No retreat laws and racial disparities
The Tampa Bay Tribune ran an article claiming, from a review of 200 court cases, that defendants claiming self defense in Florida were more likely to be acquitted if the person shot was black. I posted on this a while back, looking over their raw data, and noting that most defensive shootings were black on black or white on white, and that both defenders of both races had very nearly the same chance of being acquitted (and hispanic defenders had a much higher acquittal rate than either).
Howard Nemerov, who has far more statistical skill than I do, has done a detailed analysis and gives a solid refutation to claims of racial disparity. He's responding to claims by John Roman or the Urban Institute, who used nationwide figures from 2005-2009. Howard first notes the peculiar choice of years ... why not go for 2000-2010? Why not compare figures for justifiable homicide from States with "no retreat" to States without it? Why not compare figures to law enforcement justifiable homicides, to see if justifiable homicide rules changed generally? He concludes that States which enacted no retreat generally saw an increase in the rate at which black defendants ruled to have been involved in justifiable homicide.