If you’re like me, you may have had two academic articles with seemingly conflicting arguments run through your Facebook feed lately.  The first, an article by Cody T. Ross published via PLOS ONE uses a multi-level Bayesian analysis to conclude that there exists

evidence of a significant bias in the killing of unarmed black Americans relative to unarmed white Americans, in that the probability of being {black, unarmed, and shot by police} is about 3.49 times the probability of being {white, unarmed, and shot by police} on average

The other, written by economist Roland G. Fryer and covered extensively in the New York Times Upshot column, concludes that in the case of “the most extreme use of force – officer-involved shootings – we find no racial differences in either the raw data or when contextual factors are taken into account.”

How can such diametrically opposed claims be made simultaneously in reputable scientific journals?  While much of these claims seems to rest in a domain of advanced statistics with which anthropologists typically feel less confident, the key to understanding their different claims actually might depend on a more “ethnographic” sense of the data sets they build upon.

Sociologist Michelle S. Phelps has written an excellent breakdown of the data sets used by each, as well as the implicit biases baked into each.  I encourage you to read her entire post, but here’s one highlight:

Should we conclude then that there is “no bias” in police shootings? I don’t think so—for a couple of reasons. First, there is extensive evidence (including in the datasets Fryer considers) of large racial disparities in who gets stopped by police, even controlling for differences in crime rates (perhaps especially under policies like New York City’s “Stop-and-Frisk”). Because of this, the “hit rate”—or the percent of times a stop ends with a confirmation of wrong-doing—is often higher for whites than blacks. Even if police pulled the trigger without “bias,” this disparity in stops would produce vastly unequal death rates.

While Fryer’s analysis finds a relatively equal rate of violent police outcomes across race it doesn’t take into account that African-Americans are stopped and interrogated much more frequently for much more minor offenses, producing a much lower “hit rate” of intervening in actual criminal offenses.  In other words, his discovery of “no racial difference” in police encounters is equivalent to claiming that there is no racial bias because police shoot African-American traffic code violators at the same rate as they shoot white violent offenders.  Indeed.