Measuring Perceptions of the Therapeutic Alliance in Individual, Family, and Group Therapy from a Systemic...

This study examined the multidimensional structure of the client and therapist versions of the self-report measure, System for Observing Family Therapy Alliances (SOFTA-s; Friedlander, Escudero, & Heatherington, Therapeutic alliances in couple and family therapy: An empirically informed guide to practice. Washington, DC: American Psychological Association, 2006) across three distinct therapeutic modalities (individual, family, group). Specifically, we investigated whether the originally theorized model of four first-order factors (Engagement in the Therapeutic Process, Emotional Connection with the Therapist, Safety within the Therapeutic System, and Shared Sense of Purpose within the Family) would be reflected in a second-order factor (Therapeutic Alliance). The sample included 105 therapists who worked with 858 clients (165 individuals, 233 families, and 43 groups) in several Spanish community agencies. To control for dependent data, we used multilevel modeling. Results of the multilevel confirmatory factor analyses showed adequate reliabilities, fit indices, and factor loadings across the three therapy contexts for both versions of the measure (client and therapist). Adequate measurement invariance was also found across respondents and therapy modalities. Taken together, these results support the structural validity of the SOFTA-s, a brief and flexible self-report alliance measure that can be used reliably in clinical practice as well as in studies of individual, family, and group therapy.