Italian Version of the Relationship Profile Test (RPT-I): Temporal Stability, Construct Validity, and Cross-Cultural Comparison

Abstract

This study presents the adaptation to the Italian context of the Relationship Profile Test (RPT; Bornstein & Languirand), a self-report measure of Destructive Overdependence (DO), Dysfunctional Detachment (DD), and Healthy Dependency (HD). The RPT was administered to a community sample of 661 nonclinical Italian adults together with the Attachment Style Questionnaire, the Relational-Interdependent Self-Construal Scale, the Rosenberg Self-Esteem Scale, the Self-Compassion Scale, the Positive Affect-Negative Affect Scale, and the Toronto Alexithymia Scale. A randomly selected subset of participants (n = 67) completed the RPT again approximately 5 months after the first administration. The factor structure of the RPT obtained in the main sample was compared with that obtained in a sample of 603 adult participants from the US and was found to be similar. Internal consistency for DO, DD, and HD scores in the Italian sample fell between the acceptable to good range, and test-retest reliability coefficients were all above .70. The three scales yielded the expected pattern of correlations with theoretically related constructs, documenting good criterion validity. Findings are discussed in light of the literature on the RPT as a measure of interpersonal dependency and detachment. Suggestions for future research are offered.

Data availability statement

Data and syntax for this study have been uploaded by the authors on a secure public repository (i.e., www.osf.io) currently accessible through a private weblink. Such material will be made available upon reasonable request to the corresponding author.

Notes

1 Abuín & De Rivera (2015) used a Spanish translation of the RPT developed by Abuín et al. (2007) that is not publicly available.

2 Of note, Denckla and colleagues reported the correlation between DO and HD and the SCS, but they used a different scoring algorithm (the total score Vs. the two SCR and SUCR). Although more recent international evidence (Halamová et al., 2021) discourages the use of a single factor for the SCS, we calculated it in the Italian sample to allow for a consistent comparison of correlations between the RPT and self-compassion in the Italian and US populations.

3 In the US sample the first ten eigenvalues were 8.69, 3.19, 2.33, 1.61, 1.37, 1.09, 1.06, 0.84, 0.77, and 0.69, and the corresponding random eigenvalues were, on average, 1.43, 1.38, 1.33, 1.30, 1.26, 1.23, 1.20, 1.17, 1.14, and 1.12. In the Italian sample, the first ten eigenvalues were 4.38, 3.48, 2.17, 1.65, 1.38, 1.37, 1.19, 1.12, 1.01, and .94, and the corresponding random eigenvalues were 1.44, 1.38, 1.34, 1.30, 1.26, 1.23, 1.20, 1.17, 1.14, and 1.09. Figure 1 graphically represents scree plots and Parallel Analyses.