I’m partway through writing Gold Open Access Journals 2011-2015, as some of you are well aware. That book is based on an exhaustingve survey of journals in the Directory of Open Access Journals (DOAJ) as of December 31, 2015: their APCs (or lack thereof) and article counts 2011-2015.
When I became aware of the big cut on May 9, 2016, with DOAJ dropping some 3,300 journals that had not submitted requests to be included following the new rules–after more than a year of publicity and repeated email requests–I decided it was worth discussing at the end of the book. (It doesn’t really affect the book: these journals were in DOAJ on 12/31/15.)
That continues to be my plan, but since I put together a matrix workbook to make some tables and graphs for the book easier and more consistent, I thought I could do a quick workup now–preliminary, tentative, but probably pretty close.
Update 5/22/16: I’ve now completed a more careful matching of a slightly later DOAJ dataset, resulting in 2,948 dropped journals. That change–nine more journals that are still there–will change a few numbers, but not by much. The revised figures will show up in the book, probably within the next two weeks.
The Overall Picture
URL and journal title matching shows 2,957 journals missing on May 10, 2016 that were there on December 31, 2015.
- First good news: More than half of the journals I excluded from the study are now gone–316 of 620. That includes more than two-thirds of journals with hidden or missing APCs (and I’m guessing the rest have filled in the information) and almost half of the unreachable and unworkable journals. Unfortunately, it includes less than one-third of the journals showing signs of malware. (Curiously, it includes the only journal I couldn’t include because of translation problems–and, perhaps less curiously, more than 70% of those where it was impossible or too cumbersome to count articles by year.
- Oddly, while three-quarters of journals with no 2014 or 2015 articles are gone, as are most journals with no 2015 articles, only 38% of apparently-cancelled journals and 36% of journals seemingly too small for the new DOAJ are gone.
- Ignoring excluded journals, just under 26% of journals are gone–but, not surprisingly, that breaks down to only 1.4% of APCLand journals and 29% of OAWorld journals. (If you’re not familiar with those terms, read the current Cites & Insights.)
A Few Specifics
- Only 23% of journals with 2015 articles are gone–26% of free journals, 17% of APC-charging journals.
- The article count is down 22%.
- Dropped free journals have been declining in article count: the dropped group includes 33% of articles in those journals in 2011, down to 27.5% in 2015.
- The largest (600+ articles) and smallest (0-19) journals disappeared more frequently than midrange journals.
- Among fee-charging journals, those with lower fees disappeared more often than those with higher fees–30-31% for $2-$199 and $200-$599, 10% and 3% for $600-$1,399 and $1,400+.
- Separating APCLand as a virtual region, the highest percentage of dropped journals is not in the global South: It’s what I call Pacific/English [with apologies to Quebec]–Australia, New Zealand, Canada and the United States: 39%. Asia had the second highest percentage of dropped journals, 35% (and by far the highest percentage of dropped articles, 47%), followed by the Middle East and Latin America (both 33%, but Latin America’s article loss is much lower.) The lowest percentage of dropped journals is in Eastern Europe, at 18%. Given that OAWorld’s 29% is the baseline here, only Pacific/English and Asia had outsize losses.
- Looking at categories of publishers (explained in the book), society and university journals dropped marginally more than average and traditional and multijournal OA publishers dropped substantially less (around 17% in both cases); the biggest losses are among “miscellaneous publishers,” those with only one or two journals.
- 34 countries had no losses, although that’s only 97 journals.
- The highest journal losses (by number) come from the United States, Brazil, India, Spain and Turkey–but the highest article losses come from India (50% more than the U.S.), the United States (more than twice Brazil’s number), Brazil, China, Turkey and Japan (Spain is 12th).
- Percentagewise, among countries with a fair number of journals, Japan has the highest article loss. Looking at the five countries with the largest numbers of dropped journals, the U.S. lost 40% of journals claiming to be published here but only 19% of articles; Brazil lost 27% of journals and 17% of articles; India dropped 36% of journals and 42% of articles; Spain dropped 20% of journals and articles; and Turkey dropped 37% of journals and 47% of articles.
I suspect this will serve as a wake-up call for a fair number of university and society publishers and for publishers in some countries. In other cases…well, I see a baker’s dozen of publishers with 10 or more dropped titles (the largest is 45), and there are at least two or three of those that may not be missed.
Again, this is all preliminary off-the-cuff quickie subject to change comment. The book will be free (in PDF form) when it comes out, and that final chapter may be part of a C&I extended excerpt: those numbers should be better.
