Why not to post your working paper online
I’ve been away from the blog a couple of weeks, and while some of that can be attributed to preparation and work in Ghana with Karen Grepin, a lot of the blame can fall on a recent debate I’ve been engaged in with a journal.
For a couple of years now, I have been working on a paper with Susan Watkins and Patrick Gerland on AIDS Exceptionalism, particularly looking at the opinions of ordinary Malawians on whether HIV/AIDS programs should be prioritized. It is almost impossible to count the number of versions this paper has gone through, and the number of rejections it has faced at major journals is humbling. Finally, we got a “revise and resubmit” green light from the Bulletin of the World Health Organization (BullWHO), a/the leading journal in the global public health (see impact factor). The comments from reviewers were a bit brutal (if not fair) and it took some time to revise. The article was accepted after resubmission and I was elated that this paper which seemed the albatross around my neck would finally find a home (and a good one, at that!).
About a week after the acceptance, however, I got this email:
Dear Dr Dionne,
In the course of editing the paper that was reently [sic] accepted for publication in our journal, we found an essentially identical working paper describing the same study on the website of the University of California in Los Angeles.
Because this constitues a case of duplicate publication, we now have no choice but to reverse our decision to accept your paper for publication.
It turns out my co-author had submitted a previous working draft of the paper to her center for posting on their Working Papers site. I made a plea to the Bulletin of the WHO to reconsider (actually, I made three, but this one gives the general idea):
…You are right that the working paper posted online is a previous iteration of the paper that we submitted…
Still, the UCLA CCPR Working Papers directory is not a publication. I did not know that the paper was posted there (it was shared by my colleague, Susan Watkins, who is currently affiliated with the CCPR). CCPR asks all of its affiliates to share works in progress in an aim to show the funding organizations that the center is making progress towards its goals. Even the CCPR knows that the repository is not a resting place for papers, but expects that they will later be published in peer-reviewed journals…
But, their final decision was to rescind their decision to publish our paper:
Dear doctors Watkins and Dionne,
We understand that you have not wilfully attempted to publish your paper in duplicate. However, the Bulletin’s Managing Editor, Dr Laragh Gollogly, agrees that from our journal’s standpoint this is a case of duplicate publication, regardless of the institutional nature of the website where your working paper was posted. Furthermore, a change of title would be useless because the two versions of the text would remain nearly identical in content. Thus, we are regrettably compelled to reverse our decision to publish your manuscript but hope that this alerts you to the risks of posting unedited research papers on line prior to publication, at least in the health sciences, particularly when due disclosure is not made at the time of submission.
We regret not being able to offer you a more palatable solution and wish you well with the publication of your future research.
I think that’s a shame. You see, the paper as a CCPR Working Paper had been downloaded 21 times (this likely counts me, the BullWHO editor, and various friends on Facebook who clicked the link following my initial post about the saga there). The BullWHO claims to have a wide readership. One colleague summed up the tradeoff perfectly:
Since they have judged it worthy of appearing in the Bulletin, it seems perverse to deny their large number of readers the opportunity to read the work because a tiny few might have come across it already.
Though I appreciate the policy to reject a “duplicate publication”, there is no mention of what constitutes duplicate publication on their information for contributing authors page, and I still contend our situation does not fit that description.
There is an appeal process that we could pursue (and I’ve been advised by senior colleagues via the Facebook that I should), but because the rejection comes from the managing editor who steers the editorial board that would review the appeal process, it seems to me as a fruitless endeavor. Instead, we have decided to place the paper elsewhere.
The first journal that we have decided to send it to instead, however, has a clearly stated policy about online working paper “publications”:
Papers or parts of papers that have been published or accepted for publication elsewhere (including being posted on the Internet) are not considered.
We may decide to send it there in the end, but all of this makes me wonder about publishing in health journals. You see, the current norm in social science seems to be to share draft versions of your papers before publication (see, for example, SSRN). Simultaneously, there has been a lot of talk in leading health journals of the value of social science research of health problems (e.g., this paper published in The Lancet). The conflict here is obvious. A discipline has a norm of sharing papers before publication, and another discipline has a norm of not sharing. I suspect any journal trying to publish work from both disciplines will likely exclude the work from one.
…the necessity of going through the peer-review process has lessened for high-status authors: in the old days peer-reviewed journals were by far the most effective means of reaching readers, whereas with the growth of the Internet high-status authors can now post papers online and exploit their reputation to attract readers.
Of course, I’m not a high-status author. Worse yet, I’m pre-tenure. I actually need journal publications to prove my worth in this line of work. If only my department and my discipline counted that CCPR online working paper as a publication (as does the BullWHO)…