Based on the social-cognitive theory, this study investigated the dynamic association between counseling trainees' self-efficacy and their clients' outcome (i.e., symptom distress), and the mediating effects of therapist-and client-rated working alliance and session quality. Data set included 1,352 sessions conducted by 87 master's level therapist trainees with their 317 clients in China. Before every session, therapists completed a measure of self-efficacy and clients reported their symptom distress; and after every session, both therapists and clients rated their working alliance and session quality. Results showed that, (a) over the course of therapy, the initial level of therapist self-efficacy did not predict how quickly client symptom declined; (b) the initial level of client distress did not predict how quickly therapist self-efficacy increased; and (c) client distress decline was not related to therapist self-efficacy increase over the course of therapy. However, at the session-to-session level, higher therapist self-efficacy before one session significantly predicted lower client distress before the next session, and higher client distress before one session significantly predicted lower therapist self-efficacy before the subsequent session. Therapist and client perceptions of working alliance were both significant mediators, while their session quality ratings were not. Findings supported the dynamic and reciprocal relationship between therapist self-efficacy and client distress at the immediate session-to-session level rather than the overall trajectory level, and revealed its specific mediating mechanism through the relationship building between therapist and client. Public Significance StatementTherapist trainees' self-efficacy and their clients' distress temporally predict each other from session to session in a reciprocal way. When feeling more confident about their therapy sessions, trainees appear to build better relationships with clients, which leads to client symptom relief; in contrary, when clients are more distressed, it seems hard to have a strong working alliance, which undermines trainees' subsequent confidence.

This study aimed to investigate the multilevel factor structure of the therapist and client versions of the 12-item Working Alliance Inventory-Short Revised (WAI; Hatcher & Gillaspy, 2006) in the United States and China, and to create a three-item brief version (WAI-B3) using multilevel factor analysis (M-FA) and multilevel item response theory (M-IRT). We gathered eight data sets from two samples each in United States and China with a total of 21,623 sessions from 376 therapists and 2,455 clients. M-FA results with the first four data sets (two American and two Chinese) suggested that the 12-item WAI across therapist and client versions, and in both United States and China showed a dominant general WA factor with three specific subgroup factors corresponding to the Goal, Task, and Bond items. We then constructed a three-item WAI-B3 by selecting items 11, 10, and 9 through M-IRT, as the best representative of Goal, Task, and Bond subscales, respectively, for both the therapist and client versions in United States and China. With the other four data sets (two American, two Chinese) to test the WAI-B3, we found adequate multilevel reliability, structural validity, and convergent validity with the original 12-item WAI scores. Multilevel measurement invariance tests provided tentative and mixed support for the equivalence of WAI-B3 between the American and Chinese data sets and between therapist and client versions. We recommend that the WAI-B3 be used in routine clinical practice to track therapy process, and that the WAI-B3 be tested with other client and therapist populations and cultures. Public Significance StatementThe 12-item Working Alliance Inventory-Short Revised (WAI-SR; Hatcher & Gillaspy, 2006) is best represented by a predominant general working alliance factor and its total composite score can be used to describe both between-session changes or conduct between-therapy-dyad comparisons. Items 11, 10, and 9 in WAI-SR form a brief version WAI-B3 that has solid psychometric properties. Measurement equivalence is supported for WAI-B3 client version across United States and China, and across its therapist and client versions in the United States. Mixed support for equivalence is found for WAI-B3-Therapist across United States versus China, and therapist versus client versions in China. It is recommended that researchers and practitioners incorporate this brief WAI-B3 into their routine clinical practice to continuously track both therapist and client perceptions of their working alliance over the treatment and conduct further cross-cultural investigations.