Michael Weissman writes:

I would like to invite you to my upcoming talk titled Erroneous Statistics in Physical Review Physics Education Research taking place on 2023-02-28 (yyyy-mm-dd) at 16:00 UTC [that’s 10 am CST] as a part of the Speakers’ Corner seminar series of the Virtual Science Forum.
To see the talk abstract and register please go to VSF Speakers’ corner website or register directly using this link.

Here’s the abstract:

The American Physical Society publication Physical Review Physics Education Research focuses on questions that are essentially social science. Unfortunately, it routinely publishes and promotes papers that make egregious errors in statistical reasoning, particularly in causal inference. As a result policy recommendations based on PRPER papers can rest on causal conclusions that are unsupported by evidence and sometimes (e.g. on use of GREs) flatly contradicted by the evidence. I will describe just one typical erroneous paper to illustrate the technical issues, while providing references for critiques of many others. The PRPER editorial team has an express policy of resisting the type of error correction traditionally employed by Physical Review.