How narrative source impacts convergence of ratings from the Social Cognition and Object Relations Scale–Global Rating Method with psychotherapy process measures

This study examines the construct validity of the Social Cognition and Object Relations Scale–Global Rating Method (SCORS-G; Westen, 1995; see also Stein & Mulford, 2018) by exploring the degree of convergence across different narrative sources (i.e., early memories [EM] and psychotherapy narratives [PT]) in relation to patient- and therapist-rated psychotherapy process measures. Using a university-based outpatient sample (n = 81), we found limited convergence for SCORS-G ratings across narrative type. First, paired t tests showed that the means for six of the eight SCORS-G dimensions differed significantly between the EM and PT narratives with the majority having a large magnitude of effect. Moreover, despite 29 significant correlations between a SCORS-G dimension and either an alliance or session quality variable, only two of the eight SCORS-G dimensions significantly correlated with the same process variable across narrative type (e.g., patient-rated session depth with SCORS-G Self Esteem [SE] and Identity and Coherence of Self [ICS]). Importantly, the high degree of theoretical coherence in the associations that emerged between the SCORS-G dimensions and the process variables suggest that the lack of convergence was not due to limited validity of the SCORS-G. Instead, the results underscore the importance of multi-method assessment techniques by highlighting that the manner in which a narrative is elicited will impact the object relational content patients provide. Future research and clinical implications related to the SCORS-G, alliance and psychotherapy process are discussed.