MULTISAMPLE CONFIRMATORY FACTOR-ANALYSIS TO TEST FOR CONSTRUCT BIAS -AN EXAMPLE USING THE K-ABC

Recent research designed to investigate construct bias in the Kaufman Assessment Battery for Children served the additional purpose of illus trating several useful methods for detecting construct bias. The purpo se of this reanalysis was to illustrate a new, powerful method for tes ting the similarity of the constructs measured across groups (and, thu s, detecting construct bias): multi-sample confirmatory factor analysi s (MCFA). Our reanalysis suggested that the K-ABC measures indistingui shable abilities for Black and White children at ages 7 through 8. Tho se analyses also showed differences in the attributes measured for Bla ck and White children for ages 9 through 12. Follow-up demonstrated th at the differences among that latter group were due to trivial aspects of measurement rather than critical differences in the constructs mea sured by the test across groups. We argue that MCFA provides a more or ganized, direct, objective, and complete method for detecting construc t bias than do other methods.