I admittedly have not read the entire thread and have not dug through the article in detail, but I’m suspicious of the methodology and categorization being used here. First of all, which “reading and math test score data” did they analyze? They don’t specify. PARCC? If so, I call a huge BS. That test is no where near ready for use by anyone for any level of measuring anything reliable … other than specific test prep. In our school district, (Howard County, MD), supposedly white students are 2.5 years ahead of the average; Hispanic are 0.1 year ahead, and black students are -0.1. Our GT program puts kids supposedly 2 years ahead. For the average white student to be 2.5 years ahead would mean that almost all of our white kids are in the GT program (which is not true), or that our GT program really puts kids more than 2 years ahead of the national average, and our “normal” track is also ahead of average. And if that is true, then I weep for the future of our nation, because our national average must truly suck. I also find it suspicious that data for Asian Americans is “not available” … um … why? I am quite sure that if you plotted Asian Americans on the graph for Ho Co MD, it would be a smaller bubble at probably something like +3 or more, probably around the same or maybe slightly lower socioeconomic level than whites. Our Asian community is very tightly knit and has a very high rate of outside tutoring (e.g., Kumon) from early on in elementary school, regardless of socioeconomics.