[tomrod]

This comment is not to agree or disagree with your comment, but to informally provide an interesting finding.

Economies are more or less urban regions surrounded by rural areas with low population density. So, Nashville, DFW, Denver, etc. are regional economies.

I have found it interesting in recent research to find that the level of disparity in health inequity (called Social Determinants of Health, one of which is the 'Area Deprivation Index' managed by UW-Madison) and race/ethnicity is somewhat dependent on which regional economy you review. I can't speak for all regional economies, but the ten I looked into show interesting variation on the strength of economic disparity and the composition of race and ethnic groups.

[Loquebantur]

A correlation between ethnic groups and economic disparity is not indicative of causation, much less of what the cause might be?

People having problems with even simple statistics definitely is a cause for racism on the other hand.

I find economists often to be surprisingly bad at math.

[tomrod]

> A correlation between ethnic groups and economic disparity is not indicative of causation, much less of what the cause might be?

Correct. We typically understand the actual cause in the US -- historical policies promoting detrimental economic outcomes to targeted ethnic groups. I'm not sure why you interpreted my positive finding of relationship into a normative declaration of causation.

> I find economists often to be surprisingly bad at math.

Which ones?

[snapplebobapple]

So are you then arguing that, since the average is no difference in discrimination except small wage discrimination in favor of men and hiring in favor of women , the reality is larger discrimination against men and women that averages out mostly?

[tomrod]

Out of curiosity, what part of my prior comment of findings lead to your conclusion that I made the argument you stated?

[snapplebobapple]

The article concludes there is no difference except the ones outlined, you pointed out that sub economies have differing results. To have no difference in population but significant difference in sub population thw variance needs to be opposite enough to cancel out significance just by how the math works.