|
Items in cart [0] |
Questions? Call 888-624-8373 |
| Copyright © 2009. 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 73
EVIDENCE OF ADAPTABILITY
IN THE LABOR MARKET FOR ENGINEERS:
A REVIEW OF RECENT STUDIES
Robert C. Dauffenbach
Oklahoma State Un~versi~
and
Michael G. Finn
National Research Council
Introduction
Understanding the degree of adaptability in the labor force is important for analysis
of many important issues that need to be addressed today in planning for the needs of
tome. In the occupational domain of scientists and engineers, that need for
understanding is paramount. It is arguable that no other occupational domain is more
crucial to the present and future compeduveness of the U.S. economy.
Adaptability can be measured in a variety of ways, such as by the ability of
individuals trained in given disciplines to attain career success in alternative occupations; by
the rates of mobility among firms, occupations, and regions of the country; and by the time
and resource costs associated with retraining. The willingness and ability of individuals
trained in one field of study but working in alternative occupations would seem to be a
primary measure of flexibility in the economy. In this sense, anyway, the extent of
flexibility among science- and engineering-educated personnel would seem quite large.
The Swey of Income and Program Participation (SIPP) provides national labor force
evidence on education/occupation correspondence. While this evidence is highly
aggregated and is, given the relative small sample base for the SIPP sweys, subject to
large sample variation, it is practically the only evidence available. The education and
occupation correspondence shares for bachelor's and higher degree holders who are
employed in scientific and eng~neenng (S&E) occupations are as follows:
73
OCR for page 74
S&E Major Degree Field
Engineenng/compu~ang
Math/sta~astics
Agoculture/fores~y
Biology
PhysicaVearth science
Psychology
Economics
Social sciences
Yield to S&E
Employment
46.3%
24.9%
15.4%
19.7%
34.7%
17.1%
15.7%
_ ~5.9%
The largest of these numbers, 46.3 percent for engineers, is surprisingly small.
Note that this number reflects all S&E employment, not just simply engineering
employment (which, of course, dominates this category). The next highest value is for
physical and earth sciences, at slightly more than one-third. Social science discipline yields
are particularly low. Evidently, most individuals trained as scientists and engineers work
in jobs outside science and engineering. Unfortunately, Deere is no knowledge of dhe extent
to which such individuals are willing and able to undertake S&E jobs that is, the rate of
back flows. It is apparent, however, that among S&E graduates the yield to S&E
employment is quite low.
~ a study by Dauffenbach (1989), an analysis of venous adaptability issues in
science and engineering labor markets is presented. Analysis of the correspondence
between occupation and education is seen in his study as a primary vehicle for appraisal of
flexibility in labor markets. Dauffenbach provides several cross-tabulations of detailed
occupation by field of study for science and engineering occupations using NSF's 1982
postcensal survey data. A major finding is that while detailed field of study is a good
predictor of occupational pursuit, the amount of variance is surprisingly high.
Such prevalence of non-exact correspondence between occupation and education
can be taken as evidence of flexibility in the S&E labor market. Dauffenbach hypothesizes,
however, that such flexibility is not without attendant costs in the form of diminished
productivity. Presumably, if there are productivity differences between workers who are
appropriately credentialed and others doing the same job, these differences should show
up, systematically, in salanes. Thus, he undertakes an extensive analysis of salary
74
OCR for page 75
differentials by applying the statistical methodology of multiple regression analysis to the
aforementioned NSF survey data. These differentials, in general, support the notion that
quality differentials result from non-alignment of degree field and occupational pursuit. In
addition, because of the variety of other factors that need to be held constant in order to
have an unbiased assessment of the impact of education/occupation correspondence on
salaries, the regression results provide a general assessment of the various factors on salary
differentials, such as race, sex, primary work activity, industry, occupation, and
professional work expenence. Also, because in the 1982 postcensal survey, certain
questions pertaining to mobility were asked, it is possible to investigate the impact of inter-
f~n and ~nter-occupa~onal changes on earnings.
Degree and Employment Fields
Dauffenbach provides separate regression results for each of the major domains of
S&E employment: engineering, biological sciences, math/computer science, physical
sciences, and social sciences. He concentrates on detailed categories of field of study and
occupation in his analysis and explores the correspondence between field of study and
occupation on three levels.
First, there is the exact match level in which an individual is working In an
occupation that corresponds exactly to his or her field of study, such as a mechanical
engineer holding a highest degree from the mechanical engineering field. Second, there is
the associated-f~eld level of correspondence, such as a mechanical engineer working as an
aeronautical engineer. Third, there is the non-associated level, such as an individual with a
highest degree in education working as an engineer. Of course, there could well be
differences among major degree fields. Physical science graduates would be expected to be
more readily interchanged for engineers ~an, say, business school graduates.
Consequently, non-associated fields were divided into several major fields of study
(including health, education, business, and "all other") in addition to the major science
fields. After substantial investigation it was decided to use 35 distinct degree fields, which
were mapped into 40 S&E occupations. These categories represent the numerically
significant fields and occupations in the NSF postcensal survey.
In the five sets of regressions provided by Dauffenbach, a basic point of interest is
the extent to which individuals appear to be working in fields not directly associated with
their detailed major field of study (Table 1 ). These results are interpreted in the following
75
OCR for page 76
Table 1. Degree Field Shares, by Field of Employment (in percent)
Occupational Field of Employment
Field of Mad & PhysicalBiological Social
Study Engineering Computer ScienceScience Science
Exact Match 54.9 21.6 71.161.3 68.4
Engineering 25.1 11.0 4.80.8 1.0
Maw & Computer 3.0 23.0 1.90.3 1.2
Physical Science 5.0 5.2 8.24.2 0.4
Biological Science 1.2 3.4 7.426.1 1.3
Social Science 1.3 8.4 1.91.5 10.5
Education 1.5 5.6 1.52.4 3.2
Health 0.1 0.4 0.60.5 0.3
Business 4.6 1 3.0 0.90.5 4.9
AD Other 3.3 8.4 I.72.4 8.8
manner. Among all of the nearly 20,000 observations of employed engineers, about 55
percent had an exact match between their detailed employment field and the detailed field of
Weir highest degree earned Another 25.1 percent of the employed engineers had an
engineering degree, but their degree field did not match their employment field. This leaves
a residual of about 20 percent with a degree in a non-associated field. For engineenng, the
most prominent of these non-associated fields was physical science, followed closely by
business. Other results are read in a similar mariner. Note also that individuals with their
highest degree in engineering represent sizable proportions of both the math and computer
science and the physical science employment fields.
Earnings Differentials
Of primary interest is the impact on earnings differentials associated with not having
an exact match or having a highest degree in a non-associated field of study if, in fact,
there are real productivity differentials associated with individuals who do not match in
76
OCR for page 77
Table 2. Regression Estimates of Earnings Differentials (in percent)
Occupational Field of Employment
Field of Math & Physical
Study Engineering Computer Science
Biological Social
Science Science
. .
Exact Match 10.04*7.27* 16.42* 0.679.19*
Engineering 9.87*7.76* 18.58* 15.72*16.84*
Math&Computer 9.93*8.51* 21.61* 6.3510.62
Physical Science 5.64*5.97* 12.04* 8.9318.24
Biological Science 0.81-0.21 5.79 -1.50-12.46
Social Science 1.83-0.19 0.78 1.481.74
Education -5.03*-2.67 4.90 -4.26-8.91
Health 3.913.81 12.25 9.40-59.86*
Business 3.~*2.83 23.20* 16.516.66
All Other
* The coefficient is statistically significant at conventional levels.
tempts of degree field and employment field (Table 21. The coefficients are read relative to
the salary associated with the "all other" fields of study, which is the excluded group in the
regression analysis. Thus, we see that an exact match in the engineering employment field
pays about a 10 percent differential above those who have a degree in the "all other" field.
However, having a non-exact match but still having an engineering degree pays almost as
much, a 9.87 percent differential. Note also that having a degree in math/computer science
and working as an engineer also pays a handsome differential of 9.93 percent. Yet, having
a highest degree in biology, physical science, social science, education, health, or business
pays somewhat less.
~ the math/computer domain, the coefficient for the associated field (same general
field, but not an exact match) is actually higher than the exact-match coefficient.
Engineering graduates earn about the same; physical science graduates, slightly less. Other
fields of study are somewhat lower. Business graduates are a large contributor to this
77
OCR for page 78
OCR for page 80
OCR for page 81
OCR for page 82
Representative terms from entire chapter:
employment field
employment field. They earn 4-5 percent less than engineers working as math/computer
· .
specialists.
A total of 75 percent of those working in the physical sciences have physical
science degrees. In this employment domain, individuals with biological science degrees
are the most frequent other contributor. They earn significantly less, about 7-1 1 percent.
Engineers, another significant contributor, earn about the same amount, if not a little higher
Can the exact-match.
Also of interest in these results is the finding that in all of the major S&E
employment fields, those who have engineering degrees earn as much or more than those
who have exact matches with their employment field. This result seems especially
significant in regard to flexibility of engineers. However, as noted previously, engineers
represent a sizable proportion of only the math/computer and physical science employment
categories. Still, the biological and social sciences are large employment fields and even a
percent composition of engineers is not an insignificant number.
Mobility
Dauffenbach's results also provide ~nfonnation on the extent of mobility of venous
types: change in employer, in occupation, and in responsibilities. Table 3 provides these
gross mobility rates for the major S&E employment fields, 1976-1982. Change in
responsibilities appears to be rather frequent among the S&E categories and about 1 in 4 or
Table 3. Estimates of Mobility Rates, 1976-1982 (in percent)
Occupational Field of Employment
Mad1 & Physical Biological Social
Science
Type of
Change En~neer~ne Computer
Science Science
Employer 21.99 27.50 23.22 19.34
C
5 changed employer within the six years. Occupational mobility is substantially lower.
Those working in the math/computer employment field were the most likely to be mobile
occupationally, about 17 percent as compared to the more common 10-12 percent. These
mobility figures are derived from retrospective questions-that is, in the 1982 survey,
respondents were asked how their jobs had changed since 1976. Mobility results tend to
be substantially higher when tabulated from longitudinal data, a result that is most likely a
consequence of coding error.
These findings from Dauffenbach's study allow one to place some bounds on the
extent of mobility, its character, and earnings consequences. The results imply that there is
a fairly high degree of adaptability among engineers, math/computer specialists, and
physical scientists. This finding is validated by (1) the magnitudes of individuals having
their highest degree in one of these fields, but working in one of these other fields, and (2)
the essentially nil pay differentials among these degree fields within each respective
employment field. Other fields, especially business disciplines, contribute significantly at
time, but in general have substantially lower pay differentials. When these results are
coupled with the evidence that the majority of S&E degree holders do not work anywhere
in science or engineering, the extent of flexibility is large, indeed. A major gap in our
knowledge is the extent to which such individuals are both willing and able to return to
S&E career pursuits. This is a knowledge gap in great need of being closed.
Other Findings
Other recent studies find similar results to those of Dauffenbach and also examine
explicitly the value of an engineering degree for persons in management jobs. Korb (1987)
studied the employment of 1983-84 baccalaureate degree recipients in 1985, using a
Department of Education survey of recent graduates that included all degree fields. Not
surprisingly she found that engineering graduates reported the highest salaries and that
those with nonengineering jobs 1-2 years after graduation earned somewhat less than those
with engineering jobs. However, the engineering degree seems to be valued more highly
for the principal nonengineering jobs entered by recent engineering graduates when
compared with the earnings of other college majors in the same jobs. For example, she
found that engineers were employed as technicians after graduation much less frequently
than were biological sciences graduates (9 percent versus 40 percent), but that they reported
much higher earnings than the biological sciences graduates when they did work as
79
technicians ($22,000 versus $15,000). Only 5 percent of engineering graduates took jobs
as managers 1-2 years after graduation. They earned less than the engineering graduates
who took engineering jobs but more than business/management graduates who took jobs as
managers. In short, the Department of Education survey indicates that new engineers are
valued most highly for engineering jobs, but they are valued more highly than other majors
in all of the occupations they enter. This suggests a great deal of adaptability.
The strong labor market for recent engineering graduates is widely known' but are
persons trained as engineers as highly valued later in their careers? Evidence from another
Department of Education survey suggests that the answer to this question is "yes." James
et al. (1989) examined data from the National Longitudinal Study of the High School Class
of 1972, and used data they reported about their jobs in 1986, about 9-10 years after the
typical person In the sample had completed a bachelor's degree. Their study was
innovative in that it controlled for differences in ability as measured by SAT scores, for
differences in college grades, and for the number of math courses taken in college, all
factors that enhance earnings. Still, engineering graduates were found to have the highest
earnings in 1986, about 20 percent higher than the second highest-paid bachelor's degree
field, business. Relevant to die adaptability question is the fact that engineering graduates
who held noneng~neer~ng jobs earned salaries Hat were just as high as those with
engineering jobs. On the other hand, business majors seem less versatile: they reported
high wages only when they held jobs as managers.
We could cite other studies which document that engineers earn more when they
devote a relatively high percentage of their time to management activities (e.g., Finn,
19851. However, this is probably widely accepted and needs little discussion.
Summary
The studies cited here indicate that persons with engineering degrees do frequently
take jobs outside engineering-not only in He physical, mathematical, and computer
sciences, but also as managers and in other capacities such as sales. Salary is a simple and
imperfect summary measure of the adequacy with which they perform these jobs. Yet
findings of these studies of earnings are unambiguous. They indicate that, no matter what
the occupation studied to date, persons trained as engineers do at least as well as persons
trained in any other degree field. This might be the result not merely of the adaptability
provided by an engineering education, but also of the superior ability or willingness to
80
work hard that characterizes the students who complete engineering degrees. However, we
note that one study that was able to control for some measures correlated with native ability
and willingness to work hard (SAT scores, college grades) found evidence of the same
high degree of adaptability as did the other studies without such measures.
It would be incorrect, however, to infer that degree field doesn't matter, that all
college graduates are adaptable. The Dauffenbach study indicates clearly that persons
without engineering degrees earn less than those with engineering degrees when the job is
engineering. The one exception seems to be the math and computer science degree-
recipients. They seem to show as much adaptability as engineering degree-recipients, at
least for jobs in the broad groupings of engineering, mathematical sciences, computer
science, and physical science.
References
Dauffenbach, Robert C. 1989. The Issue of Quality in the Market for Scientists and
Engineers - A Report to the National Science Foundai'on. Shllwater, OHa.:
Oklahoma State University.
Finn, Michael G. 1985. Foreign National Scientists and Engineers in the U.S. Labor
Force, 1972-1982 (ORAU-244). Oak Ridge, Tenn.: Oak Ridge Associated
Universities.
James, Estelle, N. Alsalam, J. C. Conaty, and D. To. 1989. College quality and future
earnings: Where should you send your children to college? American Economic
Review 79(May):247-252.
Korb, Roslyn A. 1987. Occupational and Educational Consequences of a Baccalaureate
Degree. Washington, D.C.: U.S. Government Printing Office.
~1
