In the latest SPARC Open Access Newsletter, Peter Suber posts the results of research with Caroline Sutton on scholarly society publishers with open access journals. At its core is a list of open access journals affiliated with scholarly societies and various characteristics associated; the post contains some analysis. The list and analysis also considers society journals with hybrid open access options.
The information is quite interesting, and practical (for decision-makers and OA advocates). The authors note that they’d like to explore the topic in greater depth. Here, then, are my comments — hopefully useful for Phase Two.
The number of societies involved, and the number of journals published, is large, accounting for 16% of the Directory of Open Access Journals (450 out of 2900). The number of open access journals is considerably larger than the number of journals with hybrid OA options (450 vs. 73).
The geographic base of OA-publishing societies is broad — 57 countries/regions — compared to just 5 countries/regions with hybrid journals (93% of which is composed of the US and UK). (Note: I’m not clear whether the “geographic location” listing in the chart is based on the society’s location or the journal’s. For example, European Physical Journal is published by Societá Italiana di Fisica; the location is listed as “Europe”. I’d appreciate if the authors could clarify this.)
Similarly, the publisher:journal ratio for open access journals is much lower than that of hybrid journals. “Most societies publishing OA journals publish just one. [...] Only five societies publish just one hybrid journal.” To suggest some possible causes for this, it might be that societies publishing multiple journals do so because publishing is a profitable endeavor for them, and therefore they’re more concerned about loss of revenue. Or perhaps societies with more emphasis on publishing are simply more hesitant to make rash decisions with their journals. Maybe it’s because their executives receive bonuses based on the financial performance of the publishing division. These might be avenues for the authors to explore in Phase Two.
Might there be a link between the geography and the prevalence of open access? I’d expect there might be some network effect or peer pressure, but that’s not what I mean. I frequently hear that journal profits often subsidize other society activities — at least in the U.S. Perhaps this isn’t the case in other countries, where perhaps journals are more likely to themselves be subsidized. I’d like to hear thoughts from anyone who has them.
The hybrid journals seem to pay more attention to copyright issues. This seems consistent with the self-selecting nature of hybrid OA: I can imagine some authors in no-fee OA journals having no opinion about, or even ignorant of the fact that, the journal is OA; clearly that’s not the case for hybrid journals. I’d be curious to see a comparison of the fee-based OA journals (only 17% of the list) with the hybrids (all fee-based); I suspect the difference might be insignificant.
Unfortunately, the study found that only 15 of the 450 OA journals state that they use Creative Commons licenses. The exact number might be a bit fuzzy, but it looks clearly as though most OA society journals don’t comply with the BBB definition of open access. (Which means, in my view, that we shouldn’t be calling them “open access”, but rather “toll-free” or some comparable appellation.)
There is a seeming underrepresentation of social sciences, arts and humanities among OA journals: 79% of the list are from science, technology, and medicine. The accepted wisdom is that this mirrors the divide among OA journals in general, regardless of publisher. I can posit several potential explanations for this, but regardless of cause, it only increases the importance of OA for scholarly monographs; more on this in a later post.
I hope, concurrent with the Phase Two research, the authors refine this for publication in a journal. That’ll make a more powerful citation and get it seen by more readers. It’ll also force the authors to iron out some of the niggling methodological issues, including those noted here and in the original post itself.
One note about the reliability of the data: The DOAJ is known as the definitive source for information about open access journals, but I’m not sure their data about hybrid journals is as comprehensive. I’d be keen to hear from anyone who knows about this, but until I do, I’m hesitant to draw many conclusions about hybrid journals. If many simply aren’t listed in DOAJ, that could account for wide variances in results.
Some information I’d like to see in Phase Two, in addition to the topics already noted in the original post and those noted above:
- information about society publishers with policies that permit self-archiving (from SHERPA/RoMEO)
- information about society publishers whose journals are delayed OA (though I’m not sure where one would find this information)
- when these journals became OA
- whether these journals were born OA or converted from pre-existing toll-access journals
- whether these societies also publish toll-access journals or have converted entirely to OA
- a more complete listing of impact factors / Eigenfactors for these journals
- a definitive listing per journal on whether the publication meets the BBB definition of OA
- more clarity on the definition of “scholarly society” (e.g. the World Health Organization is listed as a publisher here; it seems a stretch to say the WHO is primarily a scholarly society; same with the American Federation of Teachers, primarily a labor union)
A final note: The authors mention that their list of scholarly societies that publish OA journals is significantly longer than recent lists of societies opposed to OA government policies (specifically, the US National Institutes of Health self-archiving mandate). The authors do caution against a direct comparison, but they leave out the fact that their list of scholarly societies is worldwide, while presumably only societies based (or with substantial members / activities) in the US would speak out about policies of the US government. I count 82 unique OA-publishing societies based in the US, plus 8 unique US societies with a hybrid-option journal; adding the 12 unique international OA-publishing societies, there are perhaps 102 OA-publishing or -offering societies who might reasonably be interested in US policy, significantly less than the 425 the authors were using in their comparison.
Coincidentally, of the 9* US societies with a hybrid-option journal, four signed the Association of American Publishers letter opposing an NIH mandate:
- American Chemical Society
- American Dairy Science Association
- American Society of Animal Science
- American Society of Plant Biologists
Someone with more time on their hands can do the cross-check of the 82 OA-publishing US societies.
* Yes, I said 8 in the preceding paragraph: the American Physical Society publishes both a hybrid and an open access journal, so it was counted in the OA list and therefore not “unique” for that tally.