the pager paper, sociological science, and the journal process

Last week, we discussed Devah Pager’s new paper on the correlation between discrimination in hiring and firm closure. As one would expect from Pager, it’s a simple and elegant paper using an audit study to measure the prevalence and consequences of discrimination in the labor market. In this post, I want to use the paper to talk about the journal publication process. Specifically, I want to discuss why this paper appeared in Sociological Science.

First, it may be the case that Professor Pager directly went to Sociological Science without trying another peer reviewed journal. If so, then I congratulate both Pager and Sociological Science. By putting a high quality paper into public access, both Professor Pager and the editors of Sociological Science have shown that we don’t need the lengthy and cumbersome developmental review system to get work out there.

Second, it may be the case that Professor Pager tried another journal, probably the ASR or AJS or an elite specialty journal and it was rejected. If so, that raises an important question – what specifically was “wrong” with this paper? Whatever one thinks of the Becker theory of racial discrimination, one can’t critique the paper on lacking a “framing” or have a simple and clean research design. One can’t critique statistical technique because it’s a simple comparison of means. One can’t critique the importance of the finding – the correlation between discrimination in hiring and firm closure is important to know and notable in size. And, of course, the paper is short and clearly written.

Perhaps the only criticism I can come up with is a sort of “identification fundamentalism.” Perhaps reviewers brought up the fact discrimination was not randomly assigned to firms so you can’t infer anything from the correlation. That is bizarre because it would render Becker’s thesis un-testable. What experimental design would allow you get a random selection of firms to suddenly become racist in their hiring practices? Here, the only sensible approach is Bayesian – you collect high quality observational data and revise your beliefs accordingly. This criticism, if it was made, isn’t sound upon reflection. I wonder what, possibly, could the grounds for rejection be aside from knee jerk anti-rational choice comments or discomfort with a finding that markets do have some corrective to racial discrimination.

Bottom line: Pager and the Sociological Science crew are to be commended. Maybe Pager just wanted this paper “out there” or just got tired of the review process. Either way, three cheers for Pager and the Soc Sci Crew.

50+ chapters of grad skool advice goodness: Grad Skool Rulz ($2!!!!)/From Black Power/Party in the Street

Written by fabiorojas

September 28, 2016 at 12:10 am

9 Responses

Subscribe to comments with RSS.

  1. Another potential reason… maybe it was submitted elsewhere first and, although rejected, only emerged in current form due to incorporating feedback from earlier review process?

    Regardless, as you said, “three cheers for Pager and the Soc Sci Crew.” Also, thank you for bringing the paper to my attention. I mentioned it in a class recently as evidence of benefits of organizational diversity.


  2. Three cheers indeed! But the idea that what ails the refereeing process at major soc journals is “identification fundamentalism” strikes me as far fetched.


    Pierre Azoulay

    September 28, 2016 at 10:46 am

  3. Pager’s paper is awesome. It deserves all of the praise you and others heap on it. That said, why should we be surprised that ASR/AJS sometimes reject good papers? If good papers only appeared in those two journals, our discipline would be in terrible shape. I don’t see Pager’s inclusion in Soc Science as a negative statement about ASR/AJS at all. They can’t publish every good paper that comes their way, but that’s okay because we have other journals, like Soc Science, that also publish good papers.

    The problem here may be that we simply put too much weight on the value of getting a paper in ASR/AJS. I’m a strong advocate of the don’t count/read approach. Good papers show up in lots of different journals. We should celebrate them instead of lamenting that they didn’t get published in the “top” journals. Deifying ASR/AJS just reproduces a winner-takes-all approach to publishing that I think is unhealthy for social science.

    Liked by 1 person

    brayden king

    September 28, 2016 at 2:03 pm

  4. Pierre: My claim isn’t that identification fundamentalism is the only, or even main, problem with journals. Rather, it is a comment on this paper. It was the only substantive criticism I can come up with and it’s lame..

    Brayden: That’s a good argument for 1986. In a world of limited space, it’s the price you pay for having humans edit journals. But in 2016, it raises deeper questions. If we have unlimited electronic space, why reject papers at all that have scientific merit? Also, it’s an issue about the lack of reliability of peer review, especially in sociology.



    September 28, 2016 at 3:58 pm

  5. I think Brayden makes good points about deifying AJS/ASR articles. Not every article published in those journals is well-done. Some AJS/ASR articles are rather questionable.



    September 28, 2016 at 4:28 pm

  6. Fabio: Not even Sociological Science accepts every paper that has some scientific merit.


    brayden king

    September 28, 2016 at 7:01 pm

  7. Correct but PLOS one does as do arXiv and SocRxiv. Why limit ourselves?



    September 28, 2016 at 7:13 pm

  8. Here is what I experienced as a student in a reputable graduate program; this professor googled the author/s of a supposedly anonymized paper for review in front of my face and flaunted how talented he is. Sociology isn’t mathematics where you can come out of nowhere and become famous as long as you prove something. Network, prestige, ideology, and facework are as important as the work itself, right? I bet the result would be different if some of the parameters are tweaked. Probably the editor/s just don’t like Pager’s results. Well, it’s so easy to come up with the most scathing criticism in sociology. Social science in decay.

    Liked by 1 person


    September 29, 2016 at 2:49 pm

  9. One problem with this piece is the sample size, way too small for MLE. This is interesting research, but the evidence isn’t compelling. It seems to me it’s right where it belongs.


    Fred Thompson

    October 1, 2016 at 7:40 pm

Comments are closed.

%d bloggers like this: