A recent study by the Publishing Research Consortium (PRC) postulates that the most prominent (and thus most citable) authors are more likely to make their articles available in an OA model, and that they are more likely to do so with their most important (and thus most citable) articles. (Executive Summary, 5B).
The PRC also postulates that there is an Early View effect with open access articles, which relates only to articles posted before final publication, and suggests that the period between the early posting of an article (either pre-print or post-print) and the appearance of the cognate published journal article allows for earlier accrual of citations.
If these postulations are correct, this is good news for open access, readers, and authors!
Readers will have yet another way to sort through all of the research literature that is being produced nowadays. If one does not have time to read everything, then start with open access resources. Not only are they easier to access, they may be higher quality and more important too.
For authors, what this would mean is that taking advantage of the OA impact advantage means getting your article OA as soon as possible, particularly as a preprint. If correct, this would be strong evidence that embargoes on open access are a serious disadvantage to researchers.
Whether or to what extent these hypotheses are correct is a matter I will leave for the experts in this area. From my perspective, I would not underestimate the impact of open access per se, even aside from these other factors which could be partially responsible for the OA impact advantage. After all, in order to read an article, build on the work and thus cite it, one does need to be able to read the article in the first place. OA expands readership; hence, it seems illogical in the extreme to think there would not be a correlation between OA per se and impact advantage.
To review the PRC literature review, go to www.publishingresearch.net