Given the recent publication of “Hybrid open access—A longitudinal study” by Mikael Laakso and Bo-Christer Björk, I thought it might be interesting to put together all the pieces: my study of DOAJ-listed journals, my study of “others” (using Beall’s generally pernicious lists as a source directory), and this study of hybrid articles.
Here’s what I come up with, complete for 2012 and 2013, partial for 2014 and 2015. “Questionable” for DOAJ includes journals with unstated/hidden APCs; for the gray segment, it includes a variety of things (see Table 3.4).
| 2012 | 2013 | 2013% | 2014 | 2015 | 2015% | |
| DOAJ |
438,644 |
493,475 |
69.8% |
560,036 |
566,922 |
65.0% |
| Gray/norm |
69,075 |
98,679 |
14.0% |
135,052 |
148,564 |
17.0% |
| SubNorm |
507,719 |
592,154 |
83.8% |
695,088 |
715,486 |
82.0% |
| DOAJ/Ques |
10,539 |
10,896 |
1.5% |
10,170 |
8,866 |
1.0% |
| Gray/Ques |
55,964 |
89,966 |
12.7% |
120,131 |
148,399 |
17.0% |
| SubQues |
66,503 |
100,862 |
14.3% |
130,301 |
157,265 |
18.0% |
| Hybrid |
10,802 |
13,994 |
2.0% |
|||
| Total |
585,024 |
707,010 |
825,389 |
872,751 |
I believe this is as complete a picture of gold OA as we’re likely to get, although it does omit a few thousand articles where journals have malware or are otherwise resistant to article counts. I’d suggest a 5% margin of error—and also suggest, as I’ve long suspected, that hybrid OA is still within that margin of error, less than 5% of gold OA.
