Abstract
Actor-partner interdependence modeling (APIM; Kashy & Kenny, 2000) was used to study the early therapeutic alliance in 74 clients being treated by 29 therapists to explore the relationship between the alliance and treatment progress, while prioritizing the dyadic nature of the alliance. The APIM examines collaboration/influence by modeling the impact of one dyad member's alliance ratings on the other member's session impact rating (partner effects). In terms of the alliance, the results revealed significant client-actor effects for client ratings of session depth and positivity as well as significant therapist-actor effects for therapist ratings of session smoothness and positivity. For client-rated alliance, there were also significant client-partner effects for therapist ratings of session depth. For clients who made a reliable change in treatment, an increase in client-reported alliance was related to therapist reporting more arousal in the 3rd session. For clients who did not make a reliable change in treatment, client-reported alliance was not related to therapist arousal. Limitations of the study and implications of the findings are discussed.