Browsing by Author "Brannick, Michael T."

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    A comparison of internal, external, and successive unfolding based on occupational judgments
    (1985) Brannick, Michael T.; Hahn, June
    Successive unfolding is a recently developed technique that appears to have advantages over traditional unfolding techniques. To assess its usefulness, successive, internal, and a type of external unfolding were compared with regard to the stimulus configurations recovered, the fit of the models as a whole, the fit of individual subjects, and each model’s stability in a cross-validation sample. The data were obtained from judgments of similarity of and preference for occupational titles. Internal unfolding yielded degenerate solutions and was dropped from subsequent analyses. External and successive unfolding yielded interpretable dimensions, but the nature of the dimensions uncovered differed somewhat from one type of unfolding to the other. Both types of unfolding fit the data adequately for the stimulus configuration, though successive unfolding appeared to fit better. The first two dimensions of both models were well reproduced in cross-validation samples. Dimensions beyond the second were more easily cross-validated using external unfolding. Successive unfolding proved superior to external unfolding in representing preference data as indicated by the fit of ideal points. Successive unfolding appears a viable method in unfolding research.
  • Thumbnail Image
    Item
    Estimation problems in the block-diagonal model of the multitrait-multimethod matrix
    (1990) Brannick, Michael T.; Spector, Paul E.
    The most popular method used to analyze the multitrait-multimethod (MTMM) matrix has been confirmatory factor analysis (CFA). The block-diagonal model, in which trait effects, trait correlations, method effects, and method correlations are simultaneously estimated is examined in detail. Analysis of published data from 18 correlation matrices showed estimation problems in all but one case. Simulations were used to show how identification and specification difficulties may account for these problems. Even trivial misspecification of a single parameter can prevent program convergence. These problems render the CFA block-diagonal approach to analyzing MTMM data less useful than has generally been thought. Index terms: construct validity, covariance structure modeling, factor analysis, multitrait-multimethod matrix, parameter estimation in confirmatory factor analysis.