area of nursing research.
Further di s cu s s i on 0 f the s e i ndiv i (1ua ~ s i s
presented in Chapter 7.
17
OCR for page 18
More than two-thirds of ~ he se research scientists are
empl oyed by colleges and universities (Chapter 3, Tab, e
3.2) . A good portion of the na' ion's biomedical and
behavioral research is conducted in these academic
institutions by this group of scientists with support
provided by federal agencies such as the NIH. It is
therefore clear that
Is a critical factor
the availability of faculty positions
affecting the employment patterns of
these research scientists. Faculty positions in turn are
largely determined by the pattern of student enrollments in
the biomedical and behavioral fields and by the availability
of research funds a ~ colleges and universities ~
In me dice ~ schools, clinical investigators perform
patient care duties in addition to their teaching and
research activities (Chapter 5~. In recent years, even
those clinical faculty members who are primarily engaged in
research have devoted an increasing amount of time to
patient care, particularly through the expansion of
community health care programs at university medical
centers. This trend is likely to continue and therefore
must be taken into account in estimating the future need for
clinical investigators.
The Committee views the education-training-employment
system as a dynamic one, where the f low of students into
biomedical or behavioral research careers involves at a
minimum those components shown in Figure 2. 1. Since the
preparation for careers in biomedical and behavioral
sciences generally begins with graduate study or even
earlier in many cases, and proceeds through the at tainment
of the doctorate degree, the starting point in Figure 2.1
the baccalaureate (B.A. or B.S. degree). From that point
on, two separate career paths lead to the production of a
biomedical or behavioral research scientist. The most
frequently taken path is through graduate school to the
attainment of the Ph.D. degree, a postdoctoral
appointment,and then into a position as a faculty member on
as an ~ nv~stigator at a nonacademic research organization.
In some cases, particularly in the behavioral sciences, the
postdoctoral period is bypassed and the individual takes a
position directly upon obtaining the Ph.D. degree.
The second customary pathway is through a medical
dental, or other health professional school where the
training is directed rot toward research, but toward the
diagnosis and treatment of diseases and other health
problems. At some point during this training, some of these
individuals may decide to pursue a research career. Those
that do so generally undertake an additional period of
training beyond the professional degree, at which time they
acquire the necessary research skills. A very small but
important group makes an early decision to combine research
training and medical training into a single integrated
training program, which leads simultaneously to the M. D. and
Ph. D. degrees.
18
OCR for page 19
Biomedical and
Behavioral
Scientists not
Currently Active
in Research in
These Areas
l
B.A.'s
Graduate
Enrollments I - ~ -
or
Medical and
_ Health Professional
Schools
r
r I I
Information Feedback Loop
Labor market conditions as indicated by placement
opportunities and relative salaries affect students'
decisions to pursue careers in the biomedical and
behavioral fields. Students' perception of career
opportunities affect the transfer rates at the valves
(A) that regulate the manpower flow.
_/ ~
_ Post-
doctorals
Biomed.
and Behav.
Active
Research
Labor
Force
l
FIGURE 2.1 Flow model illustrating the principal paths and decision points leading to the biomedical and behavioral re-
search labor force. Exits from the paths have not been graphically illustrated but are implied by the valves A and B. Solid
lines signify flow of personnel. Dotted lines signify information flow. Training grant and fellowship expenditures may
affect the transfer rates at points A. R and D expenditures and enrollments in undergraduate, graduate, and professional
schools are important determinants of the demand for personnel in the academic sector and therefore affect the transfer
rates at points B. R and D expenditures also affect demand in the nonacademic sector, as well as influencing the transfer
rates at valves A.
19
OCR for page 20
In the biomedical and clinical sciences, a Fcstdcctoral
appointment is commonly viewed as a transit ion stage between
the training phase and an establi shad position as a
biomedical or cliff. car science investigators But i ~ is a
vita ~ stage in wh ich much important re search i s done . In
fact, many view postdoctoral training to be a combination of
intensive research activity and an opportunity to enhance
the research -techniques of the trainee under the guidance of
an experienced investigator. It is generally considered to
be a necessary element of training for the graduates of
health professional schools, and somewhat less necessary,
but nonetheless desirable, for those with Ph.D. degrees. In
most behavioral science fields, however, where ccmE:lex
laboratory equipment and procedures are rot an integral part
of research, a postdoctoral period of training has been the
exception rather than ~ he rule (Chapter 4 , Table 4 .1 } .
Trains ng grant and fe1 lowship programs provide support
to students at the predoctoral and postdoctoral levels. The
stipends provided by tines" programs act as incentives for
students to pursue careers in those areas for which such
f· nancial aid is available. There is evidence from a recent
study by a committee of the NAS (NRC, 19 7 fib) that the
availability of funds for training grants and fellowships in
certain areas has been associated with increased graduate
enrollments in those areas relative to other areas ~ The
empirical evidence also indicates that there is
approximately a ~hree-year de, ay before variations in the
training program expenditures in an area are reflected in
the proportion of graduate students enroll ing in that area
(NRC, 1 976b, Chapter 5) . Whi le not conclusive, these
empi rical observant ions imply that the training programs have
some inf ~ uence in the long run on the supply of scientists.
To the extent that they do, their impact on the system would
occur at points A in Figure 2.1 and would tend to modify the
flow of students at these points.
An additional regulating factor in the system evolves
from students' perceptions of job opportunities and relative
salaries. As labor market conditions change, students tend
to alter their career plans accordingly. It is not clear at
this point, however, that students' assessments of the
market are accurate enough, or their reactions Prompt
enough, to prevent the development of short-run imbalances
between supply and demand.1 There are complex lead and lag
times in this dynamic system that are not yet thoroughly
understood. Under these conditions, the Committee believes
that the most appropriate strategy is continually to
appraise the current market situation and the near-term
outlook with a view to anticipating impending imbalances and
recommending steps to avoid them. Adjustments in federally
supported training programs should be made where and when
appropriate to help achieve system balance.
20
OCR for page 21
The Concept of Balance Between Supply and Utiliza~ ion
A comment on the concept of "balance" is appropriate. When
deal ing with Ph. D. ' s and M. D. ~ s, many of whom have
postdoctoral training, one would ~ not expect that many would
be unemployed for signif icant lengths of time. Such highly
trained persons wi Il al most always find work, and any study
'~hat views a cross section of biomedical and behavioral
scientists will generally find them employed. Any survey
that asks universities about their "capacity to absorb" or
" their need f or" more such personnel wil ~ almost always f ind
that capacity or need to be great, based on the potential
contribution of these highly trained professionals . Since
the expert judgment of faculty members usually produces
somewhat inf lated estimates of the need for additional
faculty, it is not considered to be a reliable indicator of
actual demand a s Irma sured by more obj ective criteria.
For an examination of "balances, " which is relevant to
policy, the concepts of need, demand, and utilization must
be related to the cost to society of producing such skilled
prof e s s iona ~ s, t o the prosper t s of the ir be ing emp loy ed ~ ~
activities that require the full range of skills imparted in
their cost, y training, and to the question of whether the
costs of this training should be prove deaf from pub, ic funds.
Thus, ill i s e ss entia l to spec i f y, or at le ash examine, the
future possibilities for emp, oyment of these professionals
at tasks commensurate with their training and to distinguish
realistic estimates of available future opportunities from
unrealistic notions of potential utilization.
It is here that confusion arises. Scientists observe an
enormous potential for basic and applied research. On this
basi s, some argue that, since there are insufficient numbers
of scientists to carry out these large research agenda,
additional scientists must be trained. This view was
expressed by several participants at the public hearing held
by Ohm Committee in November 1976. On the other hand,
others argue that potential alone is an insufficient basis
for diverting resources to training such personnel.
Although there are unlimited areas where professions. 1 and
technical manpower could perform productive research, it is
necessary to base public funding of draining of this type of
scientific manpower on the actual amount and distribution of
supper+ for research that the society can be expected to
provide through its social, economic, and political
mechanisms. In this view, it is erroneous to contend that,
if research support for particular areas such as the
biomedical and behavioral sciences, i s not adequate fully to
employ all qualified persons, nothing is lost because
Ph.D. 's and M.D. 's trained in research nevertheless will be
employed in other productive positions. Under these
ci rcums tanc e s the extra co st:s i ncurred in the i r spec i al i zed
research training may not be necessary for the adequate
performance of their duties.
21
OCR for page 22
The public policy question that is directly raised is
the extent to wh i ch the " extra " c c' s t s of t rai n i ng more
people than neces sary should be borne by publi c funds . Some
i nexactne s s in f orec as ting ne ed s i s inevitabl e . S ome
cushion of excess research manpower is desirable. The
maintenance of strong research training institutions is also
impor' ant for the future of science. All of these factors
must be taken into consideration in formulati ng Public
policy.
A change in national pal icy with respect to public
funding of research in a general area such as basic
biomedical sciences, or in a particular field, such as
molecular biology, could create grave difficulties if field-
switching by profe ss tonal manpower were limited. To the
extent that professional manpower is sufficently broadly
trained that researchers can quickly change fields, perhaps
after short--arm training such difficulties are less likely
to art s e . S. imi la r ly, i f soci ety ha s i n the pa s t produced a
"surplus" of research scientists who are 2mp:0y-6 at jobs
that require less than their full skills, these
professionals can shift to new fields of interest where
their training can be more fully utilized.
The variability of society' s demand for research, and
the value of having a reserve of trained research E'erscnnel
to meet emergent research needs, have been suggested as a
ra ti one le f or an ample or ove rsuppl y o f Sue h manpower . Yet
this rationale has significant faults. First, there are
limits to the amount of extra training costs for unused
skills that can be justified for the use of public funds.
Second, obsolescence of high-level research skills proceeds
at a rapid rate, and it is questionable whether these
reserves could perform effectively several years later
should society need them at that time. Finally, the quality
of incoming students might well be affected, if they could
not anticipate with confidence that they would, in fact, be
using the skills they must work so hard to develop.
The most promising new developments in manpower analysis
have grown out of attempts to remedy the shortcomings of
prior pro jections based on simple extrapolation of trends,
international comparisons, and surveys,. These newer
approaches, s ome o f which are implemented in later chapters,
seek to ~ dentify the factors, particularly those on the
supply side, that determine students' choices, and to
explain the system within which these factors operate.
These approaches have profound implications for the task of
the Committee.
First, economic factors, some resulting from the
interplay of market f orces and others from public policy
decisions that affect the cost of students' education, play
an important role in students' decision to embark upon
graduate study and to choose a particular field of
spec ialization. Relative earnings, cost and duration of
study, availability of stipends aria other support, and
22
OCR for page 23
pro spects o f emp ~ oyment in parti cu lar f i e ids o f study are
examples of such factors. The contention that the sp~cif ic
fill ~ of study or the decision to obtain a Ph. D. is
generally dependent upon economic considera' dons does not
ignore the noneconomic motivations in the demand f or higher
education. Rather, the argument i s that marginal choices
are ~ nf luenc=d by the relative economic attractiveness of
various alterra' ive areas of study. Thus, pro jections of
the supply of biomedical and behavioral scientists, and
particularly the distribution of this supply by f ields, must
be grounded in accurate estimates of the future economic
status of these fields relative to other fields requiring
similar amounts of tray Ding.
Second, estimation of the economic prospects for the
biomedi ca ~ and b ehaviora ~ s ci ence s depends i n turn on the
supply re spouse of future potential graduate students . That
is, the relative salaries and prospects for employment of
these scientist s depend on their number in the labor force,
a s we l ~ a s on re s ea rc h budget s and the number s of graduate
students to be trained. Thus, as Freeman ~ ~ 97 ~ ~ and others
have shown, "he dynamics of the supply and demand for any
broad area of professional manpower involves significant
interactions among these factors, for who ch the magnitude
and speed of ad justment cannot ea sily be estimated . For
example, a decrease ~ n the rate of real biomedical R and D
expenditure s can be expe cat ed eventual ly to reduce the
relative salaries of research personnel in this area with
the reductions affecting most quickly new postdoctoral
entr ant s i nt o the j ob ma rket . Thi s re duct ion in turn wi
tend ~ o discourage first-year graduate enrollments in this
area and perhaps undergraduate majors in biology, but may
al so lead to s one inc re a sed uti ~ i z a ti on of the now le s s
costs y personnel in the area who have already been trained
as researchers. The reduction in Ph.D. ~ s will tend to
lessen the decrease or perhaps even increase red ative
salaries and wil ~ subsequently af f ect future enrollments.
The key questions that these concerns raise for making
pro j ections of future needs relate to the length of time
these various ad justments will take, how complete they will
be, and how st ron g the i r i of ~ ue nc e wi ~ ~ be . We do no t have
precise answers to these questions. They probably differ
from field to field and surely are influenced by the degree
to which personne l in any fief ~ has reasonably close
substitutes in related fields.
Gi ven the s e uncert al ntie s i ~ i s not surpr i s ing that no
attempts have }'Ben made to further (disaggregate projected
utilization of professional manpower by area or fields. Yet
the NRSA Act clearly calls f or such disaggregation. To
approach this task for the areas of concern to the
Cor~-rnittee, it is necessary to make some assumptions
regarding the composition of future ~ and D expenditures, to
develop some estimates of professional manpower
requirements, and to assess the degree of potential
23
OCR for page 24
substitution among those trained in different fields for
various types of biomedical and behavioral science research.
Many individual s and groups, including this Committee,
are now investigating d~ fferent aspects of the problems
outs iced above. Organizations and agencies concerned with
manpower analyst s, particularly the National Science
Foundation, also are attempting to find be' ter ways of
handling these conce rns .
Nevertheless, it is assent al to realize that
pro j ec t ion s of ma npowe r supp ly and uti ~ i za tion i ne vitab ly
must be teas =d on judgments and assump~ ions. In subsequent
chapters these judgments arid assumptions are speci fled and
applied -o logical models of particle ar components of the
labor market to derive e stimates of future manpower supplies
and utilization.
Considerate e uncertainty also remains about what can be
said about the statu' ory mandate that the Committee specify
fiel ds that shout ~ be given priority in research training
programs. This issue is intertwined with the problem of
taxonomy--the labeling and classification of the fields
within the areas of the biomedical and behavioral sciences .
Both aspects of this issue are addressed in the fold owing
sect ion.
ESTIMATION OF NEEDS AT THE DISCIPLINARY LEVEL—-THE
TWIN PROBLEMS OF TAXONOMY AND MOBILITY TONG FIELDS
Each of the broad areas into which the biomedical and
behave oral fields have been divided in this study
encompasses a renumber of similar fields. Biochemistry is
pe rhaps the domi nant di sc ip li ne within the ba s i c biom edica
sciences, as measured by the number of scientists who
identify themselves as biochemists. But there are at least
20, and perhaps as many as 40, other fields that could be
Cal assified as basic biomedical sciences, depending on the
Ha xonomy that i s used.
minis circumstance raises the f irst ma jar problem
encountered by the Committee in attempting to estimate
manpower needs at the f ield leered. Many different
ta xonomi es have been used in the past and are currently used
by the f eder a ~ and nonf edera ~ organizations concerned with
its administration to describe the fields of biomedical and
behaviora ~ s cience.
Even within a single agency, such as the NIH, several
different taxonomists are used, and the differences among
them can be quite si gni f icant . Trainee s and f ellows of the
NTH are classified according to a detailed nomenclature
cad 1~d the Discipline, Specialty, Field (DSF) C:ode. This is
a system of more than 6 00 titles and subtitle s covering in
great detail the biological, clinical, and behavioral
sciences and, in lesser detail, the physical sciences, the
24
OCR for page 25
engineering, agricul sure, environmental, and other health-
r=la-~d fields, and most of the conscience fields (Tabs e
2. 1) .
Personnel empl oyed on, or contributing to, research
grants sponsored by the Inset tutes of the NIH are classifier
into a set of 133 disciple nes called the Field, Discipline,
specialty (FDS) code. It is related to the DSF code in that
it provides considerable detail within the basic biological
and Cal inical sciences, although it is more compact in other
areas.
A third taxonomy in use within the NIH is called the
Central Scientific Classification System (CSCS), a multiaxis
scheme designed to describe the NIH research grants and
contracts programs according to the fields, body systems,
and research materials involved in the pro ject.
There is at least one more taxonomy in general use at
the NIH. Each institute e use s a set of program areas to
describe the focus of its research and training programs.
In some cases tines" have the customary field titles such as
pathos ogy, bioengineering, or physiology, and in other cases
they describe much broader areas of interest, such as
carcinogenesis, blood resources, or the arsenical scientist
program.
ADAMHA has recent, y developed a two-dimensional scheme
for cat assifying its trainees and fed lows (Table 2. 2) . One
axis describes the f ields of training, while the other
describes the problem area. Because individua Is trained in
a par' ~ char field often contribute to research in several
problem areas, this scheme offers a reasonable e and realistic
method of Cal ass-` fication. The Committee has found that in
many cases, notably in the fields of epidemiol ogy and
bios~atistics and the area of health services research, Z the
area or field alone does not provide an adequate description
of an ~ ndividual' s expertise. But -~e use of a two-
d~ Tensional matrix such as in Table 2. 2, expanded to include
a more comprehensive set of fields and probe em areas would
largely overcome this problem by providing the additional
dimension required for adequate classification of an
inch vidual.
Organiza' ions other than the NIH are also heavily
involved in coil echoing and disseminating data relating to
manpower and expenditures for scientif ic activities and have
their own taxonomies. Again, ~ here is no formal mechanism
for coordinating these activities, and as a result no
standard taxonomy exists. This of course ~ eads to
considerate, e difficulty in merging anti comparing data
published by different organizations. The NSF usual ly
publishes i' s manpower and funding data only for broad areas
or fields such as l:)~ol ogical sciences, clinical sciences,
psychology, chemistry, physics, etc. The Office of
Education tabulates data about enrollments and degrees by
fief 0, whir e the NRC also cod lects information by field
about the doctorate population in its annual Survey of
25
OCR for page 26
TABLE: 2.1 The NIH Taxonomy for Trainees and Fellows: Discipline, Specialty,
Field Codea
Gen. med. and bio. sciences
Anatomy
Biochemistry
Biophysics
Microbiology
Pathology
Pharmacology
Physiology
Multidisciplinary
Radiation, nonclinical
Entomology
Genetics
Nutrition
Hydrobiology
Ecology
Cell biology
Zoology
Botany
Biology, not elsewhere classified
Other gent med. and bio. sciences
Clinical medicine
Internal medicine
Allergy
Pediatrics
Geriatrics
Obstetric s-gyneco logy
Radiology
Surgery
Otorh~nolaryngology
Ophthalmology
Anesthesiology
Neurology
Psychiatry
Preventive medicine
Other clinical medicine
Clinical dentistry
Other health-related activi _ _
Health administration
Nursing
Social work
Engineering, health- related
Veterinary medicine
Biosta~cistics
l~piderniology
Other health-related professions
and activities
Cormnunity health fields
Accident prevention
Maternal and child health
Dental public health
Mental health
Over community health fields
Environmental health fields
Radiological health
Water pollution control
Air pollution
Environmental engineering
Food protection
Occupational health
Environmental sciences
Other environmental health fields
Psychology
General and experimental psychology
Comparative and animal psychology
Physiological psychology
Developmental psychology
Personality psychology
Social psych.-sociological aspects
Abnormal psychology
Clinical psychology
Education, counseling, and guidance
Other psychological areas
Behavioral sciences except Psych.
Sociology
Social psych. -sociological aspects
Anthropology
Social sciences & related discip.
Math., phys. sciences, engr., other
Mathematics
Chemistry
Physics
Earth and related sciences
Agricultural f ields
Engineering
Otter f' elds
Only the major group headings are shown here. In the complete code, each
specialty may have up to 10 or more subspecialties.
26
OCR for page 27
0 id
c)
· - ~
14 In
u]
0
A]
A) ~ A)
H · -
OR ~
~ ~ O a: ~ S
·- ~ O
O ~ - ~ ~ ~
us ~ me ~ ~ :r:
U)
o
0
.,,
·,'
0
a,
Cal
o
4~
a:
Ma
U]
3
o
EN
a
ED
JJ
Ma
4)
tic
0 :~:
eC 4
m ~
· - ~
fig
P.
As
a,
~ .a
x: a;
~ a
0
s4 ~
~ 0
~ ·5;
0 U
In
·~4
Me
0
a
o
TIC
m
~4
o
AS
o
g
~ 0
41 4J
o
P'
u
o
43
~-
~ 0
s4 ~
~ ~o
r
C)
0 ~
· - ~
0 ql
U) C~
~4
~n
t) 4J
·^ ~
0 `:
~5
oo
P4
.^ .
o~ ~
0
0
m
~n
~4
o.
u
u)
0 ~ ~
U C O O
4~ ~ ~ ~ D ~ ~ ~ ~ ~ ~ ~ S ~
V ~ ~ ~ O ~ ~ ~ ~ ~ ~ ~ V~
U) t~ S ~ ~ ~ ~ ~ ~ ~ O
s a~ ~ ° ~ ~ ~ O ~ ~' ~ ^~=
V ~ ~ ~ ~ · - ~ ~ O -- O ~ ~ . - ~ ~ O
~ O ~ R I o~ ~ ~ ~ ~ ~ ~ o
·— ~ O O .~: _~ ~ _ ~ ~ ~ o ·^ ~ 34 ~ O
~ ~ ~ ~ ~ O ~ O ~ ~ ~ V ~ ~ ~ ~ ~ - ~ ~
O ~ O ~ :^ S S mC ·~1 V t~ i4 :>~ ·^ s: la ~ ~ -1 ,C i4
O ·~1 G) U) ~ ~ V ~C .0 0 Q~ .C ~ ~ ~ ~ ~ ~ ~ ~
0 m ~ ;; ~ ~ 0 :` ~ u' ~ ~ ~ ~ 0 ~ ~ v, ~ 0 5;
~ U) 0 · 4J
m ~ cn 0
27
OCR for page 28
Earned Doctorates and its biennia ~ Survey of Dcctorate
Recipe ents.
As a result of this plethora of taxonomies, the
Committee and panels have faced an especially vexing
problem. In carrying out the mandate of the NRSA Act to
determine the needs for biomedical and behavioral res march
personnel and to specify the kinds and extent of tray Ding to
be provided, it has been confront ea with the fundamental and
frustrating task of devel aping a satisfactory taxonomy and
relating it ~ o those currently ~ n use.
The issue of taxonomy is fundamental, because the
definition, label ing, and classification of scientific
research fie ids must occur before rational measurement and
analysis can proceed. The task would a be considerably
simpli f fed if a I] previous taxonomies could be ignored and
a f re sh s ~ a rt ma de . Un f ortunate l y, thi s would vitiat e t he
considerable body of data that has accumulated over the
almost 4 0 years that federally supported training programs
have existed. These data provide much of the material on
which the Committee must base its recommendations, and their
conversion to a common taxonomic base would be an expensive
ard ~ ime-consuming ~ask.
Despite the progres s the panel s have been making in
developing a satisfactory taxonomy in certain cases, a
generals y acceptable solution is not yet in sight, and
cons derab~ e further of f ort will be needed.
The Committee nonetheles s has had to proceed with its
own ~ axonomy in undertaking several surreys needed to obtain
data to carry out it s mandate. It conducted two surveys
during ~ 9 76 and ~ 977 to deve lop data on the current ~ abor
mark en from the viewpoint of the individual and of the
academic department chairmen. The first survey wa s directed
to recent Ph.D. 's in the biomedical and behavioral fields
and was concerned with ~ heir employment and utilization.
For that purpose, a taxonomy was developed consisting of the
f it ~ tithe es that de scribe ba sic biomedi Cal and behaviors
sciences in this study. Hereafter in this report, the
be omedical and behavioral sciences are defined and reported
tar the speci fic f ields listed in this taxonomy or
approximations ~hereto. The taxonomy employed in the
survey, and the corresponding set of fields from the
Doctorate Records File, are shown in Table 2. 3. In Chapters
3 and 4, estimates of the number of Ph. D. degrees awarded
annually wit 1 differ because of minor differences between
these two taxonomiPs .
The s econd survey was directed to the chairmen of
biomedical and behavioral departments in Ph. D. -granting
institutions. Since i+ was designed partly to supplement
data con 1 ect=d by the NSF in its annual departmental survey,
the NSF ~ ist of department titles was us ed to identif y the
group to be inc lured in the C ommi ttee ' s surve y . The se
department titan es are also shown in Table 2. 3.
28
OCR for page 29
u)
h
V,
-
4)
4>
.,'
a)
U)
3
U)
a,
a
...,
u
cn
o
s
GJ
m
u
a
Ok
.,'
CO
U)
a1
o
o
x
E~
uS,' OX
~ O E~
~ ~ ;`
E~ u) tn
. V
eq
C) C) ~ V
,,, ,,, ~ ~ V O '
U] ~ $:
a1 ~ ~ a,l ~ ~ °
tq Q~ ~ O v ~ ~ u~ "
o ~ ~ ~ ~ ~ ~ ~ ~ O
u ~ ~ a,, a, LO ~ V 60
.- ~ ~ ~ ~ S
~, c >, ~ ~ 0 ~ u, `m `) ~ U Q. a,
U s~ -- C)--
~q ~ 4~ u: O O ~ ~ O ~ ~ ~ ~ O h
U ~ ~ - ~ ` - ~ ~ ~ ~ ~ ~ ~ O ~ ~ ~ ~ ~ ~
u ·- ~ 0 ~ 0 0 ~ 0 u~ 0 -- a~ ~a =: 0 a~ ~ ~ O ~ O ~ ~ S4
ns ~ ~ ~ ~ -- O U ~ ~ ~ ~ U .- ~ ~ ~ ~ . - ~ ~ ~ ~ ~ O ° °
C' ~ ~ ~ >. R ~ ~ O ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ h O ~ O ~ ~ .
~ a, 0 .c s 0 0 ~ ~ 0 ~ ~ - - 0 0 ~ ~ ~ a, ~ u 0 s :, s~ 0 0
.^ ~ U
O ~ O O u ~ ~ ~ O o ~ ~ ~ 0 ~ ~ ~ .Q. - ~ ~ ~ ~ ~ ~ C) C) S
S S .- . - ~ ~ ~ O ~ ~ ~ ~ ~ ~ ~ ~ ~ O ~ O O
o m ~ m m ~; ~ o4 ~ m m ~ ~ Z ~ Z P. ~ ~. ~- r, m s P c~ <; u~ u~ m
.,~ a,
m m
~n
a
u
V
60 ~
~n v
~= u
0 0
.,l ~ ~ ~ ~ ~ C~
~ ~ U ~ o ~ P. ~
0 ~ ~ ~ C~ ~ ~ ~ ~ ~ ~ V U
0 U ~ O ~ ~ ~ S4 ~ O ~ C: ~ ~ `: ~
~ ~ ^° 8 .~ ~ ~ ~ ·^ o ac) ~ s ~Q~ ~ ~ C=, ~ U ~
u `: ~ 0 ~ g) ~ ~ C) u, 4~ _1 h, t) ~ a
~s u ~ ~ ~ ~ ~o ~ ~. ~ ~ ~~ ,a) a~ ~ >, ~ ~ ~~~ 43 0 ·,~ ~4 ,c
u) _ ~ :>' O O ~ :~ a) u ~ 0 ~ ~ ~ ~ O dJ 0
~: ~n E ~ v ~ .4 ~ ~ ~ ~ a~ ~ u e . ~ ~ ~ o~ .,~ ~, ~, ~ ~ ~., ~
C;-~ a~ ~ ·- ~ u~ 0 0 c~ 0 .c ~ ~: ~ O ~ O a~ 0 ~ "i ~ 0 0 ~ ~ ~d
a~ U] _ ~ U ~ ·— O V >t ~ ~ ~ ~ ~ ~ ~ ~ ~ S~ rl S ~ ~ O
v ~ ~ ~ ~ ~ ~ 0 ~ .~ 0 .4 ~ ~ C, ~ m.- ~ ~ 0 u 0 C ~ u a) Q) U) U ~ ~ ~ .C
O ~ ~ ~ ~ S ~ ~ ~ ~ O O ~ ~ ~ O U V O ~ ~ S 3 ~ ~ ~ ~ ~ ~ U ~ O U
k o o O o~ ~ ~ C ~ U ~ ~ ~ ~ O O ~ ~ S~ ~ ~ ~ ~ ~ ~ ~ ~S ~ m0 ~ ~ 0
~ ~ ~ -- ~ -- ~ ~ ~ ~ ·- ~ S ~ 0 ~ ~ ~ ~ ~ ~ O
o ~ ~ m m m m ~ ~ ~ H ~ P~ P ~ ~ ~ m ~; .c ~: ~ P< c
·,4 P)
~n ~
0 0 0 0 s v 3
a ~ : 0 ~ ~ ~ - ~ c - ~ ~ 0 ~ ~ ~ ~
6 ~a ~ ~ ~ V ~ ~ ~ O 0 6 ~ V ~ ~ ~ ~ ~ ~ ~ U ~ ~ ~ ~ ~ U
O ~ C ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ O ~ ~ ~ ~ O
U) ~ ·m ~ ~ ~ ~ {q ~ ~ ,~ ~ ~ ~ ~ ~ ~ oC,= o ~ ~ ~ .m ~U ~ ~ ~ ~ 0
U U ~ - - O ~ ~ S ~ ~ ' ~ ~ ~ ~ ~ ~ 1~- ~ m~ O'^
c) c ~ ~ a) ~ ~ ~ ~ ? ~ ~ ~a ~ c ~ 4J tn a~ ~ ~ ~ tJ a 54 ~ -4 ~ ~ t~ u) u) ~ t~ :~ ·,l ~ r;
r~ ~ ~ ~ ~ o ~ u ~ ~ ~ ~ ~ ~ ~a c; a ~ ~ ~ ~ ~ ~ ~ ~ ~ o o s ~ ~ ~
u) 0 c ~ o ~ ~ ~ o ~ o o :~ ~ ~ ~ 0 u ~ ~— ~ ~ ~ a) s :3 ~ o ~ ,c ~ ~ Q' ~ v :~
£ ~ u ~ ~) ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ .~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ X O
u~ ·H a~ ~ ~ O ~ ~ 0 0 u O O ~ O O ~ O O S ~ u ~ O ~ ~ O ~ l) a1 ,~ ~ ~ 0 0 0 ~ ~ ~ ~: k~ ~ a) ~ 0
.1: ~ u~ ~ c, ~: -~ ~ ~ ~ ~ ~ ~ .~ ~ 0 u ~ ~ ~4 ~ a, 0 ~ k; ~ ~ ,1 u~ c c: ~ ~ 0 ~ S ~ ~ ~4 0 ~ ~ tn ~ s4 R
u ~ 0 ~ ~ :~.o O O ~ ~ ~ -— O R O ,0 ~ _~ ~ O V O ~ ~ o ~ ·r~ 4,~ 0 ~ C _ ~ ~ ~ ~ U) U) U U ·— ~ ~ Q~ Ei O ~ G) O
'-~S2~= ~a~l4r~)~°'p°'i4,t ~'U)~-OfI3 ·-ol~4O~c=-b~~Siu;~ ^o~;rl6~Ei~s~ =~d
a' ~ 0 0 0 0 ~ ~ :~ ~ 0 c ~ ~ ~ ~ ~ ~ 4~ ~ :~ ~ i4 0 ~ ~ m V] 0 ~ V ~ ~ :r X X: ~ ~ P. P4 ~ V) 0 U ~ t~ C) >: 0 ~c; ~ ~ ~ O
b; ~ ·— ~ — a~ (P ~ ~ ~ 41) C~ t ~r] O a~ ~ ~ ~ ~ ·^ .— ~ ~ u~ 0 ~
0 ~; m m m oa ~ cl ~ ~ ~ ~ ~ ~- :t. >: ~* -~, P, ~ P, t~4 ~ ~ c
d)
m ;i5
29
OCR for page 30
In i t s prey iou s re pc rt s, the cocci ttee ha s grad e
recommendations only for the broad areas, awaiting the
outcome o f e f fort s by panels and staff to a sse s s the
squat ion at the fie 1d level. One fact is becoming clear as
a re suit of thi s e xp [oration . Con side rat le mobi lity occurs
we thin most of the broad areas. Individuals who receive
their Ph.D. in one field are frequently found to be employed
~. r a different but rat ated fief ~ within the area (these data
wil ~ be presented ~ n Chapter 3) . From the point of view of
the individual invcived, this might be call ed field-
switching; from the employers s point of view it represents
substitutability among the fief ds.
Al though there i s little suppor* for the proposition
that research personnel tray ned in dif ferent f ields are
completely interchangeable, field-sw~tching data do indicate
that, within some limits as yet undefined, substitution can
and does occur. The Committee's studies are progressing,
but are nor ye' to the Fain- where fields can be grouped
into those that represent homogeneous clusters within which
there is considerable mobility and between which mobility is
more difficult.
A second major problem then arises . If there is general
substitutabi lity among the fields of a particular area
(e. g., the basic biomedical sciences}, is there any need to
emphasize training in a particular field within that area?
would ~ a policy of providing a broad spectrum of training be
the proper on" under these circumstances?
The Committee has thus far concluded that predoctoral
trai ning sho ~ ~ ~ c ore r a broad range of f ie Ids so that the
student may be broadly exposed to vari ous aspects of the
sc ten if ic area . The identi f ication of priority f ields f or
training at ~ he predoctora~ level in general would be an
undesirable policy. If fields are to be singled out for
priority in the allocate on of training funds, this should be
done pr i marl ~ y at the po sttiocto ra ~ We're ~ .
The third ma jor problem, therefore, is to develop
criteria for assigning priorities to postdoctoral training
fields. At the broad area level (i. e., basic biomedical
sciences, clinical sciences, e4_c. ), the Committee has based
its previous recommendations on an assessment of need as
determined by the liked y near-term balance between supply
and demand at prevailing salary levels. Trenas in
enrollments, Ph. D. production, and R and D expenditures have
been prominent factors in the analysis. It is doubtful that
a similar approach would be feasible at the field level,
given the difficulty of obtaining detailed data on the
various f actors invo ~ ved . However, it i s quite possible
Ma ~ f urther study o f this problem wil ~ yi eld an alternative
fees ible procedure.
One approach that the Co~runittee has been investigating
would attempt to determine training priorities by examining
the field mix represented by the professional personnel that
have been utilized on research grants of the sponsoring
30
OCR for page 31
agencies. The basic premise of this approach is that each
of the problem areas of interest to the NIH, ADAMHA, and HFA
util i ze research personne ~ f ram a di f fer ent mix of fi elds,
and if ~ he f ield mixes were known in each case they right
serve a s gui des to training priorities ~ ~ n e s senc e, this
approach provides a direct ~ ink between training and
research programs.
How is this linkage to be constructed? Since ~ 973, the
NIH has been collecting data on the character) sties of
personnel employed on their research Grants and those
contributing to, and not paid from, grant funds. These data
are now being alla lyzed by the Committee. Although
preliminary results are available (see Volume 2) , it is
blear that some basic probe ems must be solved before these
~ inkage s c an be re ~ fed on . Among them i s the ta sk of
deci ding how the probe em a rosa s s hould be speci f ied--whethe r
the level of resolution should be the study sections that
r~vi ew the grant applications, the program areas of the
i nstitutes, or just the institutes themselves. Whichever is
chosen, it must lend itself readily to the estimation of
future funding levels, for the problem of specifying current
training priorities is closely related to the problem of
anti cipating future research needs. These in turn will be
largely determined by the level and di stribution of grant
and contract funds.
Another survey of the labor market for biomedical
scientists, conducted by westat, Inc., for the NIH in 1975,
wa s des igne ~ to d eve lop inf ormati on on both the demand and
supply sides of the market. Employers were asked to list
the number of budgeted vacancies occurring between May and
September ~ 9 75. At the sarre time, suppliers (primarily the
training program d~ rectors at colleges and universities) of
biomedical science ists were asked to list the number of
persons expected to complete thei r training during the same
per' oaf. National estimates of supply and demand were then
developed from the responses. The NIH has reported that the
survey results indicate shortages of biomedical scientists
existed i n certa in f ie Ids in ~ 97 5, parti cular ly in the
clinical area.
The Committee has followed the progress of the We stat
survey with interest and has requested details of the
sampling and estimating procedures. Substantial questions
have beer raised about the survey methodology and, in
particular, about the criterion for estimating the demand
for biomedical scientists. Unfortunately, the details of
this survey were made available just as the report was in
preparation and ~ he Committee was theref ore unable to
analyze them ~ n time for inclusion in this report. Further
study will be given to the westat survey over the coming
year, and a ful ~ discussion wi 11 be included in the ~ 978
report .
Only in the basic biomedical sciences area has the
committee felt that sufficient ~ nformation is available to
31
OCR for page 32
enable sore spec i f ~ c statements to be made thi s year about
priority f ields (Chapter 3) . AS sta ~ ed in Chapter 1, survey
resume ts and expert judgment have been the main. sources on
which the Commi ttee ha s re lied i n recommending s recta
attention be given to certain f ields.
OVERVIEW OF LABOR MARKET ANALYSES
In the f o ~ lowing two chapters data are pre sensed on the
popul anion, manpower flows, and anticipated balance between
supply and demand ~ n the broad areas of research covered in
this study. Knowledge of the market system is f ar from
complete in terms of the information desired, especially in
the areas of clinical sciences, health services research,
and nursing res earch. Nonetheless, the Committee was able
to complete an. extensive analysis of employment prospects in
the teas ic biomedical and behave oral sciences. In particular
the committee examined (~) the extent to which changing
market conditions have affected the employment patterns of
recent Ph.D. recipients in these fields; {2} the factors
that have contributed to recent expansion in the biomedical
and behavioral labor forces; and (3} the emp' oyment outlook
in these f ields, given various assumptions about f actors
af fecting the demand for Ph. D. personnel in the academic
sector.
In examining the current employment experience of recent
biomedical and behaviora ~ science graduates, the Committee
noted that other surveys (e . g ., NRC , ~ 97 3 and ~ 9 7 55) had not
revealed high levels of unemployment, even in f ields that
might seem to have a signif icant oversupply of trained
research personne ~ . The Committee decided to investigate
chan ge s in the type s o f po siti ons taken a f ter gra duet i on and
to search for shifts f rom presumably overpopulated training
f i el ds into other f i el ds with gre ater employment
opportunities . ~ n many biomedical science fie Ids,
emit oyment cliff icult' es might manifest themselves ty
prod ongation of postdoctoral study. In other biomedical and
behavioral fields, market imbalances might be reflected by
increases in the number of graduates taking positions that
di a not uti ~ i ze ~ hei r predoctoral training. Wherever
possible, these market ad justments were considered
separate' y for each field within the biomedical and
behaviors ~ s ciences.
The Committee also attempted to identify factors that
have contributed to the expansion of the biomedical and
behavioral science labor forces in recent years. An
understanding of the relative importance of such factors as
annual Ph.D. production, field-switching, and attrition
(retirement and death) has provided valuable ins ight into
the likelihood of continued growth in these labor forces.
32
OCR for page 33
Some attention has also been given to differences in the
rates of expansion in the academic and nonacademic sectors.
Finally, the Committee has developed models of academic
markets demand for Ph. D. personnel in the biomedical,
behavioral, and clinical sciences. Projections of future
needs for research personnel were made on the basis of
alternative assumptions about anticipated changes in both R
and D expenditures and total graduate and undergraduate
enrollments in these areas. The assumptions used in each
model cover what are considered to be the likely range of
possibilities f or re search funding ~ evels and enrollments in
the next five years. Projections of demand for faculty were
derived from al ~ ernative app, ications of the models and
compared wi' h expected growth in the research labor forces.
Although data in the areas of clinical science and
health services research and nursing research were not as
readily avail ate, e, some preliminary analyses of labor market
conditions were undertaken in those areas as well. In the
Cal inical science area, a model based on P< and D funding and
enrollments in medical schools has been developed to pro ject
future demand for clinical science faculty (Chapter 5).
While thi s model is not as precise as the committee would
like, it provides some insight into future demand for
clinical invest. galore. No accurate count of the number of
currently active clinica ~ investigators could be made at
this time. Nonetheless, information frog NIB training
programs and other sources has been used to estimate trends
in the numbers of M. D. ' s entering research careers in recent
years.
In the areas of heal th services research and nursing
research, some ~ nformation on the utilization patterns of
recent graduates has been collected (similar to that
avail able for biomedical and behavioral science graduates) .
This information has provided the Committee with some
indi cations of current supply i n these two areas . However,
it ha s not been poss ible to determine the s i z e o f the
populations available to do research. Personnel qualified
to work on problems fin health services research were
particularly difficult to identify since they received
doctoral training in a variety of b: omedical, behavioral,
and re ~ ated f i e ld s and since many of them. have moved into
and out of the health services research at different s' ages
in their careers . Even less i s known about the future needs
for personnel trained in areas of health services res earch
and nur s ing He s ea rch . Ne~rerthe ~ e s s, ~ he C ommi ~ tee wa s ab le
to make some comment s on the outi ook f or the s e are a s on the
basis of its knowledge of the current market experience for
recent graduates and its general impressions of
opportunities for research in these areas. These will be
found in Chap' er 6 for health services research and in
Chapter 7 for nursing re search .
33
OCR for page 34
FOOTNOTES
1. The direction of the adjustments are difficult to
foresee. For example, shortages in empicyment opportunities
in a f ie 1 d may inf luence some persons earning baccalaureates
n that f ie ld to enter graduate school rather than ~ eke a
job that does not meet their expections.
2 . ~ ~ i st of pro bleary categor i es that the Pane ~ on H ea Ith
Service s Be s "a rch ha s deve loped to def ine that area i s
pres end ed in Chapter 6 .
34
Representative terms from entire chapter:
services research
