Chart Of The Week

A reader writes:

I’m surprised that you repeatedly posted that chart about the Stand Your Ground results compared to white-on-white incidents without attribution. It’s a great chart, and the difference so stark, that it begs for attribution.

We subsequently updated the posts with a link to the study – by John Roman of the Urban Institute’s Justice Policy Center, via PBS’s Frontline, which followed up with Roman to create the chart. Another reader picked up on Roman’s study:

I was amazed by the graph you’ve posted several times this week regarding percentages 11of justifiable homicides between the races.  It was troubling enough that I decided to track down the source of the study.  In doing so, I found out an even more amazing statistic.  The study looked at all 73,000 homicides between 2005-2009.  It then separated all of the homicides where one stranger killed another stranger, similar to what occurred in the Trayvon case.  Finally, the study separated those instances by race.  Of the 73,000 homicides in that time period, only 23 were one white person killing one black person.  23!  From the media coverage this weekend, I thought the number was probably in the thousands.  The small sample size makes the significant portion of the graphic you posted basically worthless.

I don’t think it makes the chart worthless. But it’s an important piece of perspective. Update from a reader:

The reader is fundamentally confused on the difference between a random sample survey and a census or direct reporting of observed data.  For a survey, you need to reach a certain number of responses/cases randomly selected from the total universe of potential respondents or cases to achieve statistical validity.  But if you’re reporting the actual, total data for a universe of observed people or phenomena, there’s no “sample size” involved – you’re reporting the facts.

Another critique:

Income is absent from this chart. Did it occur to you that perhaps the white people were wealthier and had better lawyers? Maybe the amount paid for lawyers correlates to the efficacy of those lawyers? (It stands to reason that lawyers that can charge more get their way more often; why would anyone pay more for them if that weren’t true?) In fact, nationally, white men earn 150% of what black men earn.

Another builds on that point:

Look at OJ – a rich black guy was able to kill two white people and get away with it, even with what looked like a very good case.  How? MONEY – he had a lot of it and bought the best lawyers he could to buy his freedom.  I suspect many of the white dudes in the chart that beat their case because of their ability to put up a vigorous defense, or at least threaten to do so. So the chart does show that SYG makes it easier to beat a murder charge, but it doesn’t show that whites (or Hispanics) are getting away with murder with a sly wink of a racist society.  I’d also like to see more detail from the chart as well, such as the location and situation of the relative incidents.  It’s possible that there are a lot of other factors involved.