Christmas came early this year, with the Stanford Education Data Archive releasing a new version of their nationally-normed data from (more-or-less) every district in the country’s 3rd through 8th grade test results by race. Unlike the last version (which only showed gaps), these show mean performance within race by district and county in grade level equivalents. So it gives us a clue which places seem to be especially good at attracting or fostering good students within different racial groups, and the extent to which “good schools” are good across the board or for particular groups.
The map for white students is probably no big surprise, with the Northeast/Acela corridor states doing best, along with places with rising incomes like North Carolina, Colorado, and Minnesota, and Appalachia and the Deep South doing worst. (The units here are in grade levels ahead or behind national averages.)
The map for black students is similar, although here Michigan and Illinois stand out for doing especially poorly, and a few states with small black populations (Wyoming, Montana, and Hawaii) probably having a heavily selected group.
Hispanic achievement has poor performance for New York and most of the West, and better performance in Florida:
And Asian students have much stronger performance in the East Coast than on the West (and very little county-level data).
California and New York does surprisingly poorly across the board, while Florida and Texas look a bit better. The reliably high performers seem to be New Jersey, Virginia, Connecticut, Massachusetts, Maryland, and Colorado.
Across grades and years, the results are also pretty expected and slightly depressing. Racial gaps get larger from 3rd through 8th grades, with Asian students rising slightly relative to national grade-level norms, and white, Hispanic, and black students all dropping, black students the most. Black students go from about a grade level behind in 3rd grade to about 1.5 grade levels behind in 8th grade.
Across years, there’s even less movement; performance by each group is about the same as it was in 2009, with perhaps very slightly larger gaps more recently.
Again, all the hubbub of recent education reform policy didn’t really do much of anything, for better or for worse. Since many states switched over to Common Core tests in 2014, it may be hard to compare the last two years of data, but it certainly doesn’t look like a big switch.