An item response theory approach to punitive attitudes

Abstract

Concerns about how punitive attitudes are measured are long-standing in the academic literature. However, empirical research in this area has often overlooked variations in the difficulty of punitive items and differences in individual dispositions. To address these challenges, this paper explores the measurement of punitive attitudes through the application of Item Response Theory to a representative sample of Uruguayan citizens. By addressing the limitations of Classical Test Theory in criminological research, we provide a more nuanced understanding of punitive attitudes, distinguishing the difficulty of survey items from respondents' underlying traits. Our analysis highlights significant variation in item difficulty and discrimination, showing that assuming that all punitive measures reflect equivalent levels of punitiveness is problematic. Our findings identify gaps in the measurement of individuals with higher levels of punitiveness and suggest that frequently used survey items may fail to capture the full spectrum of punitive attitudes. This research emphasizes the need to refine survey instruments to enhance the validity and reliability of scales of punitive attitudes, thereby contributing to a more comprehensive understanding of public opinion regarding crime and punishment.