The Shedler-Westen assessment procedure (SWAP): evaluating psychometric questions about its reliability, validity, and impact of its fixed score distribution

The Shedler-Westen assessment procedure (SWAP) is a personality assessment instrument designed for use by expert clinical assessors. Critics have raised questions about its psychometrics, most notably its validity across observers and situations, the impact of its fixed score distribution on research findings, and its test-retest reliability. We review empirical data addressing its validity, emphasizing the multitrait-multimethod approach to evaluating test validity. To evaluate the hypothesis that the fixed, asymmetric score distribution artifactually inflates correlations between SWAP profiles, we conducted Monte Carlo simulations and also presented empirical data from a large patient sample. We observed a mean correlation of zero between simulated SWAP profiles, indicating that the score distribution does not impact the correlation coefficients. Empirical correlations between SWAP profiles of actual patients were small and similar to those obtained using Diagnostic and statistical manual of mental disorders, fourth edition (DSM-IV) personality disorder scales that had no fixed score distributions, suggesting that the correlations were not a methodological artifact of the SWAP. We report new test-retest reliability data (median coefficient > .85) for the SWAP's trait and personality disorder dimensions. The SWAP appears to be reliable and valid. The data do not support its primary psychometric critiques.