How Bayesian estimation might improve CBS measure development: A case study with body-image flexibility in Hispanic students

Abstract

The methods for examining questionnaires in psychology are steeped in conventional statistics. However, many within the social sciences have started exploring Bayesian methods as an alternative to the conventional approach. This paper highlights the usefulness of Bayesian methodology for factor analysis, using the Body Image Acceptance and Action Questionnaire (BI-AAQ) as a case study. In an all-Hispanic undergraduate sample (n=289), we compared techniques from Bayesian and frequentist estimation for examining the factor structure of the BI-AAQ. Results indicated Bayesian estimation was flexible and offered unique insights relative to the conventional frequentist approach. We conclude the BI-AAQ was a structurally valid measure for our all-Hispanic sample and that Bayesian methods may be fruitful for further evaluation within the contextual behavioral science community.