Find in Library

In social science research, we often use a measurement with items answered in two ways: yes or no, pass or fail. The goal of this paper is to provide explicit guidance for substantive researchers interested in using the binary confirmatory factor analysis (CFA) through Mplus to validate the psychometric properties of a measurement with dichotomous items. Although the implementation of binary CFA using Mplus is simple and straightforward, interpreting the results is not as simple as implementation, especially for those who have limited experiences or knowledge in the Item Response Theory (IRT). We use an illustrative example from the National Longitudinal Survey of Youth (NLSY) to demonstrate the implementation of binary CFA of the Behavior Problem Index (BPI) using Mplus and the interpretation of the analysis results.