Also, the high-social-mobility list of schools is dominated by schools in high CoL areas, particularly NYC. Graduates from NYC schools (including graduates wit

Also, the high-social-mobility list of schools is dominated by schools in high CoL areas, particularly NYC. Graduates from NYC schools (including graduates with non-wealthy backgrounds) disproportionately stay in NYC and make an NYC salary, which looks upper-middle-class by national standards but doesn't go very far when your rent is three grand. Therefore the claimed effect is at least somewhat spurious, being driven purely by geographic CoL variation. A better analysis would compensate for this, perhaps by normalizing by local CoL's.