The short Working Alliance Inventory in parent training: Factor structure and longitudinal invariance

OBJECTIVE: The purposes of the present study were twofold: (a) to examine the factor structure of the Working Alliance Inventory, Short (WAI-S) and (b) to investigate if factor loadings and thresholds fulfilled properties of longitudinal measurement invariance across two waves of data. METHOD: The study sample consisted of 259 Norwegian parents receiving Parent Management Training, the Oregon model. Parents rated alliance at sessions 3 and 12 during the therapy. Confirmatory factor analyses to assess the fit of a one-, two-, and three-factor model were performed using robust weighted least squares estimation for categorical indicators. RESULTS: The results showed that data provided best fit for the three-factor solution with goal, task, and bond. Furthermore, results demonstrated satisfactory invariance for factor loadings and thresholds across time. CONCLUSIONS: Overall, the results indicate that the WAI-S three-factor solution has acceptable psychometric properties for longitudinal measurement comparisons.