Abstract
This article presents the results of a confirmatory factor analysis expanding and enhancing a previously developed measure of “machismo” both to explore the dimensionality of the measure and to test the invariance of factor structure in a sample of Anglo, African American, and Mexican American adult male drinkers. Specifically, this article (a) describes the development of a multidimensional 13-item machismo measure, (b) tests the goodness of fit of a series of confirmatory factor models examining single versus multiple factor solutions (e.g., Does a unitary machismo factor fit, or must positive and negative dimensions be modeled?), and (c) tests the invariance of the resulting model across racial-ethnic groups. Results support the use of the measures for cross-ethnic studies and emphasize the importance of a multidimensional measure including both positive and negative dimensions.