However, I did say "last few decades" and not "last decade" as you claim. You even reposted my comment in italics.
So, I'm not sure if you're trying to catch me with wrong statistics or what. But I do not believe your data contradicted what I said. I mean, if it was 301 instead of 289, would you say the term "several hundred" been misleading? I don't think so. And that wasn't my intent.
My intent, was to show, that proportionate to the people convicted of crimes, African Americans have been wrongfully convicted beyond their proportion in the judicial system. I'm not talking proportion to the population. We all agree, that African Americans have a higher incaceration rate. I'm talking about, of those who are or were incarcerated, those that were wrongly so, a higher percentage are African Americans. I still stand by my statement.
I don't think the numbers do reflect that. The percentage of convictions proved to be bad is so miniscule that the difference in which ethnic groups may be overrepresented is little more than a statistical error. It's not a large enough group in total to make any reasonable conclusions.
The "tiny number" of exonerations suggests that the "epidemic of bad convictions" that Scheck suggests is "fiction," said Marquis, chief prosecutor in Clatsop County, Ore. There were 1,051,000 felony convictions in state courts in 2002, up from 829,300 in 1990, according to the federal Bureau of Justice Statistics.