Gottfredson and Hirschi’s self-control theory represents one of the most well articulated and extensively tested explanations of crime in the field of criminology. To date, studies have found considerable support for the theory’s major tenets. However, questions remain regarding the validity of self-control constructs that have been used in research. Traditionally, researchers have tested self-control theory using an attitudinal scale, measuring the risk aversion and impulsivity of respondents. Recently, Hirschi (2004) offered a revised operational definition of self-control that he defines as the “inhibitions that one carries with them wherever one happens to go”. This study examines and compares the validity of these two measures. With data from the International Self-Report Delinquency (ISRD) study (US version), we use the Rasch measurement method, a form of Item Response Theory (IRT) to test attitudinal and Hirschi’s revised scale of self-control across a sample of US middle schools. We also examine the construct validity of the two measures focusing on the extent to which variables from alternative criminological theories (e.g., strain and social learning) mediate the relationship between each self-control measure and delinquency.