I have not tried to reconcile all these numbers, which clearly are important as the basis of various percentages. However I may be able to throw some light on how there could be some way forward.
The key to all this is the ISSN-L which is a linking field in each ISSN record that provides the equivalent of a family name for the print, electronic and other formats of a serial. More on that below. Clearly Heather is doing some useful work and she might find the following helpful. The numbers come from papers given at the ISSN General Assemby which I attend as an Observer, and are cited in a paper I am giving at the UNESCO Conference this coming September: Archiving The World's E-Journals: The Keepers Registry As Global Monitor, Peter Burnhill1, Françoise Pelle2, Pierre Godefroy2, Fred Guy1, Morag Macgregor1 and Adam Rusbridge1 (1. EDINA, University of Edinburgh 2. ISSN-IC, Paris) This is the Memory of the World conference on 26-28th September being held in Vancouver (BC, Canada) - so we may get to meet Heather who I note is at Simon Fraser. First some numbers. As of the start of 2012 there were over 97,500 eISSNs issued by the national centres of the ISSN Network - of which a subset are e-journals. 100,000 would seem a good working number to use, but see http://www.issn.org/2-22638-ISSN-and-electronic-publications.php. In a paper published in 2010, we quoted that there were 60,000, http://www.era.lib.ed.ac.uk/handle/1842/3451 This number has risen considerably over the past 5 to 8 years, from about 13,000 (must find that exact date!), as the result of both the increase in e-serials (including journals, magazines of all sorts, newspapers etc) and the efforts of the ISSN Network working to improve the assignment (and use) of the ISSN for the electronic format. NB The Directory of Open Access Journals (DOAJ) requires ISSN assignment and so are all in that number, 100,000. On the matter of geography, the following may be of interest. About 20% of those eISSNs were issued by the Library of Congress with respect to USA as place of publication. UK has assigned about 10%. Netherlands, Germany and Brazil have each assigned about 4.5%. There are relatively low numbers for India and China/HK given their size. Egypt and South Africa assigned most for that continent, Egypt also having the largest assignment for the Middle East. Collectively there are a large number of eISSNs assigned in Latin America and the Caribbean. Back to the ISSN-L, for more on this see http://www.issn.org/2-22637-What-is-an-ISSN-L.php and to download the ISSN Table, see http://www.issn.org/2-24117-Download-the-ISSN-ISSN-L-table.php The Keepers Registry, found at http://thekeepers.org/thekeepers/keepers.asp uses the ISSN-L as its key field to record what e-journal content is being looked after by seven of the leading archiving organisations. This means that if the wrong ISSN is in the metadata then we can cope: ie if the print ISSN is being used for something in electronic format then the ISSN-L enables linkage to the ISSN for that. Apart from its intrinsic value - and all should send us much love (and flowers) - this illustrates the significance of the ISSN-L in enabling co-location that facilitates search, retrieval and delivery across all media versions for services like Open URL, library catalogues, search engines, or knowledge bases. Obviously it also is the way to do effective 'de-duplication' if that is seen as the problem. And I suppose I should mention the blog which is now linked from the Keepers Registry - it began as a blog for the projec called PEPRS (piloting an e-journal preservation registry service) http://peprs.blogs.edina.ac.uk/ And to come full circle, the current page of that blog makes reference to a Pecha Kucha presentation at OR2012 on "E-journal Preservation and the Archival Value of the Authors' Final Copy" - the ' intended in the plural :) Enjoy and hope this all helps. Peter On Sat, 4 Aug 2012, Niamh Brennan wrote: > The unexpectedly high figure is due to Ulrich's counting the print > version and the electronic version of a journal title as separate > journals. > > Example (using a smaller dataset): a search for active journals which > are academic/scholarly and peer reviewed/refereed filtered by Country of > Publication=Ireland shows the total number of journals as 175. > > These are listed in Ulrich's as comprising the following: > > Print (77) > Online (65) > Microform (33) > LooseLeaf (1) > > However when the duplicate titles are (manually) removed from the > downloaded spreadsheet the number of unique journal titles is 154. > > Heather Morrison has commented on this issue here: > > http://poeticeconomics.blogspot.ie/2012/05/about-30-of-peer-reviewed-scholarly.html > > where she says: > > 'we don't have a count of how many peer-reviewed journals there are in > the world. We tend to use Ulrich's as a surrogate, however this list > reflects a strong English-language / western bias, e.g. Ulrich's would > only include a very tiny fraction of the academic journals from China. > > Something else to keep in mind is that is has become more difficult to > assess the number of peer-reviewed journals in Ulrich's, because the > default search does not deduplicate for multiple formats (i.e. a quick > search for academic / peer-reviewed / scholarly journals will yield two > titles when a journal is produced in both print and electronic form). > > My latest count of active, peer-reviewed scholarly journals from > Ulrich's using deduplication from Dec. 1, 2011 is 26,746. My method and > calculations are shown here, in this appendix of my draft thesis: > > http://pages.cmns.sfu.ca/heather-morrison/appendix-c-how-many-active-scholarly-peer-reviewed-journals/ > " > > Heather's methodology looks like the best we can do with Ulrich's. > However when her method is applied to my example above with the small > dataset of Irish journals (77 print + 6 electronic only = 83) the result > falls considerably short of the actual number of unique journal titles > for that dataset (154 - deduplicated manually). Perhaps it works better > with a larger dataset. > > It would be very helpful if Ulrich's would provide a simple > 'De-duplicate' option on its search results. > > Meanwhile it would be an easy matter for Ulrich's to provide us with the > data. Let's ask them! > > Best wishes, > > Niamh > Niamh Brennan > Programme Manager, Research Information Systems & Services > Ussher Library, Second Floor, > Trinity College Library Dublin, > ________________________________________ > From: [email protected] [[email protected]] On Behalf Of Sally > Morris [[email protected]] > Sent: 03 August 2012 21:53 > To: 'Global Open Access List (Successor of AmSci)' > Subject: [GOAL] Re: Update on Ulrichs estimate of total number of active > peer-reviewed journals: 55, 311 > > I find this figure very surprising. What appears to be the same search, > carried out this past March, came up with a (to me) much more credible > figure of 27,566. > > Sally > > > Sally Morris > South House, The Street, Clapham, Worthing, West Sussex, UK BN13 3UU > Tel: +44 (0)1903 871286 > Email: [email protected] > > > ________________________________ > From: [email protected] [mailto:[email protected]] On Behalf Of > Stevan Harnad > Sent: 03 August 2012 20:08 > To: Global Open Access List (Successor of AmSci) > Cc: LibLicense-L Discussion Forum > Subject: [GOAL] Update on Ulrichs estimate of total number of active > peer-reviewed journals: 55, 311 > > For years now, I've just been re-using an old > Ulrich<http://ulrichsweb.serialssolutions.com>'s estimate of about 25,000 for > the total number of active peer-reviewed journals. > > Prompted by a recent query from someone, I've checked again, and -- unless > I've made a mistake in my search -- the number now seems to have doubled to > 55,311. > > The parameters I used to get this figure were (1) Active, (2) Journal, (3) > Academic/Scholarly, (4) Refereed/Peer-reviewed, > > A further breakdown shows that of these 55,311 active peer reviewed journals, > > 23,527 (43%) are available online > > 9,354 (17%) are indexed in Thomson-Reuters-ISI's Journal Citation reports > > 6,962 (13%) are open access journals (freely available online) (Gold OA, > presumably not including Hybrid Gold). > > 769 (11%) of the 9,354 Thomson-Reuters-ISI-indexed journals are open access > journals > > -- > > According to the last estimate of journals indexed by SHERPA/ROMEO (which > does not include all the journals indexed by Ulrichs, but does include most > of the top journals indexed by Thomson-Reuters-ISI): > > 60% of journals recognize the author's right to provide immediate, > un-embargoed open access upon self-archiving their final drafts in their > institutional repositories. > > -- > > It would be helpful if others could check and confirm these figures. > > Stevan Harnad > > > _______________________________________________ > GOAL mailing list > [email protected] > http://mailman.ecs.soton.ac.uk/mailman/listinfo/goal > ********** ********* ******** ******* ****** ***** **** *** ** * Peter Burnhill Director, EDINA national data centre & Head, Data Library Causewayside House University of Edinburgh 160 Causewayside Edinburgh EH Scotland, UK tel: +44 (0) 131 650 3301 fax: 3308 mobile: +44 (0) 774 0763 119 Email: [email protected] URL http://edina.ac.uk -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. _______________________________________________ GOAL mailing list [email protected] http://mailman.ecs.soton.ac.uk/mailman/listinfo/goal
