get a theory, would ya?

I’m currently in midst of writing reviews for 21 NSF proposals (which need to all be filed by 5:00 tomorrow before I head to the airport to fly out to Washington for the Sociology Panel…argh). Having read most all of them at this point, I can make a couple of recommendations to future writers of NSF grants, but I’ll limit myself to just one in this post. This same recommendation flows just as much from my experience editing the journal Mobilization and from watching innumerable job talks over the last decade. That advice is: get a theory.

The biggest problem with the vast majority of these grant proposals is that it is not apparent that many of the authors have a theoretical project at all, or if they even know what that means. Likewise, this has been (without a doubt) the biggest gripe from reviewers of Mobilization manuscripts over the past year-and-a-half: Where’s the theory? And, it is the number one concern raised about faculty candidates (both junior and senior) in our department since I’ve been involved in faculty searches: “The empirical stuff seemed ok, but I couldn’t figure out where s/he was going theoretically.” That question is the kiss of death.

So many candidates seemed oblivious to the importance of talking about the theory, that when I was chair, I took to tipping off candidates when I invited them for interviews. I’d say something like “make sure you spend a significant amount of time in your talk discussing your theoretical contribution. It’s possible we may over emphasize it here, but it is critical to get that theoretical piece across to the group.” Mostly, it didn’t work–either people didn’t incorporate anything about theory or they’d append a slide to their power point called something like “Theoretical Issues” and blow through three bullet points suggesting possible theoretical links with existing literature.

That’s bad enough for a junior faculty candidate, but I can’t imagine going into a highly competitive grant competition like NSF Sociology* and expecting to get anywhere without communicating very clearly to the reviewers and panel what you think the theoretical payoff is going to be, or what theoretical hole you are trying to fill. It’s true you don’t know for sure because you haven’t finished the work yet, but you need to have an agenda at least and show that you know (a) that there is some theory already out there and (b) why what you are doing might be important to the developing theoretical trajectory of the field. When rating NSF proposals, one has to label each as Poor, Fair, Good, Very Good, or Excellent. Without a strong theoretical basis or upshot, I feel that Fair is a pretty generous rating.

In contrast, I just got done reading a really good one from a very experience grant getter and, boom! By the time I got through the second paragraph of the project summary (the one page abstract of the project), I knew exactly where the project was going theoretically and why it was important.

Honestly, it’s even worse when one is submitting an article. With a grant proposal, there’s always some degree of speculation about what will come, but an article is a finished project. You had better give the reader a clear sense of the theoretical contribution or you’re dead in the water. I am uncomfortable, as an editor, issuing an R&R decision and having to ask the author(s) to develop some theory. How this can be an afterthought is beyond me–it should have been driving the whole paper forward. I have done it on occasion, but invariably, I have been disappointed with the results. If the author didn’t take it seriously in the first place, it unlikely to emerge in the revision either.

* There are obvious exceptions like the GSS which are large data-gathering enterprises that are expected to serve many different needs, so of course there is not a coherent theoretical vision driving that kind of project, but the vast majority of NSF proposals are not of that ilk at all.

10 thoughts on “get a theory, would ya?”

  1. But a proposal with all theory and no clear statement about what the research actually will be is not good, either. I found that to be a bigger problem when I did NSF reviews. Maybe we don’t disagree: for there to be a projected theoretical payoff, you need to have both a theory and a clear idea of just what you are really going to do in the research.


  2. Oh so true! I agree completely. My experience of reading these suckers though is that the sans-theory style dominates–at least in this round. That’s right people, you have to do it all!!! :)


  3. I want to put in theory. I really really do. But I feel like my graduate program has let me down in that there is a dearth of discussion on that topic in favor of massive amounts of quantitative research methods coursework.

    Let’s say for the sake of audience that someone doesn’t do well at this piece; that somehow they haven’t learned. How would you advise them educating themselves?


  4. Sociology lacks theory?! As a political scientist, I find this bizarre. We steal theories from sociology all the time.

    In the quantitative coursework, there is usually a claim that X –> Y. To answer the question, “Why does X –> Y?”, you need a theory, no. Some quantitative work fails to do this, but if you ponder the “why” question, you should get somewhere.

    Peruse ASR, follow the footnotes.


  5. Re putting in theory, it does not mean tracing everything back to the 19th century. (Not saying you thought that, just starting where some students misunderstand.) And there are lots of competing schools of what theory means & how to do it. But how I think of it is that you are trying to explain how or why things work the way they do. So you are not just saying “X is a significant predictor of Y” but, as clball said, you are embedding that empirical relation in a larger context of explaining how something in the social world works.

    The lit review is not just an inventory of everyone else’s previous findings, but structured as an argument about what is important that has been found by others and what new contribution you will be able to add.

    I would also agree that ASR/AJS give you some idea of what that means, or rather, the wide variety of things that can mean. You won’t go far wrong if you think of a successful grant proposal is basically the front end and methods section of a successful research article. I always encourage my students to come up with some preliminary data, no matter how trivial, because it lets you give an example of why your findings are likely to be important and helps give a clear shape and argument to the proposal.

    ASA also runs workshops at the convention on grant proposal writing. I have not gone to them, so I can’t say whether they are good.


  6. In our current search, one of the questions we included in our phone interview protocol was something like, “What broader ideas or themes in sociology do you see your research fitting into?” The question took some people by surprise; others had obviously thought about it already.


  7. There’s nothing secret about the composition of panels for federal granting agencies, is there? For NIH grants that I’ve been involved in, the people who were the study section that reviewed the grant have always been known (although you don’t know which particular persons were the focal reviewers for your grant).


  8. Interesting. I never thought of that because I’ve known most years who was serving on the panels. It doesn’t do much toward revealing the authors of specific reviewers anyhow because you only review a fraction of the proposals. It’s sort of like knowing who is on the editorial board of a journal. We were asked to sign a bunch of government legal junk about conflicts of interest and swearing we wouldn’t reveal the content of proposals or pass them around, but nothing about keeping our secret identities secret.


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: