From the paper, I've never been sure how this applies (my statistics skills are something I'm working on):

"In 2010, the rate of firearm homicide for blacks was 14.6 per 100,000, compared to 1.9 for whites, ..."

Is this for the same population sample? So we're comparing the whole of 325m Americans when computing these numbers in both cases, yes? Or am I normalizing to the black population and the white population separately? My gut would tell me the former, but I confess to being a rank amateur at anything but pure hard numbers.

Statistics == opportunity for con-job. :-)