Date of Award


Document type


Access Type

Open Access Dissertation

Degree Name

Doctor of Psychology (PsyD)

Degree Program


First Advisor

Michael J. Constantino

Second Advisor

Elizabeth Harvey

Subject Categories

Clinical Psychology


The quality of the patient-therapist relationship, or therapeutic alliance, is widely viewed as an important element of the psychotherapeutic process. Empirically, the therapeutic alliance is a well-established and robust predictor of therapeutic change. With its clear impact on therapeutic success, researchers have increasingly examined factors that contribute to alliance development, including patient psychological characteristics. This study examined the relationship between patients' object relations (i.e., mental representations of self and others) and alliance quality, and whether timing of the alliance rating and the rater perspective (patient vs. therapist) moderated this relationship. Participants were 73 patients and 23 therapists from two outpatient mental health clinics. Patients completed the Bell Object Relations Inventory at baseline, and both patients and therapists completed the Working Alliance Inventory across multiple therapy sessions. On average, patients perceived the alliance more positively than their therapists, and there was a small, but statistically significant, correlation between their perspectives. Patients' general object relations deficits, as well as greater alienation and insecure attachment, were associated with more negative patient-rated alliance quality. On the other hand, patients' greater egocentricity was associated with a more positive patient-rated alliance perception. Patients' object relations did not significantly predict therapist-rated alliance, but symptom severity did. Less severe patient symptomatology was associated with more positive therapist alliance perception. Symptom severity did not predict patient-rated alliance. The number of sessions in which patients engaged was positively associated with therapist-rated alliance, while patient-rated alliance remained stable across sessions. Time of the alliance assessment did not moderate the relationship between object relations and either patient- or therapist-rated alliance. The results suggest that it may be important to consider patients' presenting quality of object relations for treatment planning and for negotiating the therapeutic alliance.