Library Out Loud

Since widespread Internet access became available about twenty years ago, researchers have seen it as a simple and fast means of disseminating their work within their own communities, and beyond that to other research communities and to the public at large. Universities (and particularly their libraries) got into the game very early on, but one barrier has always been that the most defining works of academic scholarship were not available for the sorts of free dissemination that the Internet represents. In particular, peer-reviewed journal articles were owned not by their producers (the authors) or their funders (the universities) but by the publishers to whom copyright was assigned as a condition of having the work appear in prestigious and career-enhancing journals. The  defining characteristic of these journals (apart from prestige and career enhancement) was their eye-watering cost, and the consequent need to protect their income streams by the erection of large firewalls – plenty of librarians have helped with this enterprise as a condition of providing their own scholars access to the publishers’ baskets of goodies. Major hand-wringing.

This inequity of access to the products of scholarship has been responded to by the growth of open access (sorry, Open Access) in various forms. Since the mid 90s open access publishing has gone through a variety of forms and over time has moved from the free-to-publish (no author charges) and free-to-read (that’s what open access means) model to the not-free-to-publish (author charges) model. (Nick Hopwood of University of Technology Sydney ran a really interesting workshop on academic publishing here at Massey three weeks ago and followed it up with a brilliant summary on his blog of current trends in academic publishing and he deals with Open Access from the point of view of a practising researcher.) As well as Open Access publishing , the institutional repository movement has negotiated with copyright owners (that’s a fancy way of saying publishers) to gain the rights for “unofficial” (i.e. not branded or paginated) versions of peer-reviewed articles to be placed in document repositories or on university websites. The third form of Open Access (and one that is often overlooked) is the longstanding practice of many journals (particularly those of scholarly societies) of making their articles free to all Internet users after the expiry of an embargo period – this is particularly common in the life sciences.

So anyway, what does this all mean for the average non-affiliated person wanting to read peer-reviewed journal articles? I asked myself this question last year and my answer can be found in this article Total Availability of Journal Articles to Internet Users published in the latest issue of Library Review. It’s really really interesting with lots of statistics and tables and a novel approach to bibliometric sampling. (Well, I thought so.) When I looked at the literature around OA publishing I found that a lot of it dealt with the amount of Open Access material that existed, but that was not very helpful in answering my very simple question – let’s say I want to read this particular article, what are my chances of finding it on the Internet? And the answer is – ONE IN THREE!!! (Okay, the actual answer is that if the article was published between 2011 and 2013 and you were looking for it in December 2013 then you had a 32% chance of freely accessing a copy of it. Precision.) Obviously it was important to avoid skew in arriving at this figure, as the possibility exists – and was confirmed – that OA practices would vary greatly both by publisher and discipline, so in order to gather a “neutral” sample I used the surname Robinson (first author) as my sampling criterion. To get a cross-section of the quality peer-reviewed literature I used the Scopus database and pulled down a total of 1,463 articles (that’s an awful lot of Robinsons) which I then looked up on Google Scholar to determine if they were readily available.

Here’s a summary of what I found

• Of the total sample, 15% were freely available from their publishers, 13% from repositories and 5% on web pages and social networking sites like Researchgate.
• Articles from the big commercial publishers were much less likely to be available than those from scholarly societies, both in terms of being directly accessible from the publisher and through repositories. The effect of (expiring) embargoes may be an overlooked factor in OA, whereas the outcome of hybrid gold access publishing seems to be negligible so far.
• Articles from the life sciences were much more likely to be available than those from the social sciences. For example 55% of 2011 life sciences articles were available compared to 24% from the social sciences. Health sciences and physical sciences fell between these two extremes.

Obviously a relatively small study of this kind only scratches the surface, but with the amount of interest there is in OA publishing it seemed like a good idea to ask the “where are we at?” question. The 13% figure I found on material available from repositories is very similar to the 12% that Björk, Laakso and Welling  reported last year, so I don’t think I’m too far from the mark. Anyway you should read it, it’s really great. Here’s the link again.

Bruce White