How often do TT ecology profs get hired near where they did their PhDs?

More often than you (or at least I) might’ve thought. For the details, read on.

The TT faculty job market is competitive, in the sense that, at any moment in time, TT job seekers outnumber TT jobs. In ecology in the US and Canada, something like 33-50% of TT job seekers eventually obtain a TT job. For that reason, TT job seekers often are advised to apply widely, without any geographical restrictions. But that’s tough advice to follow for many people. Many people want to live in a specific city, state, or geographic region, for all sorts of reasons. So, as a TT job seeker, you might wonder: how common is it for geographically-restricted TT job searches to pan out?

I’m afraid I can’t answer that question. But I do have the data to address a related question: how common is it for recently-hired TT ecology profs to have done their PhDs near where they were later hired?

Obviously, not all faculty job seekers–even those doing geographically-restricted job searches–try to remain near where they did their PhDs. And even job seekers doing geographically-unrestricted searches sometimes get hired near where they did their PhDs. Especially since the places where lots of people get PhDs are the same places that tend to have lots of TT faculty jobs nearby. But still, if it turns out to be more common than expected by chance for ecology faculty job seekers to get hired near where they did their PhDs, that gives some reason to think that doing a geographically-restricted job search is fairly common and has some reasonable chance of panning out.

To address this question, I went back to my nearly-comprehensive list of TT asst. professors in ecology and allied fields, hired in the US and Canada during the 2017-18 job season. 158 of the hires I ID’d did a PhD in the US or Canada.* I checked how many of those people were hired into a TT position in the same state or province as where they obtained their PhD, and how many were hired into a TT position in an adjacent state or province (i.e. sharing a border). I didn’t count Canadian provinces as adjacent to US states, or vice-versa, for purposes of this exercise. Moving to a different country from where you did your PhD is a big move, no matter how far it is as the crow flies.

Obviously, “same or neighboring state/province” is a very crude definition of “near”, that won’t reflect the geographic restrictions that some job seekers use. For instance, someone who did a PhD in Maine, and wanted to stay in New England long term, could well accomplish that without taking a TT job in Maine or an adjacent state. But as you’ll see, even this very crude definition of “near” isn’t so crude as to prevent us from picking up any signal in the data.

16% of those 158 TT ecology hires were hired in the same state as where they did their PhDs, and 13% were hired in an adjacent state. So in total, 29% (48/158) were hired “near” where they did their PhDs.

Now, I haven’t done a permutation test to check whether that 29% is significantly higher than you’d expect by chance if TT hiring was geographically random. But I’m pretty sure it is. For instance, this dataset includes 15 people who got PhDs in California, 4 of whom got TT jobs in CA. So, 27% of CA PhDs got TT jobs in CA. That’s even though only 8% (13) of the 158 TT jobs in this dataset were CA jobs. And of those 13 CA jobs, 5 (38%) were filled by people with PhDs from CA or a neighboring state, much higher than the 16% of people in this dataset who got their PhDs from CA or a neighboring state. And it’s not just CA. 3/9 people with PhDs from Colorado went to TT jobs in CO or adjacent states, much higher than the proportion of all TT jobs located in CO or neighboring states. 3/6 Oregon PhDs got TT jobs in OR or neighboring states. 2/5 Florida PhDs got TT jobs in FL. 3/3 Quebec PhDs got TT jobs in QC or neighboring provinces. 2/3 Alberta PhDs got TT jobs in AB or neighboring provinces. Ok, not every state/province I checked worked that way (e.g., 0/7 New York PhDs got TT jobs in NY or neighboring states). And individually, each of these states and provinces provides a very small sample. But the consistency of results among states/provinces strongly suggests a non-random geographic association between where recently-hired academic ecologists did their PhDs, and where they landed TT jobs.

Presumably, that association arises because an appreciable fraction of successful TT ecology job seekers run geographically-restricted job searches. It does not arise because of universities hiring their own PhD graduates: almost nobody gets hired at the same institution where they did their PhD. And it mostly doesn’t arise because universities hire internal candidates, because they hardly ever do. And it doesn’t arise because faculty search committees prefer applicants from the same state or province, because come on, that’s not how search committees work.

One other, speculative tidbit: 70% of the 46 people in this dataset who landed TT jobs “near” where they got their PhDs were women. That’s as compared to 61% women among all 158 TT hires in this dataset. The overrepresentation of women among people who took TT jobs “near” where they got their PhDs could well be a blip; the sample sizes are modest. Or, it could be an indication that women faculty job seekers in ecology are more likely than men to run geographically-restricted job searches. Comments welcome on this.

Finally, it’s worth emphasizing that this dataset only includes people whose TT job searches were successful, not people whose TT job searches were unsuccessful. For that reason and others, you can’t use these data to estimate how your own odds of landing a TT position will be affected by any geographic restrictions you’re operating under. All I can tell you is that, if you’re running a geographically-restricted TT job search, you’re far from alone in that–an appreciable fraction of successful TT ecology job seekers apparently did the same.

As always, I hope these data provide some useful context to ecology faculty job seekers and anyone advising them.

Looking forward to your comments.

*I looked at PhD location rather than, say, postdoc location only because I didn’t record postdoc locations.