Although the underrepresentation of women and other marginalized groups in physics is well documented, efforts to address these disparities have always faced a certain amount of opposition. In the past few months, this opposition has turned into outright hostility from some in the USA and elsewhere.

Credit: R.Tsubin/Moment/getty

At Nature Physics, we continue to welcome and encourage not only submissions from a diverse group of authors, but also contributions that assess manifestations of inequity in physics and propose strategies to address them. Anyone who wishes to work in physics should be able to do so without facing barriers, and gender- and ethnically diverse teams have been proven to produce higher-impact science1,2.

This month we publish an analysis of one measure of the gender distribution of authorship in our journal3 and consider what we can do to tackle the underrepresentation of women authors in our pages. As Alannah Hallas points out in the Comment, for mid-career researchers in many branches of physics, being identified as the driving force behind a publication in a journal with a high reputation by being named as the last author is vital for taking the next steps in their career. Thus it is imperative that editors at those journals are fair and consistent when they decide which papers to publish.

“Overwhelmingly, the corresponding authors of manuscripts submitted to Nature Physics identify as men.”

The picture painted in the Comment is not encouraging: the number of papers where the last author is a woman is rather low — fluctuating around 8% — and does not appear to have increased in the last decade. This calls for introspection on our part.

Since January 2023, Nature Portfolio journals have been asking the author who completes the formal submission in our online system — we will call them the corresponding author in this context — to self-report demographic information about themselves, including their gender. Now, we can disclose those data from a period covering 2023 and part of 2024 (Table 1). It is important to note that these data are not directly comparable to the data collected by Hallas. Specifically, the corresponding author is often not the last author but can be a student or postdoc who carried out the work. This also does not capture instances where more than one author is designated as ‘corresponding’ on the final paper.

Table 1 Gender breakdown of the corresponding author of submissions to Nature Physics
Full size table

However, the corresponding author defined in this way is the person with whom the editor has the most direct contact and whose name is displayed quite prominently in the submission system, so it could be that the perceived gender of that author is a good proxy for analysing the potential for editorial bias.

The data are broken down into three stages. The first row of the table corresponds to the whole set of primary research papers that were submitted to Nature Physics in that time window. The proportion of respondents who did not wish to reveal their gender is below 10%, and so we are getting a reasonably clear view of the demographics of our corresponding authors. Also, the number of submissions we received from people with non-binary or gender-diverse identities was too small to make any strong inferences.

A glaring point is that the number of corresponding authors who identify themselves to us as women is low — just 8.1%. Overwhelmingly, the corresponding authors of manuscripts submitted to Nature Physics identify as men.

The second row in the table corresponds to manuscripts that were sent for peer review during that time period. If the gender breakdown of manuscripts sent for review is different from those submitted, that could be interpreted as evidence of editorial bias. However, the proportion of peer reviewed papers that have a corresponding author who identifies as a woman is very similar to the number for all submissions.

The third row corresponds to manuscripts accepted for publication during that time window. A change in the gender breakdown of the corresponding authors at this stage could suggest bias during the peer review process. But again, the percentages do not change much from those of the full cohort of submissions. In fact, the underlying data reveal that corresponding authors who identify as men have an effective acceptance rate of 8.2%, whereas for women it is 11.1%.

We need to be clear that these data are preliminary; in particular, the time period is too short to draw strong conclusions on the absence of bias in our editorial or peer review processes.

Looking to the future, we acknowledge the responsibility of the journal to provide an unbiased platform for highlighting impactful research, and we commit to monitoring this situation closely to ensure that if the proportion of manuscripts submitted by women (and people with non-binary or gender-diverse identities) increases, then the proportion of published papers from those cohorts increases accordingly. Therefore, we explicitly encourage submissions from all physicists, regardless of the demographic categories that they fall into.

We also invite anyone who feels that their authorship doesn’t belong in our pages because of their background to write to us or speak to us at conferences. We realize that the submission and publication process is not always clear, especially for first-time contributors, and we are available to answer questions and give advice. We are also open to suggestions on how we can adapt our practices to ensure that the authorship of papers in our pages reflects the wider physics community.

We aim to publish the most impactful physics research, and that can only happen with the wide range of perspectives that a diverse authorship brings.