There have been an increasing number of studies in criminology that use network analysis to understand the effect of peer relationships on delinquency.  However, relatively little attention has been given to the distinction between dyadic and triadic relationships. Given the theoretical importance of the distinction, first made by Simmel (1950), this is an important issue that needs to be addressed. Drawing from Simmelian tie theory (Krackhardt, 1999) and using the two waves of AddHealth data, this project investigates the differential effects of dyadic and triadic peer relationships on delinquency and attempts to advance the understanding of causal mechanisms of peer influence.