Family configuration and adolescent sexual behavior

Adolescent sexuality data are used to examine the relationship between family configuration and premarital intercourse. In the context of analysis of covariates models, family and parent variables are defined to control for possible spurious influences, and the family configuration variables of interest are observed against a well-controlled background. Several apparent relationships are shown to be spurious when possible confounders are controlled. However, number of brothers is shown to have a significant relationship with intercourse behavior controlling for a wide range of covariates. Number of original parents is shown to have a marginal relationship. A theoretical framework for further investigation of the number of brothers effect is suggested, including a masculinization hypothesis, a friends-of-brothers hypothesis, and a role model hypothesis.