Even papers that indicated authors had “contributed equally,” male-female listing was statistically more frequent than female-male listing.
Image: Chenspec/Wikimedia Commons
Sexism in science is nowhere near a new phenomenon. Biology nerds will never forget Rosalind Franklin’s x-ray diffraction of DNA and her largely unacknowledged contributions to Watson and Crick’s DNA double helix model in 1953. Today, women are gaining more traction in many scientific fields, but our deeply-ingrained sexism remains clear in many facets of academic research.
One of the eternal struggles of women in science is getting credit for the research they have contributed to. Many institutions measure success through the number of “first author” publications an application has, because the first listed author on a paper is assumed to have contributed the most. First authorship has huge implications for career advancement, which is especially important for early-career female scientists if they want to achieve senior positions.
Women are severely underrepresented as first authors, though. One study of high-impact medical journals from 1994 to 2014 published in the British Medical Journal in 2016 found that only 34 percent of the research papers had female first authors. And, though female representation overall has increased greatly over the past 20 years, it has plateaued and even declined in some of these journals. This type of sexism in research publications has been documented in astronomy, pediatrics, gynecologic oncology, and radiation oncology, among others.
A new paper by Nichole Broderick and Arturo Casadevall, professors at the University of Connecticut and the Johns Hopkins Bloomberg School of Public Health, respectively, shows that sexism in research publication is rife in the field of biomedical research too. The paper is currently in pre-print and hasn’t yet been peer reviewed. It was published online on December 31.
Broderick and Casadevall decided to investigate gender gaps in co-authored papers, specifically, in which authors indicated that they had “contributed equally.” Broderick told me that there has been a recent increase in co-authored articles because of more interdisciplinary and collaborative research that requires equal input from multiple researchers. Co-authorship becomes a problem when the first author listed receives more credit even when the authors indicate that they contributed equally.
Broderick and Casadevall found that co-authored papers listed men as first authors more frequently than women, despite equal contributions, and therefore demonstrated a subtle bias in the publication process. However, this bias seems to be decreasing, and women are becoming more represented as first authors, though whether the male-female and female-male authored papers have become equal yet is unclear.
To investigate whether there was a gender bias in the ordering of author names in their field, Broderick and Casadevall mainly found co-authored papers through internet searches for biomedical literature, and then determined gender by finding images of the authors.
They eventually came up with 2,897 total articles, and found that male-female listing was statistically more frequent than female-male listing, with a ratio of 1.3:1, with male authors listed first in 56.5 percent of the publications. They found a similar suggestive (but not quite statistically significant) effect for publications with three or more authors sharing credit. Broderick and Casadevall found, however, that this effect has decreased over time when they compared the data from 1996 to 2006 to the data from 2007 to 2017. The 2007 to 2017 data set did not demonstrate a statistically significant bias, while the earlier decade did, demonstrating that the recognition for males versus females as first authors has reached parity, or at least come close to it.
“To me, that’s very optimistic, because we can interpret it as: this was a problem, but it’s maybe getting better,” Broderick said.
The authorship order of many articles was not even alphabetical, so it was often unclear why the first author claimed that spot. Only one paper out of the close to 3,000 Broderick and Casadevall studied indicated how they chose to the authorship order.
Broderick and Casadevall made their paper a model for what they hope will be the future of authoring, listing themselves in alphabetical order and in order of increasing seniority, and clarifying this within the paper.
“My hope is that at the very least this will stimulate discussions. There are people today trying to write papers, to figure out how to put the authorship in, and they will look at this paper and say ‘hmm, maybe we should put an asterisk how we did this.’” Casadevall said. “That can begin to change how we approach what is going to be a really important issue.”