Confirmatory factor analysis (CFA) provides a convenient framework for evaluating measurement invariance. In CFA we specify a measurement model that defines the factorial structure underlying the observed data (e.g., item responses). The factorial structure includes the number of latent variables (i.e., factors) and the pattern of factor loadings (see Chapter 5 for a more detailed description of CFA). If the CFA model adequately captures the factorial structure, then group membership will not provide any additional information about observed variables above and beyond that explained by the latent variable. In fact, DIF has been framed as a dimensionality issue (Ackerman, 1992) and CFA makes the role of dimensionality in the analyses more explicit.