Leveraging School Environments to Shape Social Networks and Improve Adolescent Health-A Randomized Trial of a Social Networks Intervention

Project Summary

The project team evaluated the impact of Advancement via Individual Determination (AVID) – a college preparatory curriculum targeting students in the academic middle – on students’ academic performance and health outcomes. The team explored whether and how AVID changed peer networks and relationships with teachers, whether those changes led to improvements in academic and behavioral outcomes and, if so, what the relative importance of peer versus adult network changes were. In addition to examining the effects of AVID on participants, the team examined how exposure to AVID impacted the academic achievement and health of high-performing non-AVID peers.

Research Question(s)

  • Did students randomized to AVID report healthier social networks (lower proportion of peers engaging in risky behaviors and a higher number of supportive adults) than control students?
  • Did AVID students report decreased frequency of risky health behaviors (e.g., 30-day alcohol, drug use, and decreased frequency of fighting, etc.) compared to control students?
  • Were there any associations between AVID and risky health behaviors modified by gender?
  • Did peer or adult social network changes predict changes in substance use, violence, and delinquency? If exposure to AVID is associated with improved health, were these associations mediated by changes in peer or adult networks?


  • Guide school and district decision-making around implementation and allocation of resources for social networking programs such as AVID, by providing information about whether AVID improved student academic achievement and/or health of both participant and non-participant peers.


Use of tobacco and electronic cigarettes; fighting and weapon carrying; gang involvement and non-violent delinquency; proportion of peers engaging in risky health behaviors and number of supportive adults; school engagement, attendance, and grades.


Overarching study design is a longitudinal, randomized evaluation. To test rates and frequency of risky health behaviors, the team will conduct mixed effects linear regression models accounting for clustering at the school level and repeated measurements within students. Intent-to-treat analysis using mixed effects regression models will be used to evaluate whether students randomized to AVID have healthier social networks, a larger number of supportive adults in their network, and a lower proportion of peers engaging in violence and delinquency. 

Young teens in class at their desks
Grantee and Partner organizations

University of California, Los Angeles
Los Angeles Unified School District
University of Southern California

Grant status
Principal investigators
Rebecca Dudovitz, MD, MSHS
Start date
Award amount
30 months

Stay Connected