Psychometric properties of the Chilean version of the Householder Bushfire Risk Perception Scale (HBRPS-4)

Abstract Background Wildfire risk perception is a relevant component of household preparedness and protective behavior. However, brief and specific measures of this construct with psychometric evidence beyond the contexts in which they were originally developed remain scarce, particularly in Latin America. This study adapted and evaluated the psychometric properties of the Chilean version of the Householder Bushfire Risk Perception Scale (HBRPS-4). Methods A cross-sectional instrumental study was conducted with a probability sample of 1,600 adults living in six municipalities in south-central Chile exposed to moderate to very high wildfire risk. The adaptation process included linguistic and contextual review, cognitive interviews, and pilot testing. Dimensionality, item functioning, measurement precision, measurement invariance, and differential item functioning (DIF) by sex and area of residence were examined using complementary factor-analytic, network, and item response theory approaches. Results Findings supported an essentially unidimensional structure and adequate reliability (marginal reliability = 0.85; empirical reliability = 0.90). Item response theory analyses showed generally adequate discrimination and provided evidence on the precision of the scale across the latent trait continuum, with greater information across the central range of the latent trait and lower precision toward its upper end. Measurement invariance analyses supported score comparability by sex and, with greater caution, by area of residence. Although statistically significant DIF was detected for some items, its magnitude was small and did not reach practical relevance. Conclusions The Chilean HBRPS-4 showed favorable psychometric evidence supporting the interpretation of its scores as a brief measure of household wildfire risk perception. The scale appears suitable for population-based research and group-level analyses in adults living in wildfire-prone settings. Further research should examine evidence based on relations with other variables, temporal stability, and cross-cultural comparability.