Friday, June 3, 2016

What is Risky For Professors?

Faculty governance?
Public engagement?
Working in government or somewhere else that is not academia?
Investing in long term projects that might not work out?

There is an interesting piece that studies the work of economists (and finance folks) and finds that they do their best work (most publications, most cited stuff) before tenure.  This is problematic because one of the major justifications for tenure is to give professors the chance to do risky work that may not pay off in the short run but could be beneficial in the long run.  Job security is supposed to free professors from playing it safe. 

They definitely find something:

Their key measure for risky work is citation.  Huh?  This is strange because gambling on risky stuff should produce fewer citations, not more, as risky gambles, by definition, are low probability exercises, right?*  That they should fail far more often than they succeed.  Maybe I don't understand the methodology of the paper (yes, I read it), but their focus is on average outcomes, that average citations and home-run averages decline after tenure.  Perhaps because many professors are working on projects that are less safe? 
* While people probably overrate citation cartels, clearly there are strategies to maximize citation that are all about minimizing risk.  So, I am very puzzled by this indicator.

The second problem is, yes, perhaps the average person does publish less stuff and less quality stuff after tenure, but tenure is to protect the ambitious, not the average?  

The third problem is that junior professors at many places are protected from doing as much service and are also seen as less desirable supervisors, so the pressure of tenure and the reduced service load may matter here in producing more work AND make it harder for tenured professors to keep up the old pace as they serve in a variety of capacities that they had mostly dodged before tenure.  The authors take that into account but then make assumptions about how this affects the gambling of professors--that those doing more service would "presumably reduce effort on low-impact projects (p.17).  Really?  Isn't tenure again about freeing up professors to gamble, so why assume that professors would gamble less?  Didn't we political scientists get "gambling for resurrection" from the economists?

Econ is an article business as far as I can tell.  In poli sci, it varies, but it might be that folks invest more in book efforts after tenure as they can take the time to do longer projects with no six year clock above their head.  The book game is, in my experience, less painful than the article game, so maybe profs past tenure change how they publish? 

Or maybe they do other stuff that is risky but cannot be measured in citations: participating in faculty governance where one pisses off deans and provosts; spending sabbaticals focused on improving teaching rather than getting heaps of articles out (not that I know of many/any folks who have done this); doing more public engagement, which can attract the ire of politicians and boards of regents/trustees/governors/whatever.

I am not denying that incentives matter.  I do wonder if professors in unionized shops have more or less decline after tenure than those in non-unionized shops, as the former tend not to have merit increases that could encourage tenured profs to keep on publishing. I also do not deny that there are deadwood associate professors who get tenure and then kick back.  The thing is that such folks are scorned, and grad school and pre-tenure life has a powerful way of socializing folks into developing certain patterns.  Perhaps it is my confirmation bias speaking, but I really have not seen my friends and colleagues tune out much.  To be sure, the previous generation certainly had a significant share of such folks when research expectations were lower. 

Again, I come back to the basic question that this article raises but mostly misses: what the hell is risky behavior?  How do we measure it?  All I know is that measuring it by citations seems strange.  Citations are a blunt but not awful measure of productivity, but of gambling?  Not so much. 

I have already navel-gazed this week about my own research trajectory, but I didn't evaluate then whether I was gambling more later than earlier.  I do think that shifting my research from ethnic conflict stuff, where I made my name, to civil-military relations and alliance dynamics could be seen as a gamble and one that paid off (the 2012-2016 peak).  Was I more willing to gamble after tenure?  After I finished my second book and felt as if I had earned Full-ness even if my full colleagues did not see it that way?  After I moved out of that place?  Probably not.  The big change in my career has been more public engagement since I got tenure and especially since 2009 when I started blogging and tweeting.  Was that due to tenure?  Probably. 

1 comment:

Kindred Winecoff said...

"This is strange because gambling on risky stuff should produce fewer citations, not more, as risky gambles, by definition, are low probability exercises, right?"

This really depends on the shape of the distributions and how fat the tails are. Otherwise I agree!