Sunday, February 16, 2014

The Great Poli Sci Portfolio

In the dustup produced by Nick Kristof, one of the basic misperceptions keeps being repeated--that the American Political Science Review is not influential or readable enough.  The job of the APSR is not to be read by policy-makers but by political scientists.  Really?  Yes.  Let me explain.

The academic journals have their place in the profession just as those aimed at outreach do. Why have academic journals?  So that political scientists can do the science that is necessary for the generation of knowledge:
  • developing arguments in reaction to and building on pre-existing work (that would be the dreaded lit review), 
  • articulating casual mechanisms that link potential causes to that thing we are trying to explain (that would be theory), 
  • specifying how we think we know what we know (that would be methods)
  • drawing out the findings and their implications.
It is on the last part that really interests folks outside of the research university.  But to get there, we have to a heap of work.  We cannot just speculate about the future, like Sam Huntington did with Clash of Civilizations because then it is all based on air.  And certainly probably wrong. Doing the preceding stuff is what we do so that our stuff is not just speculative b.s. but based on the best approximation of science (not all of us have that as an aim, of course).

This research is fundamental to being a professor at a research university and pretty relevant to folks in other academic positions since teaching and research are actually related.  Not everyone doing this research has to be articulating their stuff to the folks beyond the university, but some people must be doing it.

And that is what has changed over the past fifteen years or so.  Social media have greatly enabled those who are interested in transmitting beyond the academy and are able to do so.  Blogging, twitter, and all the rest complement the more traditional means--policy-oriented journals, media appearances, public speaking, private networking that always existed. Indeed, as I have insisted elsewhere, there may be fewer famous public intellectuals, but in poli sci, they have been replaced by a much more diverse set of folks articulating the findings and implications--diverse methodologies, different ages, genders, races (although we fall short there still), theoretical perspectives and so on.

In the previous post and elsewhere online, I argued that blogging and such can be done by junior faculty but that they should still be judged for tenure mostly by the traditional standards of publishing peer-reviewed research.  Why?  Because that is a key part of developing one's credentials as a scholar--that one can do research, that the research can pass through the vetting process of academic journals, that it is making a contribution.  And they have enough stuff to do.  Once one is tenured, it is my belief, one has more responsibility to disseminate more widely (although again with a portfolio approach, there is no one way to do this and not everybody has to do it the same way).

The Kristof article has many problems, including having a pretty old and outdated view of the profession, but also downplaying the reality that research is necessary (pundits do not have to research but at least some political scientists have to).  Our research process could be more open (open access would be great, although who wants to read heaps of academic articles), but we need to have multiple conversations.  One of the key conversations is among political scientists about the research we do, which is then the basis for whatever it is we want to transmit to the folks outside of political science.

 Update: I forgot to mention that we have some evidence about how diverse IR is these days.  I wrote a piece using the TRIP data that showed that our circus has a very big tent with all kinds of folks doing all kinds of things.  Yes, there is more quant work now, but not less qual work.  More outlets means more articles--so quant has made a relative gain but qual has not made an absolute loss.


Brandon Valeriano said...

Come on, this is good, Duck it.

But on Huntington, I don't think we cannot make predictions about the future. We can, as long as they are grounded on solid theory, evidence, and a comprehensive examination of the recent past. There is where Huntington got it wrong because he was not even right about the near past and maintained that not even even the Iraq War was a Clash of Civ.

Theo McLauchlin said...

I actually think that more policy advice in APSR would not just be unnecessary: it would make matters worse. It would play out as still more incentives to offer crappy "policy recommendations" building on one's findings. I say crappy because any article only generally focuses on one aspect of reality, as a necessary simplification. The policy advice in an article often seems to correspond to the simplification so that the author can better show how their work "matters." But that's not necessarily good for policy advice. Most often, when your finding is limited, such thinking is of the hedgehog not the fox.

This isn't limited to academic articles. I think it's a wider habit of thought in the discipline. For example, a Washington Post op-ed in 2006 by a political scientist at a leading policy school noted that civil war recurrence is less likely after victory, so, in Iraq, the US should pick a winner, specifically supporting the Shiites to beat the Sunnis and Kurds. The finding is solid, the advice? Terrible.

But I do think this is a habit of thought promoted by the "policy advice" section of journal articles, because these create incentives to sell simple policy advice to sell the "relevance" of the small thing your article is about.

So encouraging more policy work in APSR would wind up encouraging this habit still more. Worse, it might influence what kind of findings get published in it: those with more obvious policy advice. I think the kind of model we have now--with blogs and other outlets debating implications of poli sci research on current events--is far superior: it's detached from any one idea, so you stand back and survey the field, and your interlocutors can respond quickly.