Let’s improve the ASA/Sage journal author agreement

I have spoken well of the policy that permits authors to post preprint versions of their papers before submitting them to journals of the American Sociological Association. That means you can get your work out more broadly while it’s going through the review process. The rule says:

ASA authors may post working versions of their papers on their personal web sites and non-peer-reviewed repositories. Such postings are not considered by ASA as previous publication.

The policy goes on to ask that authors modify their posted papers to acknowledge publication if they are subsequently published. That’s all reasonable. This is why SocArXiv and other services offer authors the opportunity to link their papers to the DOI (record locator) for the published version, should it become available. This allows citation aggregators such as Google Scholar link the records.

Unfortunately, the good part of this policy is undermined by the ASA / Sage author agreement that authors sign when their paper is accepted. It transfers the copyright of the paper to ASA, and sets conditions under which authors can distribute the paper in the future. The key passage here is this:

1. Subject to the conditions in this paragraph, without further permission each Contributor may …

• At any time, circulate or post on any repository or website, the version of the Contribution that Contributors submitted to the Journal (i.e. the version before peer-review) or an abstract of the Contribution.
• No sooner than 12 months after initial publication, post on any non-commercial repository or website the version of the Contribution that was accepted for publication.

This is not good. It means that if you post a paper publicly, e.g., on SocArXiv, and then submit it to ASA, you can’t update it to the revised version as your paper moves through the process. Only 12 months after ASA publishes it can you update the preprint version to match the version that the journal approved.

This policy, if followed, would produce multiple bad outcomes.

One scenario is that people post papers publicly, and submit them to ASA journals for review. Over the course of the next year or so, the paper is substantially revised and eventually published, but the preprint version is not updated until a full year after that, often two years after the initial submission. That means readers don’t get to see the improved version, and authors have to live with people reading and sharing their unimproved work. This discourages people from sharing their papers in the first place.

In the other scenario, people update their preprints as the paper goes through the revision process, so they and their readers get the benefit of access to the latest work. However, when the paper is accepted authors are expected to remove from public view that revised paper, and only share the pre-review version. If this were feasible, it would be terrible for science and the public interest, as well as the author’s career interests. Of course, this isn’t really feasible — you can’t unring the bell of internet distribution (SocArXiv and other preprint services do not allow removing papers, which would corrupt the scholarly record.) This would also discourage people from sharing their papers in the first place.

So, what possible reason can there be for this policy? It is clearly intended to punish the public in order to buttress the revenue stream of Sage, which returns some of its profits to ASA, at the expense of our libraries, which pay for subscriptions to ASA journals.

I assume this policy is never enforced, as I’ve never heard of it, but I don’t know that for a fact. It’s also possible that whoever wrote the Publications policy I linked above didn’t realize that it contradicted the Sage author agreement, which basically no one reads. I also assume that such a policy does not in fact have any effect on Sage’s profits, or the profits that it kick backs to ASA. So it’s probably useless, but if it has any effects at all they’re bad, by discouraging people from distributing their work. ASA should change this author agreement.

I will be on the ballot for the ASA Publications Committee this spring. If elected, I will add making this change to my platform, which I outlined here. If I’m not elected, I’ll try to do this anyway.

Review essay: Public engagement and the influence imperative

I have written a review essay at the invitation of Contemporary Sociology. Here’s a preprint version on SocArXiv: https://osf.io/preprints/socarxiv/v27xk/.

This is the abstract. Feedback welcome!

Public engagement and the influence imperative

Abstract: A review essay discussing three advice books for social scientists. Sociologists, in responding to the imperative to make their work more influential, must go beyond doing “public sociology” to embrace doing sociology “in public” (Healy 2017). Rather than using public engagement primarily for publicity – to make our research matter – we should use engagement to help us do research that matters in the first place. Next, I caution that the drive to be professionally rewarded for public intellectualism is fraught with conflicts that may be irreconcilable. To be a public intellectual today requires being both public in one’s intellectual life and intellectual in one’s public life, and for academics in the era of the “market university” (Berman 2011), trying to get paid for that leads to a neoliberal trap. Finally, I argue for a move beyond personal strategies toward the development of the open scholarship as an institutional response that ultimately may be responsible for sociology’s survival.

Here is the SocArXiv citation:

Cohen, Philip N., 2018. “Public Engagement and the Influence Imperative”. SocArXiv. April 7. doi:10.17605/OSF.IO/V27XK.

How I engaged my way to excellent research success and you can too

Kid photo CC from MB Photography; collage by pnc.

Too often sociologists think of social media, or online communications generally, primarily as a way of broadcasting their ideas and building their audience, instead of as a way of deepening their engagement with different people and perspectives. You see this when academics start a twitter account right when their book is coming out. Nothing wrong with that, but it’s very limited. A crucial part of being a public scholar, public intellectual, or a public sociologist, etc., is reading, listening, and learning through engagement, and digital communication can enhance the metabolism of that process. Especially important is the chance to learn from people you don’t normally interact with. For all the complaints about social media bubbles, some true, social media also offers huge efficiencies for meeting and learning from new people.

As I’m writing an essay about this, I thought of my work on divorce as an example. So here’s that thread, condensed.

A divorce story

In 2008 I was teaching an undergraduate Family Sociology course at the University of North Carolina, and included a section on divorce based on other people’s research. I was also developing a proposal for my own textbook, which at the time framed family structures and events, including divorce, as consequences and causes of inequality. I was reading research about divorce along with many other family issues that were outside of my formal training and experience (the closest I had come to a family demography or family sociology course was a seminar on Gender, Work & Family in grad school).

Then in 2009, I wrote a post on my pretty new blog criticizing something bad the Brad Wilcox had written about divorce. I was trying to be newsy and current, and he was claiming that the recession was lowering divorce rates because hard times pulled people together. We didn’t yet know what would happen in the recession. (In the comments, Louise Roth suggested it would take time for divorces “caused” by the recession to show up, which turned out to be true.)

I kept on that path for a while, criticizing Wilcox again for similar work in 2011. By then — prompted by the combination of my reading, the blog debates, and the news coverage around families and the recession — I was working on a paper on divorce using the American Community Survey. I presented it at a demography meeting in the summer of 2011, then revised and presented it at the Population Association of America the following spring. I blogged about this a couple more times as I worked on it, using data on state variation, and Google searches, each time getting feedback from readers.

A version of the paper was rejected by Demography in the summer of 2011 (which generated useful reviews). Although now discredited as not peer-review-publishable (which no one knew), my commentary on divorce and the recession was nevertheless featured in an NPR story by Shankar Vedantam. Further inspired, I sent a new version of the paper (with new data) to Demographic Research, which also rejected it. I presented on the work a couple of times in 2012, getting feedback each time. By August 2012, with the paper still not “published,” I was quoted describing my “divorce/recession lull-rebound hypothesis” in New York magazine.

The news media pieces were not simply my work appearing in the news, in a one-directional manner, or me commenting on other people’s research, but rather me bringing data and informed commentary to stories the reporters were already working on. Their work influenced my work. And all along that news coverage was generating on- and offline conversations, as I found and shared work by other people working on these topics (like the National Center for Marriage and Family Research, and the Pew Research Center). Looking back over my tweets about divorce, I see that I covered divorce and religion, disabilities, economics, and race/ethnic inequality, and also critiqued media coverage. (Everything also got discussed on Facebook, in a smaller semi-private circle.)

By 2014 I finally got the paper — now with even newer data — published in a paywalled peer-reviewed journal, in Population Research and Policy Review. This involved writing the dreaded phrase, “Thank you very much for the opportunity to revise this paper again.” (Submitted October 2012, revision submitted August 2013, second revision submitted January 2014, final revision April 2014.) The paper, eventually titled, “Recession and Divorce in the United States, 2008-2011,” did improve over this time as new data provided better leverage on the question, and the reviewers actually made some good suggestions.

Also in 2014 the descriptive analysis was published in my textbook. The results were reported here and there, and expanded into the general area of family-recession studies, including this piece in the Conversation. I also developed a method of projecting lifetime divorce odds (basically 50%), for which I shared the data and code, which was reported on here. Along the way I also did some work on job characteristics and divorce (data and code, working paper). When I posted technical notes, I got interesting responses from people like economist Marina Adshade, whom I’ve never met.

From my presentation, “The open scholarship media strategy for maximum success.”

So that’s an engagement story that includes teaching, the blogosphere and social media, news media, peer-reviewed publishing, conference presentations and colloquium talks. I did research, but also argued about politics and inequality, and taught and learned demography. It’s not a story of how I used social media, or the news media, to get the word out about my research, although that happened, too. The work product, not just the “publications,” were all public to varying degrees, and the discussions included all manner of students, sociologists, reporters, and interested blog or Twitter readers, most of whom I didn’t know or wouldn’t have met any other way.

So I can’t draw a line dividing the “engagement” and the “research,” because they weren’t separate processes.

Philip Cohen’s ASA Publications Committee platform

The American Sociological Association’s Committee on Nominations invited me to stand for election to the ASA’s Committee on Publications. I accepted the invitation reluctantly, because I am pessimistic about the opportunities to make progress on the most pressing issues before ASA, which I discuss below. I think it’s likely we’ll be in better shape for making big decisions in a few years, after the economics of publishing have gotten worse and (hopefully) alternatives such as SocArXiv and Sociological Science have gotten stronger.

But I appreciate the invitation, and I think there is something to be gained even if all I can do is raise the issues, and help disseminate important information (and my opinions) to the membership. And of course I can’t predict the future, and what the other present and future committee members will do, so who knows. Plus, it’s as a chance to spend time with the interesting and dedicated members of the committee and ASA staff.

Note on conflict of interest

I am the director of SocArXiv, a non-profit, open source, open access paper server for the social sciences, which I founded with a group of sociologists and library community leaders. I don’t draw a salary as director of SocArXiv, but if I can get grants for the project I might use them to buy off some of my time (meaning my salary doesn’t go up, but I work on SocArXiv instead of teaching), so you could say I stand to gain from promoting SocArXiv. And one of my ambitions in that role is to have ASA work with SocArXiv, for example in the dissemination of preprints and postprints from ASA journals, conference papers, and possibly innovative new projects like open peer review. I hope serving on this committee will help me advance SocArXiv’s agenda as well as ASA’s. So if that makes you uncomfortable you might not want to vote for me. In that case you might also consider the similar conflicts of interest held by the committee members who are, were, or hope to be editors of journals Sage publishes for ASA, or who otherwise benefit in their careers from the paywalled publishing status quo in academia. We’re all trying to do what we think is right and get paid and at the same time.

Basic perspective

In 2016 I wrote a short essay about ASA and publishing, and my basic perspective hasn’t much changed. Here is the gist from that essay.

The scholarly communication system is broken, and ASA lives off the money that brokenness creates. According to the 2016 budget report (I don’t see an update to this), 35% of total revenue comes from journal operations. That is $2.2 million that came from institutional subscriptions (mostly paid by the libraries of colleges and universities where ASA members work), under the contract with Sage. Increasingly, these subscriptions are part of big “bundles” of journals, in which individual libraries have little say over what they’re actually buying. Publishing the journals, in turn, costs 11% of total expenses, or about $717,000. So journal publishing produces money for other things the association does.

At the same time, ASA — like other paywall publishers — is in an increasingly defensive position, as open access alternatives spread and the cost of technologically and legally defending the paywall increases under pressure from Sci-Hub (which I wrote about here) and various other breaches. In a quasi-official statement from the ASA, publications director Karen Edwards wrote that Sci-Hub, “threatens the well-being of ASA and our sister associations as well as the peer assessment of scholarship in sociology and other academic disciplines.” Without the paywall, she implied, peer review itself cannot survive. I disagree.

More generally, ASA staff has raised alarms about the sustainability of the current model. From the Publications Committee minutes in spring 2016:  “The possibility exists that the journal world may not be as profitable in the future as it is now. The journal marketplace is shifting, and will continue to do so, so Council and EOB should keep an eye on this revenue source.”

We know that free journals could be published for a fraction of what ASA and Sage now spend and reap (one of the major expenses of any paywall publisher is developing and maintaining the technology to keep publications out of the hands of non-paying customers). That would mean giving up a substantial share of the association’s current income. And of course, it’s not a simple task. The basic goal is a future in which scholarly societies, academic libraries, and granting agencies together pay for scholarly publication, and cut out the for-profit publishers. That is instead of universities and federal agencies paying for research twice — once for the researchers, and once again for their published output — and all of it gets to be open access. This requires some institutional mechanism for collecting and distributing the funds used to produce open-access research output. Such solutions will require creativity, collaboration, and hard work. Designing a new system is relatively easy, but moving today’s institutional actors in that direction is not.

Platform

Here’s what I’d like to do, more and less implausible. These are overlapping and not mutually exclusive.

1. Adopt TOP guidelines for ASA journals

The Center for Open Science has published Transparency and Openness Promotion (TOP) Guidelines. The guidelines are incremental.  With almost 5,000 signatories — journals and associations — they include eight standards, each with three levels of stringency. Journals select which standards they will adopt and a level of implementation for each, ranging from disclosure requirements to verified replications. For example, we could say ASA journals will encourage data and code for analyses be posted publicly and require a statement about whether they are or not. We could encourage publication of replication studies. And so on. Even at low levels of implementation, they set an important tone and direction. This could be done for all of ASA, or individual journal editorial boards could implement them. In addition, by assigning openness badges to journals, authors, or papers, we can symbolically reward open scholarship practices.

2. Adopt preprints and data sharing

When people submit papers for ASA journals, they should be encouraged or required to post them on a public server (such as SocArXiv), along with data and code. When papers are published they can be linked to the preprints, ensuring readers are let to the final versions, while disseminating the research earlier, and free. As a model, consider the American Psychological Association, which has designated a preferred preprint server (PsyArXiv) and data archive (the Center for Open Science, which hosts SocArXiv and PsyArXiv).

3. Open conference papers

ASA requires complete papers be submitted for presentation at the annual meetings. We should make these papers available in a public archive, properly identified and preserved as part of the scholarly record. This is a key step in developing the working paper culture in sociology.

4. No new paywall journals

I will oppose the creation of any new paywalled journals, by ASA or its sections.

5. Flip ASA journals

The Holy Grail. Flipping journals refers to transitioning them from paywalled to open access. We can disseminate more sociology, better and faster, for less money. University libraries and university presses, and some foundations, can be mobilized to raise the money needed to launch sustainable models. For example, if a few hundred libraries would agree to give ASA what they spend on subscriptions to our journals, we could produce the same journals and open them to the public, cutting out Sage’s profits. My goal is for the ASA membership to task our association’s staff with figuring out how to pay for the journals without paywalls, which might require research and grants, and then bring proposals to the membership for approval.

6. Reduce revenue

ASA should make less money. Instead of making money from publications, we should just raise the money we need to publish journals, with the appropriate academic partners (universities and funding agencies). The many laudable things ASA does should be paid for voluntarily, by institutions or people who want to support them, rather than being paid for by publication profits from consumers (our employers) who have no choice but to subsidize ASA operations (and Sage profits).

7. Disseminate information

Too many sociologists don’t know how ASA works, how academic publishing works, and how we all fit into the overall system of research production and dissemination, the “scholarly communications ecosystem.” I would like to see us (starting with students) learning more about the politics and economics of this system, to have a better grasp of the big issues we face. On the publications committee I would hope to learn a lot, and help communicate this to our members and the public.

Conclusion

Here are some relevant experiences I’ve had. As editor, with Syed Ali (who is also running for publications committee!) of Contexts for three years (just completed), I was a non-voting member of the publications committee and got to see how it works. This included participating on a task force on the future of Contexts, which resulted in Sage agreeing to allow free downloads of the journal for the first 30 days of each issue, and in perpetuity after the first year. I’m the current chair of the Family section, and secretary treasurer of the Population section. I also successfully agitated for a new rule requiring ASA-award-winning dissertations to be publicly available. (For previous posts and activity related to ASA, visit the tag on this blog.) All that gives me some useful experience and knowledge.

I’d love for this election to help spur a more widespread discussion and debate over these questions. Without a majority on the publications committee or the ASA Council (and even with such a majority) this is a pretty daunting set of goals. I couldn’t pretend to promise results on any of these, but I pledge to at least promote these ideas, in the service of more research transparency and openness at lower cost.

The ASA election ballots will go out in April. To see what the Committee on Publications does, check out the committee records here.

Update on SocArXiv and social science without walls

Meanwhile, over at SocArXiv, we’re working on revolutionizing the research process and how we communicate about it in the social sciences. You can follow the exploits of the SocArXiv project on our blog SocOpen. There you can read, most recently:

• An interview I did as part of the OA Interviews series with Richard Poynder, describing the state of our project and where we hope it’s going, in addition to offering my opinions on lots of related issues.
• The announcement of a conference we are planning, O3S: Open Scholarship the for Social Sciences, at the University of Maryland, October 26-27.
• A statement from our steering committee on the March for Science, and the importance of social science without walls to our mission and goals.
• An update on our fundraising drive.

That’s the update!

Advice for and about ASA

Flickr / CC: https://flic.kr/p/sjBWi

Last summer the incoming American Sociological Association President, Michèle Lamont, asked me to offer some advice to ASA about open access publishing issues. It was an open-ended request, and I didn’t know how to go about it. My understanding of ASA is that it is not well outfitted as a change agent; it’s much more likely to respond to external developments in its ecosystem than to take the lead, especially when its revenue stream is at stake. Nevertheless, lots of good people work in and around the association, and it has great capacity. (I am involved myself, as co-editor of the ASA magazine Contexts, as chair-elect of the Family Section, and as secretary treasurer of the Population Section.) So I wrote a short essay on what ASA might do, or what its members might do or demand of it.

It’s not coincidental that this is posted on the SocArXiv blog, SocOpen, which is part of that changing external environment that I hope will lead to ASA adapting for the better. I believe that devoting my energy to this project is producing something tangible for research and scholarly communication, while also pressuring ASA (and maybe other associations) to move in the right direction.

I hope you’ll read it on SocOpen.

No paper, no news (#NoPaperNoNews)

In the abstract, the missions of science and science reporting align. But in the market arena they both have incentives to cheat, stretch, and rush. Members of the two groups sometimes have joint interests in pumping up research findings. Reporters feel pressure to get scoops on cutting edge research, research that they want to appear important as well as true — so they may want to avoid a pack of whining, jealous tweed-wearers seen as more hindrance than help. And researchers (and their press offices) want to get splashy, positive coverage of their discoveries that isn’t bogged down by the objections of all those whining, jealous tweed-wearers either.

Despite some bad incentives, the alliance between good researchers and good reporters may be growing stronger these days, with the potential to help stem the daily tide of ridiculous stories. Partly due to social media interaction, it’s become easier for researchers to ping reporters directly about their research, or about a problem with a story; and it’s become easier for reporters to find and contact researchers to cover their work, and for comment or analysis of research they’re covering. The result is an increase in research reporting that is skeptical and exploratory rather than just exuberant or exaggerated. Some of this rapid interaction between experts researchers and expert reporters, in fact, operates as a layer of improved peer review, subjecting potentially important research to more extreme vetting at just the right moment.

Those of us in these relationships who want to do the right thing really do need each other. And one way to help is to encourage the development of prosocial norms and best practices. To that end, I think we should agree on a No Paper No News pact. Let’s pledge:

• If you are a researcher, or university press office, and you want your research covered, free up the paper — and insist that news coverage link to it. Make the journal open a copy, or post a preprint somewhere like SocArXiv.
• If you are a reporter or editor, and you want to cover new research, insist that the researcher, university, or journal, provide open access to its content — then link to it.
• If you are a consumer of science or research reporting, and you want to evaluate news coverage, look for a clear link to an open access copy of the paper. If you don’t see one, flag it with the #NoPaperNoNews tag, and pressure the news/research collaborators to comply with this basic best practice.

This is not an extremist approach. I’m not saying we must require complete open access to all research (something I would like to see, of course). And this is not dissing the peer review process, which, although seriously flawed in its application, is basically a good idea. But peer review is nothing like a guarantee that research is good, and it’s even less a guarantee that research as translated through a news release and then a reporter and an editor is reliable and responsible. #NoPaperNoNews recognizes that when research enters the public arena through the news media, it may become important in unanticipated ways, and it may be subject to more irresponsible uses, misunderstandings, and exploitation. Providing direct access to the research product itself makes it possible for concerned people to get involved and speak up if something is going wrong. It also enhances the positive impact of the research reporting, which is great when the research is good.

Plenty of reporters, editors, researchers, and universities practice some version of this, but it’s inconsistent. For example, the American Sociological Association currently has a news release up about a paper in the American Sociological Review, by Paula England,  Jonathan Bearak, Michelle Budig, and Melissa Hodges. And, as is now usually the case, that paper was selected by the ASR editors to be the freebie of the month, so it’s freely available. But the news release (which also only lists England as an author) doesn’t link to the paper. Some news reports link to the free copy but some don’t. ASA could easily add boilerplate language to their news releases, firmly suggesting that coverage link to the original paper, which is freely available.

Some publishers support this kind of approach, laying out free copies of breaking news research. But some don’t. In those cases, reporters and researchers can work together to make preprint versions available. In the social sciences, you can easily and immediately put a preprint on SocArXiv and add the link to the news report (to see which version you are free to post — pre-review, post-review, pre-edit, post-edit, etc. — consult your author agreement or look up the journal in the Sherpa/Romeo database.)

This practice is easy to enforce because it’s simple and technologically easy. When a New York Times reporter says, “I’d love to cover this research. Just tell me where I can link to the paper,” most researchers, universities, and publishers will jump to accommodate them. The only people who will want to block it are bad actors: people who don’t want their research scrutinized, reporters who don’t want to be double-checked, publishers who prioritize income over the public good.

#NoPaperNoNews