One of the main problems in academia is to evaluate a researcher or a particular research project. The best method that scientists have come up with so far is to assess the impact factor of journals in which author publishes and/or to measure the number of citations i.e., how many times an academic paper was mentioned in other research. It naturally evokes several questions and potential issues; reciprocal citations, purely personal and/or professional reasons (beyond the quality of the research) for publishing in a journal, paid or free access to publication for readers and so on. Note that free access to a publication is ensured by the authors themselves who pay a fee to the publishing company.
Gaule and Maystre (2011) examined whether free access to a publication increases its chances of being cited more often. The data and previous studies show that open access articles achieve a higher number of citations, however, Gaule and Maystre set up a model which explains why, when choosing between open access and restricted access articles, researchers may tend to prefer an open access article if the research project is of higher quality. Their results suggest that a higher number of citations of open access articles might not be a consequence of a diffusion effect, but rather a self-selection effect.
In particular, the authors studied more than 4000 articles issued in Proceedings of the National Academy of Sciences (PNAS), where authors can choose whether to publish their article as open access or restricted access (there are some exceptions for developing countries etc.). To distinguish between a diffusion effect and a self-selection effect, the authors also controlled for the quality of the research by creating two proxy variables – the quality of authors and the quality of the article. The figure clearly shows that the number of citations achieved in the 2 years following the publication of the article is higher for open access articles, however, as the study concludes, this fact is most likely driven by the self-selection effect i.e., open access articles tend to be of higher quality. The authors argue that the diffusion effect of open access is, if any, relatively small.
Reference: Gaule, P., & Maystre, N. (2011). Getting cited: does open access help? Research Policy, 40(10), 1332-1338. Available here.