new occupy wall street paper at PLoS One


My student* Karissa McKelvey has a co-authored a paper on the geography of Occupy Wall Street with Michael Conover, Claytion Davis, Emilio Ferrara, Fillipo Menczer, and Alessandro Flammini. Analyzing Twitter traffic data, The Geospatial Characteristics of a Social Movement Communication Network explores the clustering of OWS in specific urban areas. A nice use of social media and required reading for those interested in recent movement history.

Adverts: From Black Power/Grad Skool Rulz

* Actually, she’s Fil Menczer‘s student, but she’s so awesome I claim her as well since I’m her minor concentration adviser.

Written by fabiorojas

March 12, 2013 at 12:14 am

18 Responses

Subscribe to comments with RSS.

  1. Why PLoS ONE?



    March 12, 2013 at 12:11 pm

  2. Jerry – I’m guessing: open access, visibility and interdisciplinary impact, turnaround times. Different model from our journals (though, I know many of our journals, including ASQ, are doing a much better job these days on turnaround times).



    March 12, 2013 at 5:23 pm

  3. @Jerry:

    Option 1: Spend 1 year writing a 45 page paper and send it to a regular social science journal. Wait 6 months for it to be rejected because it is “a-theoretical” and “just descriptive.” Send it to a second journal, who, after 6 months, will ask for a bunch of revisions, which it takes you 6 months to do. Then send it back. five months later, they accept but conditional on more revisions, which adds two more months. You send it back. 12 moths later you see it in print. Then the journal can only be accessed by people at either universities or those who pay $45 dollars.

    Total time from start to finish: (12+6+6+6+5+2+12= 49 months) = 4 years + limited visibility

    Option 2: Spend about two months writing a short 6-10 page paper based on a single finding. Send it to a major science journal (e.g., Nature), which rejects after 1 month. Then try PLoS One, which reviews in one month. You spend three more on the R&R and then one month on the final acceptance. It appears one month later and is available to the entire world.

    Total time from start to finish: (2+1+1+3+1+1 = 8 months) < 1 year + unlimited visibility

    And why, exactly, should people bother with regular social science journals?

    Girl, please.

    PS. Yes, I know, that you've managed to make ASQ fast – and I thank you for that. But Jerry Davis is a Special Person and not representative of the social sciences in general. Our journal system, frankly, reflects our lack of self-respect. We should be embarrassed that nuclear physics and brain surgery gets reviewed in a more timely fashion than a 20 page paper with a few regression tables.



    March 12, 2013 at 5:39 pm

  4. Thanks Teppo and Fabio, those are good reasons. I’m actually quite intrigued by the PLoS ONE model, and find their statement of aims ( to be pretty sensible:
    “Too often a journal’s decision to publish a paper is dominated by what the Editor/s think is interesting and will gain greater readership — both of which are subjective judgments and lead to decisions which are frustrating and delay the publication of your work. PLOS ONE will rigorously peer-review your submissions and publish all papers that are judged to be technically sound. Judgments about the importance of any particular paper are then made after publication by the readership (who are the most qualified to determine what is of interest to them).”

    Fabio’s question — why should people bother with regular social science journals — is one I’ve been giving a lot of thought. It’s interesting how much PLoS ONE’s motley contents parallel those of the first science journal, the “Philosophical Transactions of the Royal Society” ( Science in the baroque era was done by smart amateurs, not so much university-based scholars. Similarly, these days anyone can gather and analyze archival data online or run experiments on Mturk, publish in PLoS ONE, and get a nice write-up in the Sunday NY Times.

    So Fabio: why indeed? Not just why journals, but why the whole university apparatus, which mostly serves to reinforce existing (and perhaps arbitrary) status distinctions and overlooks potential contributions from “otusiders”?



    March 12, 2013 at 8:11 pm

  5. Jerry, you overstate my question and set up a straw man. The question is not “why journals?” or “why universities?”The answer is obvious. We want third parties to check our work and help us improve. Editors sort for us. The cynical among us strive for a stamp of approval from high status peers.

    The question was: “Why *social science* journals?” Think about my student Karissa. Talented and extremely hard working. She works at the intersection of two areas, computer science and social science. If she takes a physical science path, she will almost certainly get her work published in a year or two and get a huge reward.

    And I’m not talking about PLoS one in particular. Physical scientists have journals that referee in months, not years. Computer scientists do most of their work through conferences, which referee in terms of months. Biomedical journals do reviews in weeks. The result is a field that evolves quickly and in response to growing knowledge.

    Also, physical scientists don’t cling as strongly to peer reviewed materials. ArXiv papers and pre-prints will be cited if experts judge that they are good. Knowledge doesn’t wait until the editor of a journal gets around to writing you an acceptance letter.

    So what argument would you give to Karissa to switch from submitting to Science, Nature, and PNAS and go to the social science journals? This is a serious question. When I describe my publishing options to colleagues in the medical schools and physical sciences, they are revolted.

    I think we are slowly moving toward a fairly flat system of knowledge distribution. The very top journals will remain the same. ASQ need not change, especially with such an excellent editor! The top journals will continue to get the best papers.

    But the rest of the system is feeling pressure. Once you get your rejection from top journal X in your field, why bother with lower tier journals? Why not go to a PLoS one system or some other open source system and let the “people” decide? in the short term, the old 19th century system will be perpetuated by junior scholars who need peer reviewed CV lines, but in the long term senior scholars will erode the system, as it is already.

    Check out this link to Google Scholar ( These papers are getting hundreds of citations and they could have gotten into second or third their chemistry or biology journals. It is worth thinking long and hard about.



    March 12, 2013 at 8:32 pm

  6. Fun facts, Jerry: Look at the list of the 20 most cited scientific journals. #12 – arXiv Astrophysics. #14 – NBER working papers. (see




    March 12, 2013 at 8:36 pm

  7. @Fabio: I wasn’t intending a straw man, but to accurately represent your position (“why, exactly, should people bother with regular social science journals?”). Social science journals are generally slow, and have a few other pathologies. They’re also costly to maintain. There is a lot to like about some aspects of how knowledge is accumulated in other fields. I think there is real merit in your position, and we really should be talking about why we still have journals.

    PLOSONE provides peer review (one main rationale for “normal” journals) and lets the people decide what a good contribution is, not (generally elderly) editors and (generally middle-aged) reviewers. You therefore suggest that lower-tier journals might give way to PLOSONE, but that fancy journals might persist due to careerism (at least as long as tenure committees distinguish “A journals” from others), egoism (wanting to get approval from high-status peers), and because journals (sometimes) provide developmental guidance and make work better. From the perspective of advancing the discipline, it seems this last function is the only legit reason to keep normal high-status journals around.

    On the other hand, I can’t help but think that slow turnaround in journals might not be the main thing holding sociology back while computer science goes from triumph to triumph. (I await that PLOSONE article that finally resolves the agency/structure dialectic and provides a clear criterion for the boundary of a “field”…)



    March 12, 2013 at 9:04 pm

  8. This will be part of a longer thread, but a comment about CS vs. soc. Agency/structure is a hard argument to articulate, much less “solve” in any satisfactory way. That’s why we’re stuck on it. Some may say it’s a bit of a red herring as well.

    CS also has its fair share of insanely hard problems (e.g., P vs. NP). The difference is that CS allows its scholars to more quickly and efficiently publish normal science results. Remember, most of sociology is regular old research on topics like shareholder behavior, voting behavior and gender differences in school. There is no reason at all that this stuff should take years to process. Other disciplines process normal science in a much speedier fashion. We should learn a lot from that. Heck, they process *social science* faster than we process social science. Geez, doesn’t that tell you that something is seriously out of whack?

    PS. If I were to bet on the future, many disciplines will gravitate to a system of “generals, a few specialties, or take a gamble with PLoS.”



    March 12, 2013 at 9:22 pm

  9. Actually, here is a Science article that attempts to model quantitatively some “criterion for the boundary of a field.” (happens to also stand as a shameless plug for my advisor Fil Not so surprisingly, they find that disciplines tend to emerge from the splitting and merging of social communities in a collaboration network. Social ties between scientists are key here — maybe driving like open-access journals’ large success in the internet age.

    Something you both seem to take for granted, but is important to point out, is that the aging journal model was useful to get your papers to the broader community before digital libraries and the internet. The open access model only works because the cost to reproduce and publish has become infinitesimal in comparison to travelling reams of paper. It’s really no wonder the old model is crumbling. We need to keep communicating about the benefits of open access and low overhead with colleagues from a variety of disciplines.


    Karissa McKelvey

    March 12, 2013 at 9:33 pm

  10. @karissa: There is much to what you say. Open sources makes sense now, but not in the 1950s. At the same time, some of the pathologies of social science are recent. My adviser would tell me that in the 70s and even 80s, it was normal for people to quickly review. The culture then, for some reason, sifted into its ow dysfunctional state. That’s bad and we shouldn’t tolerate it.



    March 12, 2013 at 9:39 pm

  11. I think Fabio’s point about the speed at which we process normal science type social science papers speaks to a need to reform how we write papers as well. The fact is, the vast majority of sociology papers do not require the lengthy, ritualistic and redundant literature reviews that are the norm these days. Cutting this down would be a first step toward facilitating faster review times. If you need that much to space to explain why the findings in your regression table are important, they probably aren’t that important.



    March 12, 2013 at 10:00 pm

  12. In January 2003 I submitted a paper to AMR. After three rounds of review, the paper was published January 2007 — FOUR years later. I have a few other papers with similar profiles, though the average over all papers is probably 2.5 years, or so (and two-three rounds of review).

    I’ve also gone through the above AND then had the paper REJECTED. So, three rounds of review, two years of trying to placate reviewers – and then rejected. Happily the paper eventually found a great home.

    On Nov 26, 2012 a co-author and I submitted a paper to PLOS ONE. We received feedback within two weeks, revised the paper in two weeks, received additional feedback in two weeks, revised again and then had the paper accepted Jan 14, 2013 – all in less than two months. The paper was published Feb 25, 2013.

    Four years versus three months – that’s a big difference. (I’ll keep submitting to both types of journals – and will also try to change the dynamics at SO, where I co-edit.)

    So, time is an issue. It’s particularly an issue for young scholars seeking tenure and trying to establish a research presence – the usual three-four year cycle times make things difficult (particularly if something gets rejected after two years of work/revisions). I’ve noted that some young scholars are sending their work directly to science-type journals, by-passing the usual hoop-jumping in our journals – and their work is getting cited in social science journals as well, and it’ll be hard not to give tenure to these folks. Sure, Science and Nature-type articles are sometimes horrid (ignoring past work, lacking theoretical contribution, etc) – I recognize that.

    Also, it’s hard to cite/find things if it takes two+ years, even once accepted, for the paper to show up in a journal. I remember a friend going up for tenure and he was asked about his “impact” – in essence, how much has your work actually been cited? He had the requisite “number” of articles in top-top outlets (in fact, more than enough – though I can’t stand this type of “counting”), more importantly, was pioneering an area of research – but the problem of course was that his research had indeed taken 4-5 years to get published (he had some quite extreme HORROR stories) and thus his work, at that time, had not received many citations (it has NOW). Well, citations of course aren’t everything. Anyways, he left that institutions, and thankfully rightly received tenure at another institution. (“Early view” is of course changing this, to some extent.)

    The issues of access are also central. It’s harder to cite things if they are gated (in pricy library databases). Thankfully there is now lots of pressure on this issue.



    March 12, 2013 at 11:04 pm

  13. @Teppo: I hear you. Here’s a working paper I wrote with Klaus Weber in 2000 when he was a youthful grad student:
    and here it is at the end of its long journey (and after Mike Lounsbury joined us) in 2009 in AMJ:
    Yep, nine years…and the crazy thing is that it seemed just about as topical after a decade in “pre-publication” limbo as it did when we started.

    What PLOSONE (and perhaps Science and ArXiv) provide is quick exposure to (minimally) screened findings. What fancy social science journals provide is, one hopes, developmental advice that improves the quality of one’s contribution. Maybe the right decision rule for authors is, if you have a time-sensitive finding that might get scooped, go for PLOSONE (and mobilize your PR); if you aim for a contribution to “theory,” which would probably benefit from the guidance of old heads, go for fancy social science journals. I have a much easier time thinking of scholars who have made careers out of the latter than the former.



    March 13, 2013 at 2:29 am

  14. @Jerry: “I have a much easier time thinking of scholars who have made careers out of the latter than the former.”

    You need to expand a little. Within social sciences, it is rare, but it happens. For example, a bit of James Fowler’s work came out in fast outlets like Science and the New England Journal of Medicine, which do a lot of desk rejection and quick review for other papers. At Michigan, Lada Adamic, whose work is about social networks, has made a career out of ArXiv, CS conferences, PLoS, Physica A, and other venues that are fast to publish. She’s had huge impact (one paper has 1000+ cites).

    But what about “theory?” How are often are big theory papers improved through years of review? I asked Paul DiMaggio about the Iron Cage article. He got a rejection after a month or two from AJS, then sent it to ASR and got an accept a few months after submission.

    You have this image of “big theory” papers that improve with the tortuous social science publication process. We all have papers that have spent years in the pipeline. But are the most important papers really those that got endlessly delayed? I’d like to see some systematic evidence, especially since other fields seem to do just fine with a faster process, I honestly don’t see why sociology is so special that it needs a totally different peer review process.



    March 13, 2013 at 3:02 am

  15. Mark Granovetter used to tell a story about submitting to one top journal and receiving a horrible reject, and then changing *only* the title and submitting it to another and getting an outright acceptance. (I believe this was the Strength of Weak Ties.) The process can certainly improve work, but it doesn’t always.

    I also think JD has hit on another issue–the increasingly large and ritualistic literature review/contortion act of theoretical contribution. David Strang had a nice working paper some years back (though I don’t know what happened to it) that showed that the percentage of articles devoted to literature reviews was increasing mightily while discussion of results was dropping precipitously. Surely, a theory and connection to a literature is important, but isn’t findings the game? At least, if we’re not doing pure theory.

    But review times are the killer. Who wants to submit to, say, Theory & Society as a junior professor and wait a year or more without communication for a rejection?



    March 13, 2013 at 5:16 am

  16. @Fabio and Cwalken: This may be the crux of the issue in sociology. Fabio wrote in response to my restatement of his question “Why bother with social science journals?”: “The answer is obvious. We want third parties to check our work and help us improve. Editors sort for us.” If a journal’s function is simply to assess “Was this competently done?” [Fabio’s “checking our work”], then reviews can be quick. If we want journals to *sort* for us, to decide “Is this interesting?”, that takes slightly more work and more skilled reviewers (and PLOSONE explicitly does not address this question in its review process). If we want the review process to improve our work, that can take much more time and requires skilled and patient reviewers and editors.

    In some sciences, it might be fair to assume that journal submitters are competent at their craft and that the field generally agrees on what counts as interesting, so submitted articles can skip most of the mumbo-jumbo that JD and Cwalken refer to and get to the methods and findings. (Fabio: note that Lada Adamic’s PhD is in physics, and she teaches in our School of Information, not a social science department, so she’s accustomed to these standards.) Finance is more like this, in my experience: people in the field agree that the question “Do conflicts of interest influence how mutual funds vote?” is worth answering, so the review process focuses on methods for ruling out alternative interpretations, not establishing the importance of the question. Fast journals don’t try to improve work; that’s what graduate school is for, or what colleagues are supposed to do before you submit the damned thing.

    Sociology, on the other hand, disagrees a lot on what counts as interesting, or what’s a contribution to theory. (The following are certified interesting theoretical claims in organizational sociology: who you know matters; who you’re seen with matters; knowing people who don’t know each other can yield benefits; sometimes we do what our friends do; sometimes we do what people who control goodies ask us to; it’s nice to have high status.) If a journal is going to call itself “Theory & Society,” it’s a safe bet that there will be a lot of wailing and gnashing of teeth over the “Is this interesting?” question (or “Is this really what Polanyi meant?”), and a lot of time-consuming effort will go into “improving your contribution.”



    March 13, 2013 at 11:56 am

  17. […] show some sociologists are already aware and supportive of this future. Consider a recent interaction at the blog, in which the site’s most prolific contributor (Fabio Rojas — an […]


  18. @jerrydavis: Don’t get me wrong. I don’t think theory is mumbo-jumbo. Not at all. Rather, there’s two issues on the table. 1) Sometimes what we have is just a nice finding that deserves to get out, whether or not it speaks to a contested terrain. It’s unfortunate that most journals have done away with the research note. And 2) review times have gotten out of control in some places. I certainly agree that reviews, and repeated reviews, make scholarship better. But we need to be wary of diminishing returns, and accept that not every paper will solve every problem. Particularly since this is having deleterious effects on people’s careers.



    March 14, 2013 at 5:41 am

Comments are closed.

%d bloggers like this: