Young people participating in the study were detained at juvenile detention facilities in a mid-Atlantic State between 1992 and 1994. Most of the young people committed major property felonies, misdemeanors, or felonious assaults. The remaining offenses included burglaries, drug felonies, possession of hard drugs, status offenses, murder, rape, or arson. Interviews were conducted with the young people to examine crime-engendering variables such as strain and negative affect and strain-exacerbating and strain-alleviating variables such as peer delinquency, family communication, and coping skills. The study also included control variables to control for the possible influence of sex, race, age, and number of prior arrests. Findings revealed support for GST in that overall effects of principal variables were largely in the direction predicted by the theory. Promising support was found for the hypothesized influence of specific forms of negative affect, namely anger, in predicting interpersonal aggression but not property offending. The other measure of negative affect, namely depression, failed to exert a significant additive effect on either interpersonal or property offending. Interaction analyses between coping skills and negative affect did not particularly support GST except in two instances, where emotional and spiritual coping skills inhibited the effect of depression on property offending. Finally, the additive effect of strain remained significant even with controls for negative affect, coping skills, and several interactions. It appeared the relationship between strain and negative affect varied according to offense type, and young people who reported higher levels of social coping skills also reported engaging in both property offending and interpersonal aggression. Implications of the findings for future research on the generalizability of GST are discussed. An appendix contains the scale items used in the study. 71 references and 6 tables