[tzs]

That's a pretty big difference. However, before ascribing it all or mostly to race, it would really be necessary to look at the individual cases.

In general, murders where it is a lower class person murdering an upper class person in furtherance of a robbery or some such draw more vigorous prosecution and are more likely to get a death penalty than murders where it is some lower class person killing some other lower class person.

Since blacks have disproportionately high representation in the lower classes, and whites disproportionately high representation in the upper classes, black murderer/white victim is disproportionately represented among the death penalty attracting lower class on upper class murders.

[erikpukinskis]

I appreciate that you are being careful about the possibility of a latent variable here. I don't have any data about poor-on-rich/rich-on-poor accusations and conviction rates, so I can't speak directly to that.

However, the general conviction rate for "high class" black men is higher than the rate for "low class" white men: http://www.pewresearch.org/fact-tank/2014/07/18/chart-of-the... I can't see any reason why I shouldn't assume the same applies to interracial murders as well.

I do appreciate your point, and perhaps it is time for a civil rights movement among the poor, or among the white poor. I would support that. We should at least be getting the data.

But since we have this data about black people specifically, it seems fine to try to solve that problem from a race-based perspective. As other data rolls in that shows there are similar problems for the white poor, then I think we should try to solve that problem too.

Saying we shouldn't try to fix this problem because new data might come along that shows it part of a bigger problem is, to me, "great is the enemy of good" thinking.