Examining perceptions of peer substance use in friendships of different gender structure John H. Boman a IV*, Bryan Lee Miller b , John Stogner c , Laura E. Agnich b and Marvin D. Krohn d a Department of Criminal Justice, University of Wyoming, Laramie, WY; b Department of Criminal Justice and Criminology, Georgia Southern University, Statesboro, GA; c Department of Criminal Justice and Criminology, University of North Carolina at Charlotte, Charlotte, NC; d Department of Sociology and Criminology & Law, University of Florida, Gainesville, FL (Received 29 June 2012; final version received 5 February 2013) Peer delinquency is nearly always operationalized with an individual’s perceptions of his/her peers’ behavior rather than the peers’ actual behavior. A criticism of these measures is that there is often a discrepancy between the perception of peer behavior and actual peer behavior. The mechanism that is most often discussed regarding the reason underlying misperceptions is ‘projection,’ or the situation when one respondent uses his/her deviant behavior to describe the behavior of his/her friends. Using data from a large sample of friendship pairs in emerging adulthood, we employ self-reported and perceptual measures of substance use to determine how the gender structure of friendship dyads affects the construction of perceptions of a peer’s substance use. We examine the projection hypothesis generally, but our specific interest lies in how both the gender of the respondent and the gender composition of the dyad affect the cognitive perception of a close friend’s use of drugs. Results indicate that perceptions vary greatly across friendships of different gender structures, and those in split gender friendships generate perceptions that most closely reflect the peer’s actual use patterns. Keywords: substance use; gender; perceptions; peer delinquency; measurement Introduction While criminologists have consistently found peer delinquency to be one of the strongest correlates of criminal behavior (Warr 2002), research has questioned whether the construct frequently labeled ‘peer delinquency’ is an appropriate measure of a peer’s behavior (e.g., Newcomb 1961, Byrne and Blaylock 1963, Gottfredson and Hirschi 1987, 1990, Jussim and Osgood 1989, Aseltine 1995, Thornberry and Krohn 1997, Haynie and Osgood 2005). This is because ‘peer delinquency’ is rarely operationalized using a measure that represents the peer group’s actual delinquency, but rather is measured through one’s perception of the peer group’s delinquency. Therefore, the construct traditionally referred to as ‘peer delinquency’ may be more accurately referred to as ‘perceptual peer delinquency.’ This difference is significant. If a peer’s actual delinquency is different from the perceived delinquency then, in some authors’ views (such as Gottfredson and Hirschi 1987, 1990, Aseltine 1995, Thornberry and Krohn 1997, Haynie and Osgood 2005), the q 2013 Midwestern Criminal Justice Association *Corresponding author. Email: jboman@uwyo.edu Journal of Crime and Justice, 2013 http://dx.doi.org/10.1080/0735648X.2013.777210