Seulement 25000 revues scientifiques dans le monde, vraiment ?

Eye exam (no. 2)

Image by Noir Imp, licence CC.

Dans son livre sur l’accès ouvert  (MIT Press), Peter Suber affirme qu’il y a un consensus pour dire que, toutes disciplines et toutes langues confondues, le nombre de revues dans le monde, quel que soit leur mode de diffusion (accès restreint ou accès ouvert), s’élève à 25 000.

Je ne peux pas m’empêcher de penser que cette évaluation ne s’appuie sur aucun élément fondé. Disposez-vous d’éléments qui permettraient de consolider cette estimation?

Premiers éléments

En fait, le livre donne deux éléments qui semblent contradictoires entre eux :

  • Thèse des 25 000 

« http://www.doaj.org Most observers estimate that there are about 25,000 peer-reviewed journals in all fields and languages, making the OA portion about 26 percent of the total. There’s some evidence that the average OA journal publishes fewer articles/year than the average toll-access journal, making the OA portion (by articles rather than journals) even smaller than 26 percent. If we supplement the number of peer-reviewed articles published by OA journals with the number of peer-reviewed OA articles published by toll-access journals but disseminated with permission by OA repositories, the portion goes up again. »

Source : Peter Suber, Open access.

  • Dans les faits, « plus de 100 000 » titres en accès restreint (sans compter, donc, les revues en accès ouvert

« In 2008, Harvard subscribed to 98,900 serials and Yale to 73,900. The best-funded research library in India, at the Indian Institute of Science, subscribed to 10,600. Several sub-Saharan African university libraries subscribed to zero, offering their patrons access to no conventional journals except those donated by publishers. »

Source : Peter Suber, Open access.

 


Marin Dacos

Marin Dacos (CNRS) is the founder and director of OpenEdition. He has been awarded two Google Digital Humanities Research Awards (2010 and 2011) and the CNRS Innovation Medal (2016). He is scientific advisor for open science to the director-general of research and innovation at the Ministry for Research (France).

More Posts - Website

Follow Me:
Twitter

Marin Dacos

Marin Dacos (CNRS) is the founder and director of OpenEdition. He has been awarded two Google Digital Humanities Research Awards (2010 and 2011) and the CNRS Innovation Medal (2016). He is scientific advisor for open science to the director-general of research and innovation at the Ministry for Research (France).

3 réponses

  1. Peter Suber dit :

    No. Serials include many kinds of periodicals, not just peer-reviewed journals. I tried to get data from several different libraries on the number of peer-reviewed journals to which they subscribed, as opposed to the number of serials. But I couldn’t. The libraries I contacted didn’t break down their serials that way. If there’s good information on these breakdowns today, I’d appreciate a pointer and would be happy to add a summary to my updates and supplements [ http://bit.ly/oa-book ].

  2. Peter Suber dit :

    See this update to p. 78 of my book [ http://bit.ly/oa-book#p78.2 ] (click through for a version with live links):

    « It’s very difficult to get an accurate number for the total number of peer-reviewed journals in all fields and languages. 25k is still the most commonly used industry estimate. However, even limiting the count to titles indexed in Ulrich’s, the number is now closer to 28k. See the discussion thread on this question at LibLicense in August 2012. (If we use 28k as the total number of peer-reviewed journals, then in July 2012, 28% were OA.) Moreover, there are reasons to think the Ulrich list is itself incomplete. See for example Jack Meadows, « The Growth of Journal Literature: A Historical perspective, » in Blaise Cronin and Helen Barsky Atkins (eds.), The Web of Science. A Festschrift in Honor of Eugene Garfield, ASIS&T Monograph Series, 2000, pp. 87-107. In 1987, Meadows estimated that there were 71k peer-reviewed journals worldwide. (Thanks to Jean-Claude Guédon for the reference to Meadows.) »

    • Marin Dacos dit :

      Dear Peter, thank you for the detailed feedback. If Harvard was subscribing to 98 900 serials in 2008, should’nt we consider that there are at least 98 900 journals?

Laisser un commentaire

Votre adresse de messagerie ne sera pas publiée. Les champs obligatoires sont indiqués avec *