by

The Challenges of Gathering Data on Ph.D. Placements

We’ve saved the most complicated—and most interesting—question for our last post on findings from our informal survey for The Chronicle’s Ph.D. Placement Project: “What should we expect will be the most difficult part of gathering reliable data? How would you suggest doing it?”

Here’s what responders thought would be the most challenging:

  • Deciding what counts as an “authoritative” source: Do we gather data from Ph.D.’s themselves? From departments? From advisers? Who is most likely to respond? Both institutions and individuals may have reasons to misreport or hide information, and no one knows how many departments keep or update a comprehensive, ongoing record.

“I suspect that voluntary data collection is inaccurate because people who got the job they wanted will of course be more eager to talk about it. People who are scraping by on adjunct work will naturally be less inclined to tell former mentors and colleagues what they’re up to.”

  • Finding graduates, particularly those who are no longer in academe. Should we also search for students who drop out before earning their doctorate?
  • Maintaining a sufficiently long time frame: To get an accurate placement picture, we’ll need to capture students at all stages, including ABD, adjunct, postdoc, visiting professorship, fellowship, tenure-track, tenured, alt-ac, administrative, nonacademic, and perhaps others. This means we’ll need more than a decade of data for each cohort, as a Ph.D.’s prospects look very different upon graduation and five years later, especially as postdocs and visiting professorships have become a near-standard stop along the way in some disciplines.
  • Privacy concerns:

“Any information you gather would have to be absolutely anonymous. People currently on the job market are in extreme danger of never getting jobs and/or being blacklisted by their recommenders for voicing problems with their former departments’ handling of job placement.”

  • The distinction between departments’ roles and advisers’ roles:

“Job placement relies heavily on the role of the adviser, as well as the institution. … Who coaches students about conference attendance, publication schedules, and other essential professional activities?”

  • Standardizing data across reporting platforms. For instance, some universities use traditional tenure, while others use rolling contracts, which are considered comparable but cannot be labeled “tenure-track.”
  • Personal variation: Applicants make decisions about where to apply based on geographical preferences, family needs, research interests, and other factors, which are either hard or impossible to categorize.

It’s not all bad news, however. Responders also had some good suggestions for how to go about collection:

  • Embed basic tracking requirements in college-accreditation and public-funds opportunities.
  • Codify graduates in every entering cohort in order to capture completion rates and ABD status, in addition to placement: “Maybe there should be a ‘census’ of Ph.D.’s, held at regular intervals (annually? every other year?), with everyone having a unique ID of some kind, so they can be tracked—like migratory birds.”
  • Reach graduates through online career networks such as LinkedIn and Academia.edu, through dissertation databases such as ProQuest, and through alumni networks.
  • Keep as many data points as possible: Varying research specialties, even within small disciplines, can reflect large differences in placement rates.
  • Ask departments to publish rates of success for each graduating class: “Probably, departments can give statistics of the percentage of successful job placements among newly granted Ph.D.’s independent of familiar lists of names and jobs so that there will be no risk of ‘shaming’ those who did not get tenure-track jobs right away.”

Based on all the suggestions, we’ve come up with a couple of different pilot projects that could help us test the effectiveness of various strategies. (For example, many readers suggested using Google and LinkedIn to find graduates’ current employment status: What percentage will be findable this way in practice?) Stay tuned for more on those pilots.

Return to Top