OBJECTIVE: The Working Alliance Inventory (WAI) has made great contributions to psychotherapy research. However, studies suggest the 7-point response format and 3-factor structure of the client version may have psychometric problems. This study used Rasch item response theory (IRT) to (a) improve WAI response format, (b) compare two brief 12-item versions (WAI-sr; WAI-s), and (c) develop a new 16-item Brief Alliance Inventory (BAI). METHOD: Archival data from 1786 counseling center and community clients were analyzed. RESULTS: IRT findings suggested problems with crossed category thresholds. A rescoring scheme that combines neighboring responses to create 5- and 4-point scales sharply reduced these problems. Although subscale variance was reduced by 11-26%, rescoring yielded improved reliability and generally higher correlations with therapy process (session depth and smoothness) and outcome measures (residual gain symptom improvement). The 16-item BAI was designed to maximize "bandwidth" of item difficulty and preserve a broader range of WAI sensitivity than WAI-s or WAI-sr. CONCLUSIONS: Comparisons suggest the BAI performed better in several respects than the WAI-s or WAI-sr and equivalent to the full WAI on several performance indicators.
OBJECTIVE: The Working Alliance Inventory (WAI) has made great contributions to psychotherapy research. However, studies suggest the 7-point response format and 3-factor structure of the client version may have psychometric problems. This study used Rasch item response theory (IRT) to (a) improve WAI response format, (b) compare two brief 12-item versions (WAI-sr; WAI-s), and (c) develop a new 16-item Brief Alliance Inventory (BAI). METHOD: Archival data from 1786 counseling center and community clients were analyzed. RESULTS: IRT findings suggested problems with crossed category thresholds. A rescoring scheme that combines neighboring responses to create 5- and 4-point scales sharply reduced these problems. Although subscale variance was reduced by 11-26%, rescoring yielded improved reliability and generally higher correlations with therapy process (session depth and smoothness) and outcome measures (residual gain symptom improvement). The 16-item BAI was designed to maximize "bandwidth" of item difficulty and preserve a broader range of WAI sensitivity than WAI-s or WAI-sr. CONCLUSIONS: Comparisons suggest the BAI performed better in several respects than the WAI-s or WAI-sr and equivalent to the full WAI on several performance indicators.
Authors: Craig Warlick; Kimber P Richter; Delwyn Catley; Byron J Gajewski; Laura E Martin; Laura M Mussulman Journal: J Subst Abuse Treat Date: 2017-12-29
Authors: Stephen A Maisto; Robert C Schlauch; Gerard J Connors; Ronda L Dearing; Kelly A O'Hern Journal: Alcohol Clin Exp Res Date: 2020-03-01 Impact factor: 3.455
Authors: Davy Paap; Yasmaine H J M Karel; Arianne P Verhagen; Pieter U Dijkstra; Jan H B Geertzen; Grieteke Pool Journal: Front Psychol Date: 2022-07-15