[ The following text is in the "utf-8" character set. ] [ Your display is set for the "iso-8859-1" character set. ] [ Some characters may be displayed incorrectly. ]
Thank you, Marc, for these useful remarks. I quickly based myself on the statement found at the beginning of the document but failed to do the sampling that Marc did. He appears to be right as to the number of journal entries. I will enquire further with the authors and report back to the list. As far as peer review is concerned, I believe all journals in Web of Science and in Scopus are peer reviewed. The ERIH and AERES lists are comprised of peer reviewed journals exclusively, so far as I know. As a result, I believe we can trust the scientific and scholarly (i.e. peer reviewed) status of these journals. With regard to the DOAJ list, this is another important point. I have already pointed out to the authors that further lists could be consulted such as Redalyc, Open JGate, etc. Australian lists also exist already and a new one will apparently appear soon. In short, much further work needs to be done. The list is patently incomplete and the authors are aware of this. The point of my remark was not to clarify the proportion of OA journals compared to the total number of journals. It was to question the assertions commonly encountered that place the total number of scientific and scholarly journals (peer reviewed) at somewhere between 14,000 (Michael Mabe's estimate) and 21-24,000 (Stevan Harnad's estimate). Personally, I suspect that it must be closer to 40 or even 50,000 journals, many of whom are part of the "lost science" coming from research in so-called peripheral countries.Many scientific and scholarly journals do not appear in the major bibliographies, including Ulrich's. But then many if not most of the major bibliographies originate in OECD countries (i.e. rich countries) with the possible result of a "Western" or "developed" bias. It would be useful to organize a centralized database that would verify and collate all these sources. In this fashion, a reliable list of journals would gradually be built. But Marc's correction is a useful correction to the authors' figure. I simply took it at face value, and I should not have. And I also agree that this document would be more useful in a database format. Jean-Claude Guédon Le dimanche 16 août 2009 à 11:11 -0400, Couture Marc a écrit : On August 4, 2009, Jean-Claude Guédon wrote: > > A 721-page list of social science and humanities journals comprising around > 20,000 titles has been compiled. > This list is limited to SSH journals > I downloaded and examined the 721-page document compiled by JournalBase and available at http://www.cybergeo.eu/index22492.html One looks forward getting access to the database (as promised by the authors) instead of a huge text-based table, but one can readily draw some conclusions upon simple inspection: - The number of entries stated by Jean-Claude and given on the Web page (20 000) would mean an average of 27 titles per page. One can easily verify that the actual number is much lower. In fact, based upon a 15-page sample, I obtained an average of 12 different titles per page, for a total of the order of 9 000 titles (still quite a large number). One indeed obtains an average of about 30 when one includes the multiple entries one finds for most journals (one entry for each category, plus some journals appearing twice). - Although this is a fairly intuitive conclusion, the list appears indeed to comprise mostly peer-reviewed journals. - Although the authors indicate that the list "includes the information on open access journals indexed in the DOAJ", JournalBase features only 350 DOAJ journals, while one can estimate the number of social science journals in DOAJ to be more in the range 700-900 (depending upon the way one defines an SSH journal, and interprets the keywords and categories in DOAJ lists). It seems that they didn't use DOAJ as a source for journal titles (and DOAJ is not listed in the "Sources" column), but used it to check the OA status of the journals they found in other lists (Scopus, etc.). More data and analyses are thus needed to get a reliable estimate of the percentage of OA scholarly journals. One gets 17% if one uses DOAJ's and Ulrich's data (4000 OA journals over a total of 24 000), but only 4% with JournalBase data. Although the ratio for SSH journals could well be lower than the overall ratio, I don't think we should but too much emphasis on either figure. Marc Couture Télé-université (Université du Québec à Montréal) mcout...@teluq.uqam.ca http://www.teluq.uqam.ca/spersonnel/mcouture/home.htm