Validation of an instrument to measure community capacity building for prevention

Abstract

Abstract Aim: In Germany, the CTC-EFF study, a replication study of the U.S. Community Youth Development Study (CYDS), was rolled out in 2021. This article aims to examine the validity of a translated and adapted version of an instrument that measures five constructs of community capacity for prevention. Subject and Methods: Twenty-two a-priori matched intervention and comparison communities participate in the study. Two to ten community key informants (n = 182) were interviewed for each community. Exploratory (EFA) and confirmatory factor analyses (CFA) were conducted to assess structural validity. Reliability was assessed at the individual level using McDonald’s omega and at the community level using a generalizability coefficient (GC). Inter-rater agreement was measured using an intraclass correlation coefficient (ICC). Concurrent construct validity was assessed using bivariate correlations and mean comparisons. Results: CFA showed good model fit (CFI = 0.964) for the hypothesized five constructs model. Reliability analyses showed good internal consistency at the individual level (omega = 0.86 – 0.87) and insufficient to moderate reliability at the community level (CG = 0.00 – 0.62). Inter-rater agreement ranged from insufficient to substantial (ICC = 0.02 - 0.32). Consistent with theory, three constructs correlated positively at the individual and community level (r = 0.24 - 0.42). There are no associations between these constructs and interviewer ratings about the respondents’ cooperativeness, trustworthiness, and understanding. Conclusion: Three of the five constructs were measured with good validity. Two constructs show insufficient reliability. These will be revised until the next CTC-EFF data collection wave. Further validation is indicated. Trial registration: This study was registered with the German Clinical Trial Register: DRKS00022819 on Aug 18, 2021.