Viviendo entre dos culturas: Evaluating the psychometric properties of the Psychological Acculturation Scale among Latinx caregivers.

Abstract

This study evaluated the structure, internal consistency, language-based measurement invariance, and construct validity of the Psychological Acculturation Scale (PAS), which assesses psychological acculturation via emotional attachment and belonging-based indicators. Analyses were conducted on a sample of 511 Latinx caregivers of youth with emotional/behavioral problems living in the United States (U.S.). Caregivers provided survey data either in English or Spanish on key sociodemographic factors (e.g., age, binary-gender, generational status in the U.S., language preference) and on psychological acculturation. Findings suggest that the PAS unidimensional factor structure had excellent internal consistency (α = .96, English α = .96, Spanish α = .90) and strong factor loadings (.75-.88) but fit indices for a confirmatory factor analysis were mixed. Post-hoc exploratory factor analyses were conducted to determine whether the model fit could be improved through the identification of a multidimensional structure, which uncovered two factors: (a) cultural identity and social belonging, and (b) cultural competence and familiarity. However, the two-factor solution fit better but had a high degree of overlap/redundancy (covariance = .923). A bifactor analysis suggest that the cultural identity and social belonging scale was unstable after accounting for the global scale and associated fit indices supported a unidimensional structure. The PAS unidimensional structure did not achieve configural invariance across English/Spanish versions, which suggest that differences in scores between language forms should be interpreted with caution. Our findings support the cautious use of the PAS as a brief unidimensional proxy measure of psychological acculturation to facilitate research with English/Spanish speaking Latinx caregivers.