Uncovering Measurement Issues in Collective Efficacy:A Rasch-Based Evaluation

Survey items for collective efficacy are commonly included and used as composite scores in crime studies, serving as the basis for comparisons of groups, residential areas, and time periods. We argue that examining the measurement properties of items and assessing their quality is important to justify such comparisons. This study evaluates the measurement properties of a set of twelveitems designed to assess collective efficacy in the City of Stockholm Crime Survey.The analysis is based on three random samples of Stockholm residents aged 16 to 79, comprising a total of 47,906 individual responses. The Rasch partial credit model were used. Results indicate significant issues with the current item composition. Neither the use of all twelve items to measure collective efficacy nor the two theoretical subdimensions (social cohesion and trust, and informal social control) could be shown to be psychometrically justified.This study provides valuable insights into the measurement properties of items intended to assess collective efficacy. A significant strength of the study lies in its rigorous psychometric evaluation and the several limitations it reveals about the validity of the existing scale. The psychometric methods applied are also relevant for other surveys aiming to combine items measuring a latent variable.