Thanks for the interesting link, I'll read Fryer's studies. One thing suggested by the last chart in that story is that use of force stats depends strongly on

Thanks for the interesting link, I'll read Fryer's studies. One thing suggested by the last chart in that story is that use of force stats depends strongly on whether the data sets come from police sources or civilian sources. I guess that's not much of a surprise, but it's another confounding variable to partial out.