Trainee's anxiety and counseling self-efficacy in counseling session

The purpose of this study was first to develop and validate the Trainee's Anxiety in Clinical Work (TACW) scale. A total of 235 counselor trainees recruited nationally participated an online study at two different time points. The dataset was divided for exploratory factor analysis (Sample A; n= 118) and for confirmatory factor analysis (Sample B; n= 117). Three factors were identified, Supervisor's Negative Evaluation (alpha = .90), Advanced Counseling Skills (alpha = .81), and Client's Negative Evaluation (alpha = .77). The coefficient alpha for the TACW was .87. The TACW significantly predicted counseling self-efficacy over and above trait anxiety. The test-retest reliabilities ranged from .66 to .74 for the TACW and its three subscales. A paired-samples t-tests indicated that TACW is sensitive to detect the change for counselor trainee's anxiety in clinical work from the beginning to the end of practicum. Next, based on the Social Cognitive Model of Counselor Training (SCMCT) proposed by Larson (1998), this study conducted a cross-lagged panel model to examine the causal relationships among trainee's anxiety in clinical work, supervisory working alliance, and self-compassion and how those factors impacted trainee's counseling self-efficacy. The results found that supervisory working alliance at Time 1 and self-compassion at Time 1 would contribute to lessening trainee's anxiety in clinical work at Time 2, which in turn, would relate to their counseling self-efficacy at Time 2. The result for this model was after controlling for number of completed practicum. Finally, results from hieratical regression analyses indicated that after controlling for number of completed practicum, the interaction of supervisory working alliance at Time 1 and trainee's anxiety in clinical work at Time 1 significantly predict counseling self-efficacy at Time 2. However, the interaction of self-compassion at Time 1 and trainee's anxiety in clinical work at Time 1 did not significantly predict counseling self-efficacy at Time 2. Specifically, results from simple effects indicated trainees with higher anxiety in clinical work (Time 1), the association between supervisory working alliance (Time 1) and counseling self-efficacy (Time 2) was significantly positive. However, for trainees with lower anxiety in clinical work (Time 1), the association between supervisory working alliance (Time 1) and counseling self-efficacy (Time 2) was not significant. In other words, counseling self-efficacy (Time 2) would remain relatively high no matter the levels of supervisory working alliance (Time 1). Alternatively, results from simple effects revealed that trainees with lower supervisory working alliance (Time 1), the association between trainee's anxiety in clinical work (Time 1) and counseling self-efficacy (Time 2) was significantly negative. However, for trainees with higher supervisory working alliance (Time 1), the association between trainee's anxiety in clinical work (Time 1) and counseling self-efficacy (Time 2) was not significant. That is to say, counseling self-efficacy (Time 2) would remain high regardless the levels of trainee's anxiety in clinical work (Time 1).