Featured Thoughts, News and Microarticles from the Scientific Community.

The open access "movement" has existed for decades, but with the coming of the digital age in the 90's it truly bloomed as the distribution cost of electronic media dropped to essentially nothing. Some groups completely embraced the paradigm-shifting potential of the internet; open-source software and piracy being the main examples which come to mind. And while individual scientific disciplines took advantage of the new, more efficient way of searching for and distributing journal articles, the science community as a whole has still not tapped in to the full potential offered by the near limitlessly connected world in which we now live.

I am personally a big fan of the concept of a truly open scientific community that shares findings and raw data via the internet in a melting-pot of consilience. I'm aware that such a vision is extremely naive, not least because of the strangle-hold the big journals currently hold on what is perceived as "proper publishing" and the financial motivators so often required to secure funding for good science. None the less, websites such as ReasearchGATE demonstrate the possibilities which can be unlocked by simply linking together people and ideas. And open access journals do exist and can be made to work. However what impact does making a research paper open access have?

A 2008 paper, "Open access publishing, article downloads, and citations: randomised controlled trial" (Davis et al) took a suitably scientific approach to this question by randomly assigning 247 articles published in 11 journals of the American Physiological Society to open access status between January and April 2007. The control group of 1372 articles was articles available via subscription, the traditional access model for the American Physiological Society’s journals for the first year of an articles life. The researchers looked at a number of related variables- the number of abstract and full article downloads, and the number of times a given article was cited in the year following its publication.

Unsurprisingly removing subscription restriction lead to a significant increase in full text downloads (89% higher for open access articles compared to subscription articles). Interestingly, however, despite the increased number of downloads this did not translate into an increased level of citation. While 59% of the open access articles were cited a year later, 63% of the subscription articles were cited. The authors concluded that open access to articles in fact reduced the chance of the article being cited (although not significantly) compared to subscription articles, and that "open access increases the readership of articles but has no effect on the number of citations in the first year after publication."

The main issues with these findings, however, is that the researchers made the articles open access at random, when in fact there is nothing random about the articles made available in the real world. Indeed a number of studies have shown that open access articles are consistently cited more often than subscription ones (an interesting look at the issue here), with the suggestion being that it is this very non-random nature of article availability which drives the perceived benefit of making a paper open access. In other words, there is self-selection bias in open access papers, in that only the "best" papers are made available this way, and hence they inevitably will be cited more. 

However a paper from the start of this year, "Self-Selected or Mandated, Open Access Increases Citation Impact for Higher Quality Research" (Gargouri et al, 2010) has questioned this very assumption, by comparing the impact of self-selective self-archiving (i.e. researchers choosing to make their articles freely available), with that of mandated open access (i.e. research institutions forcing their staff to make all research freely available as it is produced) for 27,197 articles published between 2002-2006 in 1,984 journals. As expected, when controlling for a number of other variables linked to citation, open access articles were cited significantly more than non-open access articles.

Most interesting however was the finding that it didn't matter whether the article had been chosen to be make open access, or was merely made available due to institution policy, the open access citation advantage was still found. The authors concluded that it is "highly unlikely that the OA advantage is either entirely or mostly the result of an author bias toward selectively self-archiving higher quality – hence higher citeability – articles." Higher quality articles were cited more, but this was independent of whether the author had chosen to make the article available themselves.

Making your research open access gets it cited more. Is there anything more a scientist could ask for? The dissemination of knowledge is one of the core tenets of science and should be, in my opinion, one of the core principles of any society. As an added bonus, with original research freely available online anyone making dodgy claims, such as poor media reporting, could be fact checked far more easily by members of the public. It's win-win in my eyes.