For what it's worth, this data is from 1997. (E.g., see table 3.4 from the book being discussed, No longer separate, not yet equal: race and class in elite coll

For what it's worth, this data is from 1997. (E.g., see table 3.4 from the book being discussed, No longer separate, not yet equal: race and class in elite college admission, by Espenshade et al.) Also - and this is not exactly a problem, just a methodological feature to be aware of - it appears that they produced their models by aggregating the data together across schools.