We present an in-depth portrait of the training, careers, and work of recent Ph.D. physical scientists. Use of specialized training varies widely, with about half often using knowledge of their Ph.D. specialty area in their jobs. The use of specialized training does not, however, correlate with job satisfaction. In this and other important measures, there are relatively few differences between “academics” and “nonacademics.” Important job skills for all employment sectors include writing, oral presentation, management, data analysis, designing projects, critical thinking, and working in an interdisciplinary context. Rankings given by respondents of graduate training in some of these skill areas were significantly lower than the importance of these skills in the workplace. We also found that the rated quality of graduate training varies relatively little by department or advisor. Finally, although nonacademic aspirations among graduate students are fairly common, these do not appear to be well supported while in graduate school.
REFERENCES
1.
E. Boyer, Scholarship Reconsidered: Priorities of the Professorate (Princeton U.P., Princeton, NJ, 1990).
2.
A. E.
Austin
, “Preparing the next generation of faculty
,” J. Higher Educ.
73
(1
), 94
–122
(2002
).3.
Graduate Education and Postdoctoral Training in the Mathematical and Physical Sciences (Mathematical and Physical Science Directorate, National Science Foundation, Arlington, VA, 1985).
4.
Characteristics of Doctoral Scientists and Engineers in the United States (National Science Foundation, 1997), NSF 00-308.
5.
The 1997 SDR contained an additional question asking the extent to which the respondent’s doctoral education adequately prepared them for their career in a number of skill areas. These data have not yet been published.
6.
Starting Salaries of Chemists and Chemical Engineers (American Chemical Society, Washington, DC, 1993).
7.
G. Blake, “Skills used in the workplace: What every physics student (and professor) should know,” paper presented at AAPT summer 1995 meeting, Spokane, WA, 1995.
8.
D. Rosdil, What are Masters Doing? (American Institute of Physics, College Park, MD, 1996), Pub R-398.1.
9.
About 200 of the respondents had obtained a Ph.D. degree and 328 had master’s degrees in some field (from Ref. 7).
10.
M.
Nerad
and J.
Cerny
, “Postdoctoral patterns, career advancement, and problems
,” Science
285
, 1533
–1535
(1999
).11.
These surveys were conducted by scientific societies and coordinated by the CPST (see www.cpst.org).
12.
The sample was limited to those who resided in North America at the time of the survey. The eight universities included both public and private institutions. The major geographic regions of the US were represented.
13.
For further project information and a copy of the survey see <http://spot.colorado.edu/∼phdcarer>.
14.
The response rate for the mail portion of the survey was 55%. Most of the additional responses were to a phone version of the survey that omitted some questions. The response rate does not include 11% of the cohort who resided outside of North America and another 11% for whom we did not have a valid address. The survey reference date was October 10, 1998.
15.
The boundaries between these categories are fluid. Respondents within the same institution sometimes placed themselves in different categories.
16.
SESTAT on-line database (NSF 2000), <http://srsstats.sbe.nsf.gov/>.
17.
It is difficult to compare these figures directly, particularly because of our addition of “Government Research Laboratory” as a separate employer category. Several large laboratories that employ physical scientists are operated by universities and SDR respondents there may have chosen “university” as their employer. If half of the respondents in our government labs category were moved into our university category, the distributions would agree to within a few percent.
18.
The generalizability of our results is also indicated by the similarity in skill use across sectors, that is, Table VI and accompanying text.
19.
Quoted phrases represent direct quotes from the wording of either the question or answer scale. Descriptions not in quotes are paraphrased or inferred.
20.
Only one of the nine departments in our study with a sufficient number of responses to test had a statistically significant difference in graduate satisfaction. We did find that “other” physical scientists, that is, not chemists or physicists, as a group had a higher level of satisfaction with their graduate experience, a result that was significant at the 5% level (chi-squared test).
21.
Of the 8% who reported the highest level of pragmatic motivations for graduate school (obtaining a Ph.D. was a “pragmatic goal,” answer scale =5), one-third were dissatisfied with their graduate experience (p<0.1%; chi-squared). No other category on the pragmatic versus intrinsic scale, that is Table II, showed a statistically significant difference.
22.
Removing those who reported the highest level of pragmatism, there was no difference in overall satisfaction with graduate school between those who had “definite” career goals at the beginning of their Ph.D. and others in the sample. Those with definite career goals, however, were somewhat more likely to be “very satisfied” with their graduate education.
23.
M. Nerad, D. Gupta, and J. Cerny (private communication). For those fields with employment patterns closest to those of physical scientists, the data from Ref. 10 find that 65%, 53%, and 36% of graduates from biochemistry, computer science, and electrical engineering, respectively, reported academic aspirations at Ph.D. completion.
24.
The overall stability of career goals with time in the data from Ref. 23 can best be tested by examining those individuals who had specific aspirations at Ph.D. start. Here there was only a slight decrease in the overall percentage who held academic aspirations at Ph.D. completion (76%) as compared to the start of the Ph.D. (81%). Some category crossing did occur, with individuals who initially had nonacademic goals changing to academic goals and vice versa. The largest change, however, was due to the 28% who initially had no specific goal (and, in this question format, therefore did not specify either an initial academic or nonacademic goal), most of whom developed some more specific goal by the time of Ph.D. graduation.
25.
M. Neuschatz, P. J. Mulvey, and S. Nicholson, “Physics and Astronomy Senior Report: Class of 1998,” AIP Pub No. R-211.30, AIP, College Park, MD, 1999, <http://www.aip.org/>.
26.
If our results for academic aspirations are divided into two categories, they split at 53% (= the sum of categories [5] + [4] + 0.5[3], see Table II).
27.
C. M. Golde and T. M. Dore, “At Cross purposes: What the experience of today’s doctoral students reveal about doctoral education,” A report prepared for The Pew Charitable Trusts, Philadelphia, PA, <http://www.phd-survey.org/>.
28.
When chemistry graduate students were asked about interest in faculty jobs with a three category scale (“not at all,” “possibly,” “definitely”), Golde and Dore found 31%, 49%, and 20%. respectively. For chemistry graduates, on our five-point scale (see Table II) we found: 25%, 22%, 13%, 24%, 15%, which appears to be consistent with the findings of Golde and Dore—given the difficulty of comparing disparate answer scales.
29.
The full text was “A temporary position is defined as one that had an end date or fixed term at the time you were hired. This position includes post-docs, fellowships, visiting faculty, internships, etc. Please mark tenure track positions as permanent.” This approach is similar to that taken in the surveys of recent graduates in Ref. 11. Note that we did not use the term “post-doc” in our analysis. There is no standard operational definition of a “postdoctoral position,” particularly outside of academia, and some might consider this term pejorative. Our qualitative analysis indicated that the definition used here for “temporary position” would most accurately reflect the varied situations of the respondents.
30.
Only half of those employed in four-year colleges or universities who were not on a tenure track reported a high (answer four or five) level of job stability as compared to 90% of those tenured or on the tenure track.
31.
We did not see a difference between men and women in this percentage.
32.
The “none of the above” category did not appear on the survey, but consists of those who did not report that they “often” used knowledge in any of the previous categories.
33.
The quality of graduate training scale ranged from 1 = “None'’ to 5 = “Excellent.” The importance to work scale ranged from 1 = “Do not do” to 5 = “Very Important.” In Table V, responses of 1 and 2 = low, 3 = medium, and 4 and 5 = high.
34.
Deborah
Olsen
and Mary Deane
Sorcinelli
, “The pretenure years: A longitudinal perspective
,” New Dir. Teaching Learning
50
, 15
–26
(1992
).35.
Howard R. Bowen and Jack H. Schuster, American Professors: A National Resource Imperiled (Oxford U.P., New York, 1986).
36.
The similarity with responses for the graduate satisfaction question (Table I) for the dissatisfied and “neither” categories is largely fortuitous. Most respondents did not answer the same way on both questions in these categories.
37.
The parameters of a logistic regression model are estimated using the maximum-likelihood method, with estimates produced by an iterative algorithm. The coefficients produced by logistic regression analyses can be interpreted as the change in the log odds of the dependent variable associated with a one-unit change in the independent variable. To aid interpretation of the coefficients, we also present the inverse natural log of the coefficients, which results in estimates of odds rather than log odds. The odds ratio, therefore, tells us how a one-unit change in an independent variable increases (or decreases) the odds of the event occurring.
38.
W. H.
Rogers
, “Regression standard errors in clustered samples
,” Stata Tech. Bull.
13
, 19
–23
(1993
).39.
D. J. Brogan, “Pitfalls of using standard statistical software packages for sample survey data,” in Encyclopedia of Biostatistics, edited by P. Armitage and T. Colton (Wiley, New York, 1998).
40.
We note that all of these variables depend on perceptions, because our data are self-reports from respondents. As in any survey work, we implicitly assume that respondents, on average, can consistently report quantities such as “intellectual challenge.” The second set of variables, however, are direct measures of respondent’s perceptions and attitudes while the first set of variables can, at least in principle, be objectively measured.
41.
The latter result can be interpreted as follows: those graduates who have done the most data analysis work since graduation are less satisfied with their jobs, all other factors being equal.
42.
E. A. Locke, “The nature and causes of job satisfaction,” in Handbook of Industrial and Organizational Psychology, edited by M. D. Dunnette (Rand McNally, Chicago, 1976), pp. 1297–1349.
43.
R. D. Arvey, G. W. Carter, and D. K. Buerkley, “Job satisfaction: Dispositional and situational influences,” in International Review of Industrial and Organizational Psychology, edited by C. L. Cooper and I. T. Robertson (Wiley, Chichester, 1991), pp. 359–383.
44.
Y.
Ting
, “Analysis of job satisfaction of the federal white-collar work force: Findings from the survey of federal employees
,” Am. Rev. Public Admin.
26
(4
), 439
–456
(1996
).45.
46.
47% of the general population report that they are “very satisfied” with their jobs. (Source: On-line data extraction, variable SATJOB, <http://www.icpsr.umich.edu/>.)
47.
V. S.
DeSantis
and S. L.
Durst
, “Comparing job satisfaction among public- and private-sector employees
,” Am. Rev. Public Admin.
26
, 327
–343
(1996
).48.
Daryl E. Chubin, S. Tobias, and K. Aylesworth, Rethinking Science as a Career: Perceptions and Realities in the Physical Sciences (Research Corp., Tucson, AZ, 1995).
49.
Reshaping Graduate Education (National Academy Press, 1995).
50.
Peter S. Fiske, Put Your Science to Work (American Geophysical Union, Washington DC, 2001), p. viii.
51.
Administered by the ACS Committee on Professional Training (CPT), <www.acs.org>.
This content is only available via PDF.
© 2002 American Association of Physics Teachers.
2002
American Association of Physics Teachers
AAPT members receive access to the American Journal of Physics and The Physics Teacher as a member benefit. To learn more about this member benefit and becoming an AAPT member, visit the Joining AAPT page.
