Abstract:

Psychotherapy supervision is a core pedagogical method in professional training, often described as a “master-apprentice” or mentoring relationship that supports lifelong learning, reflective practice and the development of complex relational skills. Yet empirical supervision research, particularly on process-outcome links and supervision as a structured tutoring/mentoring practice, remains relatively scarce. Recent work across helping professions shows that a strong supervisory working alliance is associated with greater trainee engagement, programme satisfaction and self-efficacy, underlining the educational importance of this relationship.

This study is part of the national project “Psychotherapy Supervision Effectiveness and the Dynamics of Relationships in the Supervision Process (SUPER)” (2023–2026), which examines the effectiveness of psychotherapy supervision and its relational processes across different therapeutic orientations. We report data from 60 mental health professionals (30 supervisees, 30 supervisors) representing six psychotherapy approaches (Existential, Psychodynamic, Analytic, Cognitive-Behavioural, Gestalt or Adlerian). Both groups were predominantly women, with supervisors (34–62 years, M = 50.3) being more experienced than supervisees (26–57 years, M = 38.3). All supervisors had completed formal psychotherapy training, and most supervisees were enrolled in corresponding programmes. Data were collected twice during ongoing supervision: the first time at the beginning of the supervision process (during the 6th–10th supervision session), and the second time one year later. 23 supervisor-supervisee dyads continued work after one year and participated at both waves. The Supervisory Working Alliance Inventory (SWAI) and the Therapeutic Supervision Satisfaction Scale (TSSS) were used to assess supervisory working alliance and satisfaction with supervision, respectively.

The results show that supervisees consistently rated the supervisory working alliance higher than supervisors at both waves, particularly for the SWAI total score and the rapport and client focus dimensions at the first wave. At Wave 1, supervisees had a higher SWAI total mean rank than supervisors (41.40 vs. 19.60), U = 123.00, Z = −4.84, p < .001, and this pattern remained at Wave 2 (31.52 vs. 18.04), U = 126.00, Z = −3.33, p < .001; supervisees also scored higher on rapport (U = 280.00, p = .012) and client focus (U = 223.00, p < .001) at Wave 1. In contrast, group differences in supervision satisfaction (TSSS) were small and not statistically significant. For both supervisors and supervisees, SWAI and TSSS scores remained high and stable over time, with no significant changes from the first to the second measurement.

These findings highlight that, within psychotherapy training, supervision functions as a stable mentoring/tutoring context in which trainees perceive particularly strong relational support. The discrepancy between supervisor and supervisee views of the alliance suggests that supervision training should explicitly address shared goal setting, feedback and meta-communication about the relationship itself. Framing supervision as a structured mentoring pedagogy, this study supports the development of supervision curricula that place the supervisory working alliance at the centre, as a key educational outcome and a core quality indicator in tutoring and mentoring.

Keywords:

Mentoring, psychotherapy supervision, longitudinal study.