CROSS-CULTURAL VARIATIONS IN PERCEIVED CONTROL AND ANXIETY

Anxiety affects millions of adults in the United States. A variety of models have been proposed that describe subtypes of anxiety as well as causes and effects of anxiety. Perceived control has been identified as an important component of some of these anxiety models (e.g. Chorpita & Barlow, 1998). These models include different interpretations and conceptualizations of anxiety and of perceived control. Consequently, there are a variety of operational definitions and measures for both anxiety and perceived control. However, there is limited research that investigates the validity of the measures and the structural relationships between these constructs, especially across different populations. The purpose of this study was to evaluate the measurement and structural invariance of anxiety and perceived control across three adult populations: African Americans, Caucasians, and Hispanics/Latinos. Participants were 210 students from a large, ethnically diverse, urban University. Forty-four participants identified as African American, 67 as Caucasians, and 99 as Hispanic/Latino. Multi-group structural equation modeling (SEM) was used to evaluate measurement and structural models relating perceived control to anxiety. Strict measurement invariance was established in both the anxiety and perceived control models. Additionally, factor variances were invariant across the three groups. However, the structural relationships between the latent factors as well as the latent means were not invariant across groups. Hispanic/Latinos had a significantly higher correlation between the DASS (somatic and arousal symptoms of anxiety) and the SOC (personal and interpersonal control) than African Americans and Caucasians. Hispanic/Latinos also reported higher levels of perceived personal control than African Americans and Caucasians.