The concurrent external validity of the diagnosis of cocaine abuse, based on the DSM criteria, is shown to be established to an unexpectedly high degree, based on a method of determining the association between a summary score derived by quantifying the DSM criteria and another summary score derived from weighting several measures of frequency of use of cocaine and a measure of its mode of use. The degree of validity was cross-validated by performing the same analysis on two study samples: one of inpatients (N = 179) and one of urban-community African-Americans (N = 204).