Influence of peer networks on health risk behavior

Study Design and Specific  Aims:

We propose to use data from the National Longitudinal Study of Adolescent Health (Add Health), an ongoing nationally representative longitudinal study of youth in grades 7-12 in 1994 in 132 randomly selected high schools and middle schools in the U.S. Data on peer networks come from the in-school survey, collected in 1994.  All students attending school on the day of the self­ administered questionnaire in each of the schools were surveyed.  Each student filling out the in­ school questionnaire was asked to nominate up to five of their closest female and five of their closest male friends. They identified their friends by name from school rosters and entered a corresponding identification number.  Because each student attending school was surveyed, it is possible to link all of the students together by friendship ties and create a global school friendship network. Following the in-school surveys, four longitudinal waves of data were collected (in-home interviews). These data allow us to address the following aims:

 

1.   Identify the influence of the density (i.e., cohesion / integration) of the adolescent's peer network and of the global school network on health risk behaviors.

2.   Identify the influence of adolescent's prestige within the school friendship network (i.e., popularity, isolation) on the adoption and maintenance of health risk behaviors.

3.   Identify the influence of adolescents' position (i.e., core, periphery, bridge) in their peer network on the adoption and maintenance of health risk behaviors.

4.   Determine how network density (aim 1) prestige (aim 2) and position (aim 3) interact with peers' health risk behaviors and health attitudes to affect adolescent health risk behavior.

5.   Investigate potential heterogeneity by race/ethnicity and gender in 1-3 above and consider how adolescent peer networks contribute to race/ethnic and gender differentials in adolescent health risk behavior and health disparities in young adulthood.

6.   Investigate how network properties and health risk behaviors evolve over time using longitudinal network data and methods that account for the interdependence among relations.