The recent increase in the intensity of competition for jobs in astronomy has captured the attention of the astronomical community. The "oversubscription rates" for nationally-advertised postdoctoral positions and for NSF and NASA grants are high, and current projections suggest that, profession-wide, perhaps only about 1/3 of students receiving their Ph.D.'s in astronomy in the 1990s will ultimately obtain a permanent job in academia or at a laboratory or observatory, whereas the figure once was about 2/3 (estimate given at the AAS Minneapolis workshop, November 1996).
We have been here before. The cover letter to a report of the Astronomy Manpower Committee, chaired by Leo Goldberg, to the NAS in February 1975, refers to ". . . the reality that only a small fraction of the new Ph.D.'s can look forward to permanent positions in Ph.D.-granting departments, as they could in the past." The unexpected growth in job opportunities during the ensuing decade appears to have been a temporary respite, and we are now again face to face with this reality. One positive development is that a serious movement is underway to adjust expectations, educational strategies, and perhaps even norms of the profession to address and accommodate the situation. This has included a call for factual information about career success and outcomes of recent Ph.D.'s. This poster presents a summary of such information, for 78 students who earned their Ph.D.'s in astronomy from the University of Texas in the past 12 years.
The Astronomy Department of the University of Texas is one of the 30 "large" departments that produce 75% of all U.S. astronomy Ph.D.'s (NRC statistics from J. Voytuk). Texas has 20 faculty, about 12 postdocs, and is closely affiliated with McDonald Observatory, which has 12 Research Scientists of faculty-equivalent rank. The number of graduate students rose above 50 in the mid-80s, but has since settled back to 30-35. The most recent NRC rankings placed the department as number 10 (tied with two others) among U.S. astronomy departments; previously it ranked in the top 20. Thus, the track record of recent Ph.D.'s from the University of Texas should be representative of the national situation. In addition, a review by this author of the job status (in 1995) of students who entered the Texas astronomy program during 1980-1989 showed general agreement with statistics for a similar group from the University of Arizona examined by D. Kelly (see table).
During fall 1996, I surveyed all students who earned astronomy Ph.D.'s from the University of Texas from 1984 to 1995. All but five out of a total of 78 responded; information for the others was obtained indirectly, from classmates or advisors. The respondents were asked to provide professional histories as well as information about their current positions. I subdivide the sample into four-year "cohorts," as in the NRC's national surveys. This enables us to allow for "stage of career" effects as well as to compare with national statistics. Table 1 defines the cohorts, while Table 2 presents other demographic information.
In Table 3, I summarize the present employment status of all Ph.D. recipients, grouped by cohort. Perhaps the most significant results are summarized on the yellow-highlighted lines. Of students who received their Ph.D.'s within the past 1-4 years, only 27% now have "permanent" (tenured or tenure-track) positions in astronomy, whereas 59% are in postdoctoral or other temporary positions. On the other hand, fully 72% of those who received their Ph.D.'s 9-12 years ago have permanent positions. This is not unexpected, since a "typical" career path consists of earning the Ph.D., 2-5 years of postdoctoral work, followed by a permanent job. Some members of our Cohorts B and C followed this path. Others jumped immediately, or within one year, to a permanent position at a federal laboratory, national observatory, or a small college or university. The Texas sample has a significant number of students who are not U.S. nationals, in contrast (for example) to the Arizona sample; however, the Texas group may be more representative of the broader situation. This is relevant because citizens of other countries often return to their home country with a high success rate at obtaining faculty or permanent observatory staff positions there (see second line of Table 3).
The progression across this table is presumably dominated by the "stage of career" factor. In order to address the question "Has the job situation gotten worse for astronomy Ph.D.'s in the last 12 years?" we need to compare the cohorts at the same stages of their careers. We turn to this comparison next.
In Table 4 and Table 5, I have compared the Texas cohorts at equivalent stages of their careers. For example, in 1996, the students in Cohort A were 1-4 years post-Ph.D., and so should be compared with Cohort B as they were in 1992, and Cohort C evaluated in 1988. In addition, I have interspliced these columns with national statistics from the NRC, as presented by J. Voytuk at the Minneapolis AAS workshop on Graduate Education. For ease of comparison, I present percentages only, and have grouped the respondents into fewer categories than in Table 3. Also, the national statistics do not separate faculty positions at U.S. academic institutions from those in other countries, so I have combined these categories for the Texas cohorts as well.
In comparing the columns, one must bear in mind that the Texas sample is relatively small, especially once subdivided. The presence of a few students from other countries who graduate to immediate faculty positions can have a strong impact on the statistics. However, a number of trends are notable. In Table 4, the fraction of Ph.D. recipients holding postdoctoral positions ranges from 36 to 61%. This can be compared to the proportion of U.S. physicists holding postdocs six months after receiving their Ph.D.: 63 ± 1% for each year in the period 1990-1994 (AIP statistics). While the total fraction of astronomers in temporary positions has not changed significantly from 1988 to 1996, the total fraction who have permanent positions in astronomy has declined, and the line "industry and other" has increased.
In the "canonical" academic career path, the crucial period for the transition from postdoc to permanent position is 5-8 years after the Ph.D. In Table 5, we compare the two Texas cohorts and two national groups at this stage of their careers. We see a slight decline in the fraction achieving permanent positions (from 44% to 39%), and an increase in those still holding temporary positions (from 30% to 39%). I might note that here are a few individuals in the Texas sample who finally obtained a permanent job in academia or government labs after holding a Ph.D. for 8-10 years. There are others who left astronomy, mostly to work in computer-related industries, after holding postdoctoral and other temporary positions in astronomy for a similar period of time. In addition, there are a few cases of persons electing to change their "domain," such as leaving a permanent government laboratory position or a tenure-track faculty position in order to start their own companies, work in industry, or live in a different country.
At the beginning of this poster I
asserted that the Texas sample was reasonably representative of
the national scene. The national scene is, after all, directly
reported by the NRC statistics. However, these do not separate
out certain distinctions in employment which may be important,
such as U.S. vs. home country faculty positions (for non-U.S.
nationals). Table 4 suggests that the Texas group may have, on
the whole, fared better than average, but this may be weighted
by the relatively large contribution of "successful"
non-U.S. nationals. This group includes students from Korea, Canada,
Mexico, China, Brazil, Argentina, and the United Kingdom.
What exactly do we learn from these statistics? Of course mere numbers do not convey such important factors as the degree to which these individuals are satisfied with, or frustrated with, their current employment situations, and whether they feel that their original career expectations have been met. Another topic of current debate is whether graduate training in astronomy should be modified or changed in any way, in order to prepare students better for the situation that they will face upon receiving the Ph.D. I hope to follow up this basic "career facts" study with another survey that addresses some of these issues which are certainly important, but are more difficult to quantify.
A few weeks after this meeting, I intend to post the contents of this poster on the World Wide Web. My home page is:
I would appreciate any comments that
you have on this poster, as well as suggestions about additional
ways to examine and compare the data that is summarized here.
Acknowledgment: I greatly appreciate the helpful replies and cooperation of all the individuals who participated in this survey; they made it possible for me to prepare this summary, and many also sent informative and thought-provoking comments on the job situation in astronomy, from which I have learned much.
return to Harriet Dinerstein's main page