On the structure of adolescent self-concept.

Abstract

The purposes of the study were (a) to test assumptions underlying the multifaceted, hierarchical structure of self-concept (SC) as hypothesized by Shavelson, Hubner, and Stanton (1976) and (b) to determine whether academic SC can be discriminated from academic grades. In the process, we examined SC interpretations of the Self Description Questionnaire III (SDQ III) and the Affective Perception Inventory (API), two relatively new measures. With 991 (516 young men and 475 young women) 1 lth- and 12th-grade students and multiple measures of general, academic, English, and mathematics SC facets, we found support for a multifaceted, hierarchical interpretation, with one minor exception. A single dimension of academic SC, as well as subject-specific SCs, were discriminable from grades. Although we empirically confirmed the four subscales of the SDQ III, we could not confirm the Student Self subscale of the API. In recent research on self-concept (Byrne, 1986; Fleming & Courtney, 1984; Marsh & Shavelson, 1985; Shavelson & Bolus, 1982), earlier criticisms that most studies (a) lacked a clear theoretical basis, (b) used psychometrically inferior instruments, and (c) used inappropriate methodological procedures (Shavelson, Hubner, & Stanton, 1976; Wylie, 1974) have been addressed. More specifically, researchers have sought to validate both the conceptual structure of selfconcept and interpretations of the instruments designed to measure it (see Byrne, 1984). In broad terms, the purpose of the present study was to provide new evidence bearing on the properties of self-concept (SC) structure.