self improvement

A while ago I noted that I had not read many of the “famous” articles in sociology. I knew what was in them; many I had skimmed. But I hadn’t actually taken the time to seriously read them myself (you know, taking notes on them, thinking about them, etc.). So I decided to correct this. And I read the 25 most referenced articles in ASR and AJS. I found these articles here, with the help of Jim Moody. Today I completed this task. I went from bottom to top (least to most referenced). So what did I learn?

1.) There are lots of famous articles I had never heard of. And lots of debates I didn’t know existed. Some of this is because these things are older and not referenced that much today. But more of it is because I don’t read as widely as I thought I do.

2.) If you read older articles, you find that “new” ideas (things done today) are often re-articulations of “old” ones.

3.) Most importantly, many of these articles did not actually say what I thought they did. And even more importantly, they did not always seem to say what the literature attributed to them. This is most often the case in attributing “findings” to papers that have none. Take the two most referenced papers: DiMaggio and Powell’s “The Iron Cage Revisited” and Granovetter’s “Economic Action and Social Structure”. These papers don’t find anything. Though D&P may have a series of hypotheses, they don’t have any data. Neither does Granovetter. This is not a complaint or critique. It’s just an account of my own surprise. Neither paper demonstrates much; rather they seem to say, “here’s how we might think of things…” Now clearly, they were on to something. Their ideas are incredibly useful — some of the most useful in the history of American journals. But I must say I expected something different in Granovetter than a discussion of Hobbes (via Parsons) and Williamson, or in D&P than a list of 12 hypotheses that might be looked into. If you had asked me, before undertaking this task, what it took to write a blockbuster article I would have said, “A kick-ass study.” It seems the answer is something different. “A good (or useful?) idea.” Or as my colleague Josh Whitford just said to me, “Open a question, don’t close one.”

At first I was worried that some of these articles didn’t say what I thought they did (or what others told me they did). But now I don’t worry about that so much — particularly when I remind myself that “impact” isn’t as much about an article as it is a space in the field where research is thriving (so provided that that research is going somewhere, who cares about my concern?). It did make me think that a definitive finding might not be good for citation counts. And a (ambiguous?) set of ideas might, under the right conditions, might be really good.

5 thoughts on “self improvement”

  1. You also just selected on the dependent variable, so you have no way of knowing how many articles that just propose a set of ideas (even under the right conditions) don’t reach the top 25. Further, nearly all (or maybe all, based on a cursory reading) of the articles in this list published in 1993 or later are substantive as well as theoretical. As such, the “right conditions” may not exist anymore.


    1. Or, it could just be a very high risk paper. I certainly didn’t claim that all idea papers are high impact. The overwhelming majority are probably low (or never published). It’s just that many of the highest impact papers ask more questions than provide answers. Which is interesting to me.

      As for post-93: there aren’t many to look at. It would be interesting to see… (say, look at the highest impact papers since 1995).


  2. Have you seen Mizruchi and Fein (1999)’s take on how DiMaggio and Powell (1983) has been (mis)used? It’s a fun paper, and it notes how ‘mimetic isomorphism’ is often cited as the finding of D&P ’83, even though the original paper puts equal weight on normative and coercive, and that later authors rarely try to distinguish between the three. It’s a neat analysis.

    Beyond that, I would argue that almost no one remembers that D&P distinguish “competitive” and “institutional” isomorphism – that is, they start off by saying that some times organizations converge on the same form because it’s the best for whatever system they are competing it! They then go on to make some (theoretical, data-free) arguments about why that sort of isomorphism might be less important than it used to be, and why the other sort (the institutional variety) might matter more now. But how often do you see a neo-institutionalist argue that something might be a competitive isomorphism or even try hard to disprove that null?

    Note: I’m not really an Orgs person at heart, so please correct me on the last bit if my limited and unsystematic sampling of Neo-I arguments is totally off!


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 )

Google photo

You are commenting using your Google 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.