Last week, I was contacted by Elizabeth Gibney, who writes for the Times Higher, with a request for comment on the recent Science-Metrix report, and particularly their findings that papers that were OA had higher citation rates, but gold was not as high as green. Elizabeth has very kindly included some of my response in her article, but I wanted to share my full response below.
As the authors of the Science-Metrix report note, these new findings on gold citation rates are not standalone data; they require extremely careful interpretation and we should be wary. Firstly, it is important to point out the positive fact that this study is merely the latest in a long line to confirm that open access research is more widely used by researchers; "all the fields derive an OA citation advantage". Secondly, as the authors correctly note, it is perhaps not surprising that long-established journals that open up a green route are more frequently cited; they have a historical readership who trust them as a known quantity. In short: the comparison here is not entirely valid -- it is unfair to take a journal that has a decades-long history and compare its citation rate to a new gold-OA journal, particularly in the humanities where the time from submission to publication is often exponentially longer than in scientific disciplines and where readers often have "go to" journals and known routes to discovery.
That said, there are some additional important observations that should be made. The foremost of these is that new gold OA journals need to ensure that they are discoverable. Simply being online and available for free is not enough. Researchers have long-entrenched (and good) research-specific library discovery methodologies and resources (such as SFX) which do not always lead to gold-OA publications. While many of the underlying software platforms for gold-OA, such as PKP's commendable Open Journal Systems (OJS), have worked hard to ensure integration with these systems, it is not the case universally. It is also important for new gold-OA journals to gain inclusion within respected indexing services, including those run by scholarly societies.
Finally, I remain of the opinion that if we were to start from scratch today in a quest to meet our scholarly publication/communication needs, and were it not for the constraints of the history of publishing, gold OA (making our material freely available for dissemination over the internet with a form of peer review) would simply be called "publishing". That said, this is not the way things stand; history cannot be unwritten. Instead, authors are motivated by strongly conservative systems of accreditation and, in several fields, remain wary of gold OA for many of the same reasons as above: career progression is obtained by publishing in known, long-established journal brands that maintain their journal-level reputation through the replication of print-scarcity, even when they have a citation disadvantage, and this will take a long time to change. In short: researchers are motivated by reputational capital and many new gold-OA journals have not yet accumulated enough for researchers to take the risk (as well as the fact that much confusing and false rhetoric has circulated to equate gold with "author pays", which is not the case in the majority of gold journals listed in the Directory of Open Access Journals). Overall, however, researchers should be careful of shying from gold OA based solely on these findings and, with apologies to John Lennon, should give gold a chance.