Walt at Random

I’m adding a prefatory wrap to this post, having realized that I was hasty in apparently attacking the statements in a UK factsheet–when what I’m actually questioning is the definition of the universe involved.

I’m not changing the post’s title because that just gets confusing–and I’m not withdrawing “Are facts irrelevant?” because it’s a question that keeps arising when I read various discussions. I do believe that a small sampling operation done by academics or official bodies will be treated considerably more seriously than a 100% scan done by the great unwashed (“independent scholars” like me), and that “facts” can be very fuzzy things, especially when definitions of universes are involved.

And I do believe that it is at best misleading to treat every journal whose publisher says “Sure, if you pay us a huge sum of money we’ll make your article OA, but we won’t lower subscription prices” as being equivalent to a true Gold OA journal in defining the OA market. Misleading, but not technically falsifiable. If it’s true that UK scholars are gravitating toward the double-dipping approach, that’s a shame–but maybe UK universities and libraries are so well-funded that it’s not an issue. I wouldn’t know; poor American universities like Harvard and UC clearly aren’t.

Now that I’ve confused the situation even more, and felt every day of my 70 years in the process, I’ll step away and provide the original post.

The original post follows, with some updates at the end.

First off, a little deal: Lulu will pay for the USMail shipping (or give 50% off ground shipping) if you use coupon code FGA915 now through September 21, 2015. Not a big savings, but something (apparently only one person took advantage of the 20% sale, but that’s 100% of the book sales to date.)

Then this: “Monitoring the transition to Open Access,” a September 2015 “factsheet” from UniversitiesUK, includes this paragraph:

Levels of Article Processing Costs (APCs) vary widely. Most journals charge between £1,000 and £2,000; only small minorities, concentrated in a few publishers, charge either less than £1,000 or more than £2,000.

I don’t find qualifying statements (“most journals that UK authors publish in…” or “most biomed journals that UK authors publish in…”) so I’ll take that at face value.*

My spreadsheet, based on actual visits to every journal in the Directory of Open Access Journals as of June 2015, shows the following (assuming $1,550 to $3,100 as the dollar equivalent of the ranges in that paragraph):

• More than £2,000 ($3,100): 23 journals publishing 3,091 articles in 2014.
• Between £1,000 and £2,000 ($1,550 to $3,100): 548 journals publishing 73.609 articles in 2014.
• Some APC but less than £1,000 ($1 to $1,550): 1,899 journals publishing 198,996 articles in 2014.

I am unsure of the arithmetic system in which 1,899+23 is a “small fraction” of the universe of DOAJ-listed APC-charging journals (2,470 in this count). It looks like 77.8% to me (and 73% of the articles) to me.

Here’s the thing: They’re funded. I’m not. They’re official. I’m not. And that “fact” sheet will get a lot more coverage than my work: count on it.

Am I going to respond directly to them? I’m getting old and tired, and wouldn’t really know who to respond to anyway. And there’s probably some huge unstated asterisk in their paragraph that makes it correct for some set of circumstances.

Anyone wanna bet that it will not be used as though there was no such asterisk?

*UPDATE: OK, I now realize the hidden asterisk: This factsheet is treating “hybrid” journals as part of the overall universe, I don’t doubt that “hybrid” journals typically charge very high APCs–after all, that’s part of their game.

Therefore, what we really have is further evidence that “hybrid” OA is mostly a scheme to make sure that high publisher profits aren’t disturbed by OA.

What I study is Gold OA–journals that immediately make all of their peer-reviewed articles available for online reading. And apparently UK authors are more comfortable with “hybrid” OA.

Second update: 

On one hand, I’m tempted to pull this post as overreacting, since there is a way of defining the universe that may make their assertions correct (if half of subscriptions are actually hybrid–which seems high, and if so impresses me that the UK group was able to determine the status of 20,000+ journals, then, yes, assuming most of those APCs are in the somewhat outrageous £1,000 to £2,000 range, it only takes, say, 6,000 such hybrids to make 1,922 into a “small fraction.”

I simply question whether it’s meaningful to call “hybrid” journals OA journals in any real sense, unless they’ve actually published a significant number of OA articles and stated a clear policy for adjusting subscription prices to reflect the extra income from APCs. To me, double-dipping isn’t OA; it’s a scheme to protect publisher profits.

So I’m going to leave this post up. I believe gold OA is an important part of OA. Serious gold OA–best defined by the contents of DOAJ–is what I look at. You can define the universe in such a way as to swamp that area of study, to be sure.

Zoology includes veterinary medicine as well as marine biology. The group includes 178 journals, which published 9,581 articles in 2013 and 9,176 in 2014.

Grade

Table 29.1. Zoology journals and articles by grade

Table 29.1 shows the number of journals and 2013 articles for each grade; free, pay and unknown numbers; and average articles per journal. Boldface percentages are of the whole group; others are of the grade above.

As usual, journals with APCs tend to publish more articles than free journals (except for the anomalous questionable free journal with lots of articles).

The relatively small number of D journals includes these subgroups: C (apparently ceased), six journals with a total of two articles in 2013; D (dying), two journals, 17 articles; E (erratic), four journals, 90 articles; H (hiatus?), four journals, 157 articles; S (small), nine journals, 39 articles.

Article Volume (including all of 2014)

Table 29.2. Zoology journals and articles by date

Table 29.2 shows the number of free and APC-charging journals that actually published articles each year, the number of articles published and the percentage that didn’t involve APCs. “Unknown” journals (those that apparently have, but do not clearly state, APCs) are omitted. There are some journals that don’t publish any articles in any given year.

The percentage of free journals is slightly below average for STEM and has stayed fairly steady; the percentage of free articles is somewhat above average for STEM, and seems to bounce back and forth.

OA activity in zoology does appear to have slowed down since 2012 and, very slightly, since 2013.

On a journal-by-journal basis, 80 journals published more articles in 2014 than in 2013; ten published the same number (including three with no articles in either year); 88 published fewer articles in 2014. Looking at significant changes, 58 (33%) published at least 10% more articles in 2014; 43 (24%) published about the same number; 77 (43%) published at least 10% fewer articles, including a dozen journals with articles in 2013 but none (so far) in 2014.

Table 29.3. Zoology journals by peak article volume

Table 29.3 shows the number of journals in each size category, 2013 articles for those journals and the no-fee percentages. There are no prolific zoology journals, and as usual the free percentages go down as the volume of articles goes up.

Fees (APCs)

Table 29.4. Zoology journals and articles by fee range

Table 29.4 shows how many journals are in each fee range and 2013 articles for each fee range; this table also omits unknown journals. For the full study, fee ranges are based on actual quartiles, so each of the first four %Fee figures should be around 25%. Notably, zoology journals tend to be lower-priced.

There is essentially no correlation between APC level and article count, either peak count or 2013 count: the numbers are 0.01 and 0.03 respectively.

Starting Dates and the Gold Rush

Table 29.5. Starting dates for zoology OA journals

Table 29.5 shows zoology OA journals by starting date and the percentage of journals started in each date range that currently doesn’t charge APCs. The “gold rush” for DOAJ as a whole is a rapid increase in APC-charging journals beginning in 2006 and slowing down after 2011; for zoology, the rapid growth started in 2002-03 and resumed in 2006-11.

Figure 29.1 shows essentially the same information in graphic form, omitting unknown journals.

Figure 29.1. Zoology journals by starting date

Table 29.6. Zoology articles per journal by starting date

Table 29.6 shows journals that published articles in 2013, the number of articles and average articles per journal. It appears that journals founded in 1994-97 publish a lot of articles, with other deviations as well.

Overall, zoology and related topics show fairly typical STEM patterns but a somewhat atypical drop in activity since 2012.

Definitions and notes

See The Open Access Landscape: 1. Background for definitions and notes

This completes the series of subject posts based on my study of 6,490-odd journals–but I’ve now completed a much broader study, encompassing essentially all of DOAJ as of early June 2015 and using Google/Chrome translation to make sense of 3,00h0-odd non-English journals (although 20 still couldn’t be handled). The study–The Gold OA Landscape 2011-2014–is now available as a PDF ebook and will shortly be out in paperback.

ALA TechSource has now mailed out copies of Open-Access Journals: Idealism and Opportunism to subscribers of Library Technology Reports–and if you’re not a subscriber, you or your library can purchase the whole report or chapters of it from the LTR website.

You can read the first chapter for free, including the big numbers (it’s standard practice for LTR to make first chapters open access).

I think this is worth reading and buying. It’s the distillation of the first general phase of my multiphase investigation into the realities of gold OA–looking at all the journals in the Directory of Open Access Journals that have English-language interfaces, and tracking their articles from 2011 through the first half of 2014.

It’s concise (38 pages) with selected tables and graphs to provide a quick but thorough summary of gold OA at that time, with concluding chapters on sideshows and leftovers (specifically a certain “blacklist”), dealing with OA journals (including a quick way to spot most troublesome journals and “journals”), and libraries and OA journals.

It’s even been professionally edited.

I’m proud of the report and think it’s worth your time. (Admittedly, I’m within days of releasing a much broader report, covering thousands more journals and all of 2014, but that report won’t be as concise and will completely lack the key additional material in Chapters 6-8. You need both.)

Technology journals were distinguished from engineering journals based on broader subject coverage or the word “technology” in the title. As with engineering, it’s a somewhat fuzzy group. The group includes 138 journals, which published 9,738 articles in 2013 and 10,718 articles in 2014.

Grades

Table 28.1. Technology journals and articles by grade

Table 28.1 shows the number of journals and 2013 articles for each grade; free, pay and unknown numbers; and average articles per journal. There are no free C journals or unknown D journals. Boldface percentages are of the whole group; others are of the grade above.

These journals are unusual in that, other than the high-priced A$ journals, free A and B journals actually average more articles than those charging APCs—and unfortunate in that the handful of questionable C journals published lots of articles.

D journals include these subgroups: C (apparently ceased), six journals with no 2013 articles; D (dying?), five journals with 28 articles; E (erratic), four journals with 183 articles; H (hiatus?), six journals, 156 articles; S (small), three journals, 19 articles.

Article Volume (including all of 2014)

Table 28.2. Technology journals and articles by date

Table 28.2 shows the number of free and APC-charging journals that actually published articles each year, including all of 2014; how many articles those journals published; and what percentage was free. Journals with unknown APCs are omitted, and there are always some journals not publishing in any given year.

The percentage of free journals is fairly typical of STEM—but unusual in that it hasn’t changed significantly. The percentage of articles in those journals is high for STEM, and has actually increased since 2012, the only year in which a majority of articles involved APCs.

Technology OA is clearly increasing, at a reasonably brisk rate that has slowed down slightly since 2012. Looked at on a journal-by-journal basis (and including unknowns), 67 journals published more articles in 2014 than in 2013; 13 published the same number (in five cases, no articles in either year); 58 journals published fewer articles in 2014. For significant changes, 59 journals (43%) published at least 10% more articles in 2014 than in 2013; 28 (20%) published roughly the same number; 51 (37%) published at least 10% fewer articles in 2014, including six that published articles in 2013 but not (yet) in 2014.

Table 28.3. Technology journals by peak article volume

Table 28.3 shows the number of journals in each size category (based on peak articles in 2011-2013), articles for journals in that group, and what percentage doesn’t involve APCs. There were no prolific technology journals in 2011-2013, but one journal has passed the 1,000-article mark in 2014. APC-charging large journals published proportionally more articles than free ones, where APC-charging sparse journals published slightly fewer articles than free ones—but these numbers don’t particularly follow typical patterns except for the lower free percentages in large journals.

Fees (APCs)

Table 28.4. Technology journals and articles by fee range

Table 28.4 shows the number of journals in each fee range and 2013 articles for those journals. For this study as a whole, the first %Fee column will be about 25% in each cell, as the fee ranges are based on actual quartiles. Here, very few journals charge the highest fees—but those journals publish a disproportionately large number of articles. Oddly enough, journals with nominal fees also publish a disproportionately large number of articles.

There’s a slight positive correlation between peak article levels and APC amount and a slight negative correlation between 2013 article counts and APC amounts. Neither correlation is large enough to be statistically meaningful (0.21 and -0.18 respectively).

Starting Dates and the Gold Rush

Table 28.5. Starting dates for technology OA journals

Table 28.5 shows technology OA journals by starting date, including the percentage of journals started in a given date range that currently charge APCs. Note that there were only three journals prior to 1998, all of them free. The gold rush for APC-charging journals isn’t terribly evident in Table 28.5 (for DOAJ as a whole, the percentage of APC-charging journals rises considerably in 2006 through 2011), but there is a surge of sorts later on, as shown in Figure 28.1. Note that there are very few APC-charging journals until 2008-09—but more new APC-charging journals than free journals after that. (Figure 28.1 omits 1960-69 and 1990-97.)

Figure 28.1. Technology journals by starting date

Table 28.6. Technology articles per journal by starting date

Table 28.6 shows journals that published articles in 2013, how many articles they published, and average articles per journal. The anomalies are obvious: the very large journal from the ’70s and large journals from the turn of the century.

Overall, OA technology journals tend to have a higher percentage of articles in free journals than most of STEM, but a fairly typical percentage of non-fee journals. The topic shows continuing gains in article count (with article counts in non-fee journals actually increasing faster than in APC-charging journals since 2012).

Definitions and notes

See The Open Access Landscape: 1. Background for definitions and notes

If you’re interested in a book-form version of this material (with an additional bonus graph and probably some additional analysis added in each chapter), let me know, either in a comment or by email to waltcrawford at gmail dot com. If you’d like to see this research continued, please contribute to Cites & Insights.