The role of peer influence in the etiology of delinquent behavior has been a central concern in criminology for well over a decade. Despite this heritage, questions remain regarding the operational mechanism of social influence in the causation of delinquency as well as concern over the measurement of the central variable: peer delinquency. Using data from the Netherlands Institute for the Study of Crime and Law Enforcement School Project, I build on prior research in this area by examining perceptual measures of peer delinquency and social network measures of peer delinquency.
My primary findings suggests that past research seeking to resolve the debate regarding the source of the correlation between peer and individual behavior, has been contaminated by the use of perceptual measures of peer delinquency. I show that an individual’s current perceptions of peer delinquency are mainly the product of past perceptions and one’s own delinquent behavior, questioning the conclusions of prior research which have assumed that perceptions accurately measure peer behavior.
I also examine how distortions in perceptions of peer delinquency serve as a mechanism of social influence. I find that a variety of individual characteristics and social network characteristics lead individuals to overestimate the extent of delinquency and that these same individuals are more likely to engage in delinquency as a result of these misperceptions. Lastly, I examine the extent to which social influence can account for a difficult empirical fact: the age-crime relationship. I show that within-individual changes in delinquency with age are mainly produced by changes to perceptions of peer delinquency and contextual changes that occur during adolescence. Overall, this dissertation illustrates the implications of using different measures of peer delinquency to examine social influence.