Alliance Check (teaching instrument)
Practise soliciting and interpreting alliance feedback using Bordin’s three components.
Complete it
Thinking about today’s session.
My therapist and I are working towards what I actually want to change
Goals
We agree about what is most important for me to work on
Goals
What we did today made sense as a way of helping
Tasks
I understand why we are doing things this way
Tasks
I felt understood today
Bond
I could say something difficult here if I needed to
Bond
Address directly at the start of the next session, before anything else.
Psychometrics
The construct comes from Bordin’s pantheoretical reformulation of the alliance as agreement on goals, agreement on tasks, and the affective bond. The alliance–outcome association is among the most replicated findings in psychotherapy research, with a correlation of roughly r = 0.28 across hundreds of studies — modest in size, remarkably robust across modalities, measures and raters.
Cautions and misuse
Patients systematically over-rate the alliance to the person they are rating. Treat a score of 22/24 as a normal result and a drop of two points as a signal. Ask about the lowest item aloud rather than accepting the total.
Therapist and patient ratings of the same session correlate only modestly, and the patient’s rating is the better predictor of outcome. When they diverge, believe the patient.
Sources
Bordin, E. S. (1979). The generalizability of the psychoanalytic concept of the working alliance. Psychotherapy: Theory, Research & Practice, 16(3), 252–260
articleFlückiger, C., Del Re, A. C., Wampold, B. E., & Horvath, A. O. (2018). The alliance in adult psychotherapy: A meta-analytic synthesis. Psychotherapy, 55(4), 316–340
meta-analysis