Stanford University sociologist Molly M. King and colleagues set out to determine the frequency of self-citation in scholarly publication and whether the practice varies by gender (Nature 2016, arXiv: 1607.00376). To do so they analyzed a dataset covering 1.5 million research papers across a variety of fields from 1779 until 2011.

The researchers found that about one in 10 references is a self-citation. And who are the worst offenders by discipline, you wonder? It turns out that, of the areas included in the study, molecular biology is the major academic field showing the highest self-citation rate at 17.6%, versus classical studies at 5.6% at the other side of the spectrum.

It’s probably also not unexpected that men self-cite 56% more than women do, even after accounting for gender differences in authorship. What I found surprising is that in the past two decades alone, men self-cite 70% more than women. And women are 10% more likely to not cite their own past work at all.

It looks like self-promotion is more common for men, and they are getting better at it. The authors speculate that this tendency could be due to a variety of factors including men self-assessing their own abilities more positively than women, men perceiving fewer “social penalties” associated with self-promotion, men being more likely to specialize in niche academic subfields for which the universe of relevant papers to cite is smaller, and men publishing more papers over their careers and thus having more work to cite.

Whatever the reasons for this gap in self-citation, what is clear is that greater self-citation increases citation count, thereby increasing the perceived quality of a paper. Citations and the ability to publish in a high-impact-factor journal are—rightly or wrongly—some of the productivity measures that are regularly used to assess the work of research scholars. Of course, it would be naïve to think that gender disparity in self-citations has no serious implications; in fact, it is likely to have a significant impact for women in academia in terms not only of visibility, but also of pay and career progression.

As the authors state, self-citation should not be demonized. After all, it is one of the “few direct ways an academic can increase his or her citation count.” Like everything, it should be done in moderation and only when appropriate. But there is an opportunity for female academics to make some adjustments in this area. If nothing else, it is useful to have a benchmark of what is acceptable. More broadly, these results should encourage female scholars to look at self-promotion in a different light. Men are more proactive with it and it pays off. This, in combination with a greater ability to articulate their aspirations and ask for raises and promotions means they tend to advance faster and higher up the academic chain. Of course, there is more to this than meets the eye. But it is clear that female academics must give their achievements greater visibility if they seek to advance (see, for example, page 35). Well, you can start now: When writing your next paper, consider your references and “set your cites high.”