Construct Interpretations of Three Self-Concept Inventories

Three self-report measures of self-concept were administered in counterbalanced order to 488 sixth grade children as the basis for a multitrait-multimethod study of self-concept measurement. The subscales on these instruments were clustered into facets representing physical, social, and academic self-concept. The analyses indicate that all three facets demonstrate some degree of convergent validity, particularly physical self-concept, but discriminant validity is not established. Results also indicate that the response requirements of the instruments affect construct interpretations. Although developers of self-concept inventories strongly imply the existence of significantly different facets of self-concept by the separation and naming of subscales, there is little evidence for this implication.