Measurement Invariance and Criterion Validity of Comprehensive Knowledge about HIV/AIDS Prevention Scale in the Uganda Demographic and Health Survey Tool

Background Limited information exists on the functioning of comprehensive knowledge about HIV/AIDS prevention scale in the Uganda Demographic and Health Surveys. Objectives This paper aimed to: (i) examine measurement invariance of comprehensive knowledge about HIV/AIDS prevention scale across men and women groups in Uganda; and (ii) evaluate the criterion related validity of the scale using HIV testing as an outcome variable. Methods The study was based on cross-sectional Uganda Demographic and Health Survey data of 2016. Measurement invariance was investigated using confirmatory factor analysis in the framework of structural equation modelling while criterion-related validity was investigated by fitting a binary logistic regression model that explained the relationship between HIV testing and comprehensive knowledge about HIV/AIDS prevention. Results The results show that, the construct is invariant across men and women groups at the dimensional, metric and scalar levels, however, all models presented poor fit. Furthermore, criterion-related validity of comprehensive knowledge about HIV/AIDS prevention with HIV testing, was confirmed. Conclusions The findings of this study underscore the need to revise items included in the comprehensive knowledge about HIV/AIDS prevention scale in order to improve its performance.