I’m seeing some push back today that science is science and we shouldn’t attack a paper that presents valid data. Here’s the thing- science is typically (not always) based on hypotheses (1/n)

Which means you need to collect and analyze your data in a way that allows you to ask: is working with a woman good or bad for your career development? Or maybe: does the historical under-citing and under-valuing of Sciecne done by women have knock on effects for their trainees?

Once you have the question, you can test it.... but even then you have a role to play as an author. How do o present the data in a fair and unbiased way? Way story do I tell when presenting my results in order to frame them appropriately?

And that is where it is CRITICAL to reflect on the limitations of the data, alternative explanations, and overall utility of your results. This is where reviewers can help IF you listen to them and take the process seriously.

And in the end, your data are facts but your interpretation of those data is in your control, and if your interpretation is likely to do harm, there is NO reason not to stop, rethink, and change your messaging (or not publish!)