Mallinckrodt Brent, Tekie Yacob T
a Department of Psychology , University of Tennessee , Knoxville , TN , USA.
Psychother Res. 2016 Nov;26(6):694-718. doi: 10.1080/10503307.2015.1061718. Epub 2015 Nov 7.
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).
Archival data from 1786 counseling center and community clients were analyzed.
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.
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.
工作联盟量表（WAI）对心理治疗研究做出了巨大贡献。然而，研究表明，客户版的7点反应格式和三因素结构可能存在心理测量学问题。本研究使用拉施克项目反应理论（IRT）来（a）改进WAI反应格式，（b）比较两个简短的12项版本（WAI-sr；WAI-s），以及（c）开发一个新的16项简短联盟量表（BAI）。
分析了来自1786名咨询中心和社区客户的档案数据。
IRT结果表明存在交叉类别阈值问题。一种将相邻反应组合以创建5点和4点量表的重新计分方案大幅减少了这些问题。尽管子量表方差减少了11%-26%，但重新计分提高了信度，并且与治疗过程（会话深度和平滑度）及结果测量（残余增益症状改善）的相关性总体上更高。16项BAI的设计旨在最大化项目难度的“带宽”，并比WAI-s或WAI-sr保留更广泛的WAI敏感性范围。
比较表明，BAI在几个方面比WAI-s或WAI-sr表现更好，并且在几个性能指标上与完整的WAI相当。
