Skip to content

Golden fleece sought, not found

Lee Sigelman writes that Senator Coburn of Oklahoma is proposing to zero out funding for the National Science Foundation’s political science program. It’s hard for me to believe this will even come close to happening, and conflict of interest prevents me from saying anything at all trustworthy on the subject–I’ve had nearly continuous NSF funding for the past 23 years–but I’ll tell ya this: I clicked through to Sen. Coburn’s list of NSF-funded projects that he’d like to cut, which included:

– $91,601 to conduct a survey to determine why people are for or against American military conflicts.

– $8,992 to study campaign finance reform, with the stated intent of providing “a basis for assessing future proposed changes to campaign finance regulations.

– $958 for a direct mail survey of the residents of Celebration, Florida regarding their feelings of living in privately operated city.

The first two topics seem pretty important to me. And, as for the third . . . hey, it was only $958! Maybe that was a typo? $958 doesn’t seem like a real NSF grant.

I can see a lot of good arguments for cutting NSF funding: For example, as Sen. Coburn’s press release points out, we could do without the National Election Study–there are a lot of other polls out there. And lots of NSF money goes to university professors, who are mostly Democrats. I can see why a Republican senator might not like that. Beyond this, many would prefer government spending to be reduced for all purposes, and the National Science Foundation is just a particular example.

But, really, the list of “wasteful projects” seems pretty lame to me. Golden Fleece material, it ain’t.

I also wonder why Sen. Coburn is picking on political science. His main point seems to be that this isn’t real science like physics or geology, and it’s not anything helpful like medical science. But then why isn’t he talking about shutting down sociology, economics, etc etc? Or is poli sci just the first step? Maybe it’s just a publicity grab, but if so I think he could’ve found some better examples of wasteful funding than the examples above. Isn’t there some Federal funding for transgender theater or something like that, something more headline-grabbing than a study of campaign finance??

P.S. No, I don’t think that the National Election Study’s funding should be zeroed out. I think we learn a lot from having this regular survey. And, especially with all the difficulties nowadays with survey nonresponse, it’s good to have some baseline surveys such as NES and the General Social Survey that put in lots of effort to reach people.


  1. Anonymous says:


    I think his argument is just that the marginal product of political science is lower than (say) medical research, and so he wants federal research funds directed there. If the total budget of NSF stays fixed, do you think the value of political science research is above the (marginal) medical research done?

  2. Andrew Gelman says:

    I don't know. A lot of medical science research money is wasted too. I think of it as a portfolio kind of thing. Much much more money is spent on medical research than on political science research, which makes sense. But maybe it doesn't make sense to send political science research to zero. I mean, if you want to go this route, you could say that there's still hunger in the world so we shouldn't be funding any research at all. Maybe so, but that's not a rule we apply in many other aspects of life.

    In any case, as I noted above, I'm too compromised to be in any position to make a strong argument in favor of NSF funding. I just thought it was interesting that the senator's office couldn't come up with some items that sounded more boondoggly.

  3. Bob Carpenter says:

    I don't think a conservative (i.e. non-Bayesian) senator's going to like this one:

    Estimating Congressional District-Level Opinions from National Surveys using a Bayesian Hierarchical Logistic Regression Model

    But maybe their anger will be mitigated upon learning the government will get the money (US$12,000) back after the awardee uses it to "…purchase data from the US Census Bureau to obtain reliable estimates of the joint population distributions of sex, ethnicity, age, and education at the congressional district-level".

  4. NSF seems to have been a bit lax recently in its IT practices, and there's always someone who will try to make hay at your expense. My guess is that Coburn couldn't find the "viewing porn on the internet" line item in NSF's budget, so he picked a convenient target. (