Psychometric properties of the self-reflection and insight scale (revised) in German-speaking clinical psychology trainees

Self-reflection is a core competency for clinical psychologists and mental health professionals. Given its essential role in clinical practice and personal development, it is important to assess self-reflective capacity reliably and efficiently. This study aimed to translate and examine the psychometric properties of the Self-Reflection and Insight Scale-Revised (SRIS-R) for use in German-speaking contexts.

The SRIS-R was translated using a back-translation procedure. A total of 257 participants completed the questionnaires via an online survey.

Both the original 14-item factor structure and revised German model solutions were tested. Confirmatory factor analyses indicated that the original English factor structure could not be satisfactorily replicated in the German sample. Therefore, exploratory factor analyses using polychoric correlations were conducted and suggested structural differences in the German version. Based on these findings, a parsimonious revised 13-item two-factor scoring solution and a methodologically refined two-factor model including a cross-loading and a residual correlation were examined. The modified 13-item version showed improved internal consistency (α = 0.86 for Self-Reflection and α = 0.84 for Insight). The parsimonious revised 13-item scoring solution retained the conceptual distinction between Self-Reflection and Insight while being feasible for practical questionnaire scoring. Convergent and discriminant validity evidence supported theoretical associations of Self-Reflection with related constructs, particularly self-insight and reflection, whereas Insight showed negative associations with rumination and private self-consciousness. Additional ESEM analyses supported a two-factor representation while indicating relevant item-level cross-loadings. The 13-item structure was subsequently tested without modification in an independent sample (N = 161), showing comparable model fit and high factorial similarity across samples. Multigroup analyses further supported scalar invariance across the two samples. Together, these findings provide cross-sample support for the distinction between Self-Reflection and Insight, while indicating remaining structural complexity at the item level. Given the limited support for a general factor, interpretation of the two subscale scores rather than a total score is recommended.

However, further research is needed to replicate the revised structure in broader and more diverse German-speaking samples. A culturally adapted and efficient measurement, like the one presented here, is essential for fostering and assessing self-reflective skills in clinical professionals.