STUDIES:
295/
A PNAS paper finding no racial bias confuses (i) probability of black person being shot by police with (ii) probability of shooting victim being black. “Johnson et al.’s analysis cannot recover these shooting rates because all observations in the data involve shootings."
Making Inferences About Racial Disparities in Police Violence
Proceedings of the National Academy of Sciences (PNAS)
Vol. 117 Issue 3, pp. 1261–1262,
2020
Comment on this study on its Twitter thread.