Using complementary methods to test whether marriage limits men's antisocial behavior

Married men engage in significantly less antisocial behavior than unmarried men, but it is not clear whether this reflects a causal relationship. Instead, the relationship could reflect selection into marriage whereby the men who are most likely to marry (men in steady employment with high levels of education) are the least likely to engage in antisocial behavior. The relationship could also be the result of reverse causation, whereby high levels of antisocial behavior are a deterrent to marriage rather than the reverse. Both of these alternative processes are consistent with the possibility that some men have a genetically based proclivity to become married, known as an active genotype-environment correlation. Using four complementary methods, we tested the hypothesis that marriage limits men's antisocial behavior. These approaches have different strengths and weaknesses and collectively help to rule out alternative explanations, including active genotype-environment correlations, for a causal association between marriage and men's antisocial behavior. Data were drawn from the in-home interview sample of the National Longitudinal Study of Adolescent Health, a large, longitudinal survey study of a nationally representative sample of adolescents in the United States. Lagged negative binomial and logistic regression and propensity scorematching models (n = 2,250), fixed-effects models of within-individual change (n = 3,061), and random-effects models of sibling differences (n 618) all showed that married men engaged in significantly less antisocial behavior than unmarried men. Our findings replicate results from other quasiexperimental studies of marriage and men's antisocial behavior and extend the results to a nationally representative sample of young adults in the United States.