This study examined the psychosocial predictors of nicotine dependence, as defined by a variant of the criteria employed in the DSM-IV—specifically that of the University of Michigan Composite International Diagnostic Interview (UM-CIDI)—and the Fagerström Test for Nicotine Dependence (FTND). The study was conducted with a community sample of African American and Puerto Rican young adults (N = 475; mean age = 26). Predictor variables included physiologically based psychosocial (i.e., depressive symptoms and family problems with smoking) as well as social–behavioral psychosocial (i.e., rebelliousness and partner's problems with smoking) predictors of nicotine dependence. Using multiple regression analyses, UM-CIDI-defined dependence was predicted by each of the four psychosocial variables, while FTND-defined dependence was predicted only by the social–behavioral variables. These findings bear out the disparate dimensions of nicotine dependence each measure taps. Research and clinical implications of the findings are discussed, and the study's limitations are noted.