The construction and psychometric evaluation of the self-structure scale: a bidimensional model of self-structure

Although psychoanalytic and personality theories have consistently highlighted the central role of self-structure, existing psychometric tools predominantly measure isolated personality traits or narcissistic subtypes, with scarce instruments dedicated to evaluating the organizational mechanisms governing self-related functioning. This study puts forward a bidimensional theoretical framework of self-structure and constructs the Scale of Self-Structure (SOSS), aiming to empirically verify the measurable manifestation of the proposed theoretical dimensions.

Two sequential empirical studies were performed to construct and psychometrically validate the SOSS. Study 1 covered item pool development, exploratory factor analysis (EFA), internal consistency testing, and test-retest reliability verification. A total of 1,954 participants were recruited for initial scale refinement, and an independent subsample of 128 respondents completed retest assessments to evaluate temporal stability. Study 2 recruited 3,805 participants split across two separate cohorts: Sample 3 (n = 2,570) and Sample 4 (n = 1,235). This sample was used for cross-sample structural validation and comprehensive psychometric examinations, including confirmatory factor analysis (CFA), exploratory structural equation modeling (ESEM), internal reliability evaluation, multi-group measurement invariance tests, discriminant validity estimation, and person-centered latent profile clustering. Supplementary narcissism measurement data collected from Sample 4 further supported analyses of convergent and nomological validity.

EFA yielded a 28-item SOSS encompassing four correlated first-order factors: Positive Self, Self-Center, Idealized Identification, and Vulnerable Self. In Study 2, competing measurement models were compared. The four-factor exploratory structural equation modeling (ESEM) solution exhibited the best fit to the data, implying that self-structural processes consist of interrelated yet partially overlapping components, rather than following a strictly hierarchical structure. The SOSS showed acceptable internal consistency and temporal stability, supported measurement invariance across gender, and displayed adequate discriminant validity together with theoretically consistent correlations with grandiose and vulnerable narcissism. Subsequent person-centered analyses tentatively revealed four distinct profile configurations within the two-dimensional structural space defined by the SOSS.

This study provides preliminary empirical evidence supporting the SOSS as a multidimensional instrument for assessing individual differences in self-structural organization. Established based on psychodynamic self-organization theories and validated through systematic psychometric examinations, the SOSS may serve as a tentative framework for understanding self-investment, relational functioning, and self-cohesion beyond traditional trait-based indicators. Given the nonclinical and culturally homogeneous sample of the present study, future investigations incorporating longitudinal designs, demographically diverse cohorts, clinical populations, and cross-cultural validation are required to further elucidate the developmental implications, generalizability, and clinical applicability of the self-structural dimensions captured by the SOSS.