Update! 📉📈 Adding an omitted variable to the model in our PNAS paper on physician-patient racial concordance changes the result. Grateful to Borjas & VerBruggen for continuing to advance the science & to Ted Joyce for his commentary in Related Materials. www.pnas.org/doi/10.1073/...
The racial gap in infant mortality is a pressing public-health concern, and [B. N. Greenwood et al., Proc. Natl. Acad. Sci. U.S.A. 117, 21194–21200...
Doesn't this just identify the causal variable though? Or is there some other process mediating birth weight that I don't see?
From one scientist to another, your graciousness and integrity in promoting this reanalysis of your study is truly inspiring.
I encourage anyone interested to read the papers & the Joyce comment. They're concise but contain a lot of analysis pushing towards and away from various possibilities and answers to many questions.
Is it properly viewed as a confounding variable?
Appreciate the gracious behavior all around.
Look forward to reading. Given that low birthweight is significantly associated with race and plausibly post-treatment, does the paper address post-treatment bias as a concern?