Where assistant professors got their PhDs

[Ziz]

I found this on polisci job rumours. It's interesting...of course, it doesn't take into account all the people who didn't land TT positions or those who were awarded tenure - just those who are currently in assistant professorships.

From the other site...

Apologies if this has been gone over before. I stumbled across a dataset which purports to have counts of articles published by each assistant professor in U.S. political science departments, plus a host of control variables. The date range is a bit hazy, but I think it covers about 1993 through 2006. Being a big nerd, the first thing I did was run a negative binomial regression on the number of articles published by each scholar. IV's included number of years with PhD, a series of dummies indicating subfield, a gender dummy, and two dummy variables concerning the rank of the PhD-granting institution: one indicating that it's a top-5 school, a second indicating that it's ranked at all (both derived from US News rankings). Findings:

- Time that the AP has had the PhD is positively related to # of pubs. Duh.

- Female APs publish significantly fewer articles than males (about 1 less on average, which is pretty big since the mean # of pubs over the entire dataset is 2.5).

- American Politics faculty have the most pubs, followed by comparative, then IR, then theory. The American-IR, & American-theory differences are statistically significant, but the American-Comparative difference is not.

- Getting a PhD from a ranked school is worth about 2/3 of a publication over getting one from an unranked school, and the difference is significant.

- There is no statistically significant difference between APs getting their PhD from a top-5 school and APs getting theirs from lower-ranked schools. In fact, the coefficient on the Top-5 dummy is negative and close to being significant (p<.08, 2-tailed).

Results are robust to different operationalizations of the DV: adjusting for co-authorship; pubs/year; top-3 pubs only, etc.

I make no claims about the accuracy of the data -- in fact, I'm in there and it missed one of my own publications. But I guess it's better than no data at all. If you want to play with the data yourself, they're here:

politicaldata.org/assistingprof/assistingpolitics_Feb_2007_update.xls