Under the two different hypotheses, the social context may impact juvenile delinquency either through a spreading effect, that is, increasing the risk of delinquency among all youths in socially unstable neighborhoods, or through a multiplicative effect that implies only certain youths will be affected in socially unstable contexts. The purpose of the study was to analyze the relationship between family interaction and self- reported delinquency among students in schools in socially stable and unstable neighborhoods, respectively. Data were taken from the Stockholm Project's School Survey, which involved approximately 340 boys in seven junior high schools. A tentative conclusion is that both a spreading effect and a multiplicative effect concerning students' delinquency were discernible; however, a somewhat stronger support is found for the multiplicative hypothesis. Under the multiplicative hypothesis, certain students in unstable contexts should have a higher probability of deviant behavior; the data show that students with weak family interaction have a much higher probability of deviant behavior in the unstable areas than similar students in schools in the more stable areas.