From Peter Suber's December 2011 Issue of the SPARC Open Access Newsletter.
How many scholarly societies publish OA journals, and how many OA journals do they publish? Four years ago (November 2007), Caroline Sutton and I released the first edition of our inventory answering those questions, and today we release the second edition.
Cutting to the chase: Our 2007 list turned up 425 societies publishing 450 full or non-hybrid OA journals. Our 2011 list shows 530 societies publishing 616 full OA journals.
We're sure we overlooked some society OA journals in 2007 and we're sure we're still overlooking some today. If it weren't for that, we could say that the number of societies publishing OA journals grew by 25% in the last four years, and the number of their OA journals by 37%. Nevertheless it's hard to avoid the conclusion that both numbers are growing significantly.
The second edition of the list is a Google spreadsheet under a CC-BY license.
The 2007 edition is a downloadable Excel spreadsheet under a CC-BY license.
Also see the article Caroline and I wrote to accompany the 2007 list: "Society publishers with open access journals," SPARC Open Access Newsletter, November 2, 2007.
To help keep the new list accurate an up to date, we've opened the Google spreadsheet for public editing. If you notice any errors or omissions, please take a moment to fix them directly in the spreadsheet or send me an email about them. Because the new edition of the list is publicly editable and continuously updated, we don't plan to publish a third edition. However, we may publish new reports now and then on the current tallies.
The 2007 list counted full and hybrid OA journals separately. The 2011 list doesn't count hybrid OA journals at all. Hybrid OA journals are so risk-free for publishers, and consequently so numerous, that including them would have taken most of our time and overshadowed the more interesting phenomenon: societies so committed to OA that they would skip over the safety and low uptake of the hybrid option and publish full-OA journals.
For brevity, we'll say that a journal is a "society journal" if a scholarly society, publishes, sponsors, or owns it, or has adopted the journal as an official publication.
The 2007 list found scholarly OA journals published in 57 countries and regions, and the new list in 67 countries and regions. Among the countries joining the 2011 list and missing from the earlier list are France, Germany, Hungary, and Portugal.
The top 10 countries for society OA journals were the US (125 journals), India (92), Japan (89), Canada (20), UK (18), Brazil (10), Croatia (10), Korea (10), Romania (10), and Australia (8). At the other end of the spectrum, 17 countries published just one society OA journal each: Bangladesh, Belgium, Chile, Czech Republic, Denmark, Estonia, Hungary, Indonesia, Libya, Lithuania, Norway, Pakistan, Portugal, Russia, Singapore, Taiwan, and Ukraine. Some countries notable for academic publishing fell in between, for example China (5), Egypt (4), France (5), Germany (3), the Netherlands (2), and South Africa (2).
In 2007, only 15 (3%) society OA journals used CC licenses. In 2011, 92 (15%) do so, a small fraction but a distinct improvement. An additional 45 journals (7%) let authors retain copyright but do not publish under open licenses. Despite the improvement from four years ago, these are deeply disappointing numbers. As of last week (November 25, 2011) 1,727 or 24% of all the OA journals listed in the Directory of Open Access Journals used CC licenses. Hence, society OA journals use CC licenses at an even lower rate than OA journals in general.
Some background, for context: "OA repositories are rarely in a position to obtain the permissions needed for libre OA. Hence, we can't criticize or complain when most of their deposits are gratis, not libre. But OA journals can easily obtain the permissions needed for libre OA. When they don't offer libre OA, they have no excuse. This is one of the largest missed opportunities of the OA movement to date."
In 2007, 148 or 33% of society OA journals didn't describe their copyright policies on their web sites. In 2011, the percentage fell slightly (191 journals or 31%), a minor improvement. Again, despite the improvement, the numbers remain disappointingly low.
In 2007, most society OA journals (356 or 79%) were in the STM fields. In 2011 the ratio is about the same (483 or 78%). In 2011, 71 (12%) are in the social sciences, 43 (7%) are in the humanities, 10 (2%) are multidisciplinary, and 6 (1%) are in the arts.
In 2007, we found 75 society OA journals charging publication fees (17%), and in 2011 we found 141 (22%). In 2007, we found 12 (3%) charging submission fees, and in 2011 we found 4 (<1%). In 2007 we found 8 (2%) charging both publication and submission fees, and in 2011 we found 9 (1%). Putting these together, the numbers from both years confirm every other survey showing that the vast majority of OA journals charge no author-side fees of any kind. Overall, 450 (73%) society OA journals charge no author-side fees.
The most recent survey of all the OA journals in the DOAJ is Stuart Shieber's from 2009, which showed that 70.3% of OA journals in the DOAJ charged no publication fees. Our new numbers show that society OA journals charge publication fees even less often than OA journals in general (23% v. 30%).
Some publishers publish more than one society OA journal. In 2011, 17 publishers published two society OA journals each, five published three each, two published four each, and two published seven each. One (Springer) published nine society OA journals, one (Copernicus) published 15, one (WASET) published 21, one (BioMed Central) published 33, and one (MedKnow) published 64.
Conversely, some OA journals are associated with more than one society. In 2011, nine society OA journals were associated with two societies at once, three OA journals with three societies at once, one journal with five societies, and two journals with 10 societies.
In 2007, 182 society OA journals also published priced, print editions (40%). In 2011 the number grew but the percentage declined (233, 38%). This is a difficult business model to make work; hence the declining percentage makes sense. But the rising absolute number invites further investigation. Are more societies finding ways to succeed with dual (OA + TA) editions? Are more societies willing to go into the red to support dual editions?
The Journal of the Statistical and Social Inquiry Society of Ireland has an OA backfile stretching back 164 years, to 1847, the longest OA backfile of any society OA journal. Five society OA journals have OA backfiles of 100 years or more, 12 journals of 75 years or more, 21 journals of 50 years or more, 43 journals of 25 years or more, and 277 journals of 10 years or more.
We didn't collect data on how many non-OA society journals give standing permission for author self-archiving, i.e. how many are green rather than gold. Nor have we seen data on this elsewhere. If we're overlooking a relevant survey, please drop me a line.
Our purposes for this second edition of the list are the same as our purposes four years ago. First, we want to draw attention to empirical evidence correcting the widespread misimpression that scholarly societies are opposed to OA. Some are and some aren't. Or some are and many aren't. Second, we want to make it easy for society publishers thinking about OA to identify OA-committed societies in roughly similar circumstances, especially by field and nationality. We expect that society publishers deliberating about OA will prefer to consult with OA-experienced society publishers than with OA advocates who are not publishers or OA publishers who are not society publishers.
It appears that significantly more societies publish OA journals than join public statements or trade organizations opposing OA policies by funding agencies or universities. As Caroline and I put it in 2007, "[T]he length of our list changes the question from what makes the OA society publishers rare and special to what makes the OA holdouts hold out. How many of the objections or fears will turn out to be myths that can be answered by the publishers with actual OA experience?"
I thank Caroline Sutton for her efficient, long-term collaboration on this project. In 2007 she was a founding partner of Co-Action Publishing, an OA publisher of books and journals based in Scandinavia. Today she's the Director of Sales and Marketing at Co-Action and the President of the Open Access Scholarly Publishers Association (OASPA). Together we thank David Solomon and Bo-Christer Björk for the ISSNs to more than 300 of the journals on our new list, Emmeli Sjölander and Katharine Dunn for their assistance in preparing the new list, and Emily Kilcer for her assistance in digesting the numbers from the new list.