“…We are also the first to calculate desegregation trends using disaggregated data on a representative sample of Southern school districts. Previous research has relied on published state aggregates of the data used here (e.g., Rosenberg, 1991;Orfield, 2000) or constructed regional segregation statistics from a small individual-level survey (e.g., Boozer, Krueger, and Wolkon, 1992;Ashenfelter, Collins, and Yoon, 2006). 32 Using district-level data, we are able to examine formally how other district characteristics, in addition to enrollment, were related to different desegregation trends within the South.…”