Crime and physical activity measures from the SAFE and Fit Environments Study (SAFE): Psychometric properties across age groups

• Main Authors: Scott C. Roesch, Christina M. Patch, Caterina G. Roman, Terry L. Conway, Ralph B. Taylor, Brian E. Saelens, Marc A. Adams, Kelli L. Cain, Loki Natarajan, James F. Sallis
• Format: Article
• Language: English
• Published: Elsevier 2021-06-01
• Series: Preventive Medicine Reports
• Subjects: Crime; Physical activity; Measurement; Psychometrics; Age
• Online Access: http://www.sciencedirect.com/science/article/pii/S2211335521000711

Valid and reliable measures are needed to better understand the relationship between physical activity and crime. This paper provides a comprehensive psychometric evaluation of measures developed in the Safe and Fit Environments (SAFE) Study to assess a crime-PA conceptual framework. In addition to assessing the basic psychometric properties of each measure (e.g., variable distributions [item/scale level], internal consistency reliability), this study formally examined the measurement validity and invariance of measures across four age groups using confirmatory factor analysis. The sample (n = 2173) included 336 Adolescents (aged 12–17), 532 Young adults (aged 18–39), 838 Middle Age Adults, and 467 Older Adults (aged 66+). The psychometric evaluation of (sub)scales showed consistent factorial validity and internal consistency reliability across the majority of the measures and across the four age groups. Specifically, 14 of the 17 measures displayed statistically and practically significant factor loadings and internal consistency values in the overall sample and across the age groups. The pattern of correlations for each (sub)scale with other (sub)scales/indexes largely did not exhibit redundancy across measures. The findings expanded upon the test–retest reliability evaluation reported in Patch et al. (2019), and clarified key aspects of the construct validity of these indicators. The latter bodes well for potential utility of these indicators in future predictive models.