Self-Concept: The Interplay of Theory and Methods.

Abstract : The purposes of this paper were to advance self-concept theory and to present recent methodological advances for doing so. With respect to methodology, the analyses of covariance structure, compared to simply an analysis of correlations, enabled us to test competing models and to understand the origin of the observed correlations. The conclusions that would have been drawn from the correlational data--e.g., multifaceted, hierarchical structure with increasing stability of constructs toward the apex--were modified and clarified on the basis of the analysis of covariance structure--e.g., lack of support for increasing stability. The covariance structure analytical technique also permitted us to test casual relations between latent constructs rather than between observed variables. Casual relationships among constructs, of course, cannot be tested on the basis of zero-order correlations. Clearly the covariance structure technique is a major methodological contribution to the development and testing of psychological theory in education. With respect to self-concept theory, the following conclusions seem warranted on the basis of our sample of 99 middle-class, junior high students and the literature reviewed. Self-concept is a multi-faceted construct. General self-concept can be interpreted as distinct but correlated with academic self-concept. Furthermore, subject-matter specific facets of self-concept can be interpreted as distinct, but correlated with one another and with academic and general self-concept.