A self-report scale to measure adolescents' involvement in parental figures' conflicts, disagreements, or relationships, a phenomenon known as triangulation, was developed, and initial evidence for reliability and validity was obtained using a three-study format.In Study 1, the 24-item Triangular Relationship Inventory (TRI) was developed using classical test theory in a sample of college-aged students and found to have a multidimensional factor structure. In Study 2, the multidimensional factor structure was replicated in a new sample of college-aged students. Evidence of reliability was supported with excellent internal consistency. Construct validity was supported by replicating the multidimensional factor structure of the TRI. A strong correlation between the TRI and the Intergenerational Triangulation subscale of the Personal Authority in the Family System Questionnaire indicated criterion-related validity. Convergent validity of the TRI was indicated by a negative correlation between the TRI and the Differentiation of Self Inventory-Short Form, a positive correlation between the TRI and the Hopkins Symptom Checklist-21, and a positive correlation between the TRI and the State-Trait Inventory for Cognitive and Somatic Anxiety. Discriminant validity was evidenced by a low correlation between the TRI and the Satisfaction With Life Scale. Study 3 found the test-retest reliability of the TRI over a 4-week period to be strong, at .84. In summary, the TRI appears to have good psychometric properties and contributes to the literature in family systems theory by enabling a more nuanced measurement of triangulation than has been previously available.

Based in Bowen theory (Kerr, 2019; Kerr & Bowen, 1988), this study examined differentiation of self as a moderator of the relationship between anxiety and current triangling among a sample of adolescents. It also explored the relationship between specific triangle subtypes and psychological symptoms. We hypothesized that higher levels of anxiety would be associated with higher levels of triangling, and that this relationship would be weaker for participants who report higher levels of differentiation of self. Additionally, we hypothesized that psychological symptoms would be significantly lower for those who indicated a balanced triangle configuration to characterize their family structure (as compared to those selecting scapegoating, cross‐generational coalition or mediator patterns). Results partially supported our hypotheses; differentiation performed as a moderator in two analyses, but not in others. Higher levels of psychological symptoms were associated with experience with mediator and cross‐generational triangling.

No abstract