An unacceptably large number of adolescents continue to practice sexual and contraceptive behaviors that place them at risk of HIV/AIDS and other STD infection, as well as the risk of unintended pregnancy. A promising approach to more fully understanding adolescent sexual risk behavior is a """"""""contextual"""""""" approach and its focus on the larger spatial social contexts of communities, neighborhoods and schools in which adolescents are embedded. However, strong empirical evidence of the existence and nature of contextual effects is still lacking. The purpose of the proposed study is to further examine the linkages between extra-familial context and an adolescent's risk of engaging in sexual and contraceptive behaviors that lead to acquiring HIV and other STDs. Previous research has been rather unsuccessful in establishing these linkages by failing to account for the multi-dimensionality of contextual influences, by not fully examining how family and individual characteristics condition these contextual influences, and by lacking sufficient data sources. Our study is designed to overcome each of these obstacles. In particular, we apply the theoretical framework developed by Jencks and Mayer in which five key dimensions/mechanisms of extra-familial context are posited to affect an individual's behavior. Adopting this framework, we will define multiple-indicator measures of each of these underlying contextual dimensions that are theoretically relevant to an adolescent's sexual and contraceptive risk behavior. We will examine how these contextual dimensions affect each of eight aspects of this behavior, as well as how the contextual dimensions and changes in those contexts affect changes in the behaviors. We will also examine the extent to which an adolescent's age, gender, race/ethnicity, and family structure and process affect her or his susceptibility to the effects of the contextual dimensions on each risk behavior defined in our study. To accomplish these aims, we will estimate multilevel mixed-effects models. The study will be based on Wave 1 and Wave 2 data from Add Health. By applying the Jencks and Mayer framework and using this unique data set, the ultimate goal of the proposed study is to provide a theoretically driven and policy relevant explanation of how context influences an adolescent's sexual and contraceptive risk behavior.
