Activists want to retract AlShebli's paper finding a female protégé's research had less impact under female vs male mentors, citing methodological flaws.
Where are the calls to retract her paper using the same methods finding that ethnic diversity improved research impact? 🤔
Where are the calls to retract her paper using the same methods finding that ethnic diversity improved research impact? 🤔
This depends on where the claims of methodological errors arise. Perhaps they approved of the way ethnicity relationships were defined, but not how mentorship was defined in the other paper. But the authors properly qualified their definitions on the mentorship paper.
Though this article suggests people are claiming to have problems with both how they defined mentorship and how they quantified scientific impact. Though they also think the paper is flawed because it didn't consider non-binary mentors... 🤦♂️ sciencemag.org/news/2020/11/a…
Just to add, I have no issue with post-publication reviews if issues surface later that the reviewers did not catch. And if something like coding errors occurred that nullify the results, retraction is on the table. But this isn't what we saw immediately following publication.
What we saw first was outrage and people calling for it to be retracted because the results weren't politically correct. Then some people raised methodological issues, which is fine, but don't warrant retraction per se.
Unless there's data fraud or some serious coding error or error in the data itself that nullifies the result, retraction is not justified. Do post-publication review, write a letter to the editor with your criticisms, reanalyze the same data a better way and publish the results.
This is how science is supposed to work. Make it a transparent debate, don't just try to make a paper you don't like disappear.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
