Prior research examining the effect of self-control and delinquent peers on crime suggests that both variables are strong correlates and that controlling for one fails to eliminate the effects of the other. Yet prior research was based on indirect and possibly biased indicators of peer delinquency. Recent research using direct measures of delinquent peers, as reported by respondents' peers themselves, indicates that the relationship between peer delinquency and self-reported delinquency is smaller than when respondents report on their peers' behavior. The present study extends this line of work by examining the effect of self-control on delinquency when controlling for these two measures of delinquent peers. The results indicate that the effect of self-control is greater in magnitude in models using the direct measure of peer delinquency relative to models that rely on the traditional measure of delinquent peers. An interaction between self-control and the direct measure of peer delinquency was also found. Implications for future theory testing are discussed.