Links from emotional distress to adolescent drug use: a path model.

Administered anonymous surveys asking about drug use, emotional distress, and peer drug associations to 11th and 12th grade high school students (N = 563). Emotional distress variables accounted for only 4.8% of the variance in drug use. The addition of peer drug associations as a predictor variable increased the variance accounted for to 43.4%. A path model of adolescent drug use based on peer cluster theory was tested using LISREL, and this provided a good fit with the data. As predicted, peer drug associations dominated the prediction of drug use and mediated the effect of emotional distress on drug use, with the exception of a small residual path directly from anger to drug use. The hypothesis that young people take drugs to alleviate emotional distress does not hold up well; emotional distress variables, with the exception of anger, produced only very small and indirect links to drug use.