The study found that how peer deviance is measured has a strong influence on the magnitude of the link between self-control and deviant behavior. When an indirect measure of peer delinquency (using a perceptual measure from the subject regarding his/her best friend) is replaced with a direct measure (self-reported deviance from each subject's self-identified peer), the links between self-control - attitudinal and behavioral measures - and deviance/criminal behavior were consistently larger. This suggests that the strength of the link between self-control and deviant/criminal behavior is dependent on how peer deviance is measured, regardless of whether self-control is measured by attitude or behavior. Data from 796 friendship pairs were used to estimate a series of regression models that regress respondent deviance onto indirect and direct peer deviance and attitudinal and behavioral self-control measures, while controlling for elements of the social bond and demographic characteristics. 5 tables, 12 notes and 65 references