When you're working in a niche area of research or academia, it's understandable that you might on occasion have to cite your own published papers as referential authority for a specific point.
But researchers in the US looked through more than 200 years' worth of published studies and found an interesting trend: male researchers self-cite their own papers more than their female counterparts.
After trawling through records of some 1.5 million studies published between 1779 and 2011 on academic database JSTOR, a team led by sociologist Molly King from Stanford University found that men cited themselves on average 56 percent more than female researchers.
Most remarkably, despite ongoing efforts to increase the representation of women in science and academia generally, this pattern has actually remained fairly stable over the last half century, and even gotten more pronounced in the past 20 years, with male researchers citing themselves 70 percent more than women in the last two decades.
You might be able to write off this trend – amusingly dubbed "manciting" by Stephen Buranyi at Motherboard – as a harmless behavioural difference between male and female researchers, but the sad thing is, it's not really harmless.
"I think there's this idea that all we need to do is get more women into science and that will solve all the problems," King told Motherboard.
"But if women aren't citing themselves, there's a further visibility gap that comes from that, which makes it harder for women to get tenure. So it's making it even harder for women to narrow the gender gap in academia."
In other words, this self-citation quirk – where men refer to their own research more than women do – is just another thing that favours male academics in terms of getting them noticed in their careers.
Meanwhile, female researchers – who are 10 percent more likely than men to not cite their own previous work at all – aren't getting noticed in the same way, and that's something that would hurt their professional prospects.
"Every citation a paper receives will attract additional future attention from other scholars," King explains to Dalmeet Singh Chawla at Nature.
What's also interesting is seeing just how big a chunk of the citations in published papers are actually self-citations. According to the researchers, it amounts to about 10 percent overall.
While that sounds high, the team say that most of these are appropriate self-references and not attempts to game the system.
It's worth pointing out that the study hasn't been peer-reviewed as yet, so we need to consider that the research might still be amended by its authors. The paper was originally presented at a meeting of the American Sociological Association last year, and was uploaded to arXiv.org in June, so the researchers could take feedback from the sociology community.
One limitation with the study could be the manner in which the researchers gauged gender from JSTOR records. The team discarded gender-neutral names (understandably), but also dropped authors listed with only an initial for their first name.
This method might have disproportionately skewed the results, especially if some were female authors who were trying to keep their gender hidden by obscuring their first name.
"Only 56.4 percent of all the authors in their database were able to be assigned a gender," data scientist Adrian Letchford from the University of Warwick in the UK, who wasn't involved with the study, told Nature. "That's a very big portion left out, especially considering that women may be actively hiding their gender."
The researchers also acknowledge that male academics tend to publish more – which gives them more of their own papers to cite.
They're also generally more senior in academia and therefore more likely to be specialised, which could encourage them to refer to their own work in narrow fields.
But despite this, it's important that we look into disparities like this, to help inform future policy in science and academia, and help eradicate inequality – even if it's in something as seemingly obscure as the rate at which researchers drop their own names in new research papers.
"What's most surprising is there's no decrease over time," King told Motherboard. "We were expecting to find that, as women have entered science and gotten into more senior positions, we would see that ratio get smaller, and we're still not entirely clear on why that isn't the case. I look forward to trying to figure it out."