I conduct a series of tests of peer influence in juvenile incarceration facilities motivated by three mechanisms: criminal skill transfer, the formation of new criminal networks, and the social contagion of crime-oriented non-cognitive factors. Identifying peer influence off natural variation in small cohorts within the same facility, I find evidence consistent with social contagion: exposure to peers who come from unstable homes and have high levels of aggression leads to an increase in crime after release, as well as an increase in crime-oriented attitudes and behaviors. This effect persists despite controlling for the criminal experience and gang affiliation of the cohort, and is found in settings where youths are unlikely to interact after release.

Citation
Megan T. Stevenson, Breaking Bad: Mechanisms of Social Influence and the Path to Criminality in Juvenile Jails, 99 Review of Economics and Statistics 824–838 (2017).