Confusion exists with regard to the empirical and substantive link between self-concept and self-esteem in elementary school children and their relationship to self-description, self-evaluation, and global beliefs and feelings about oneself as a person. This study reports the results of investigating the relationships between these self-constructs using 957 elementary school children in Grades 3 to 7. The evidence suggests that self-concept is comprised of both descriptive and evaluative beliefs that children hold about certain characteristics, whereas self-esteem can be viewed as the global feelings and beliefs that children have about themselves as people.
Even though the terms self-concept and self-esteem appear frequently in the personality, educational, and psychological literature, thorough reviews have highlighted the lack of (a) definitional specificity regarding the constructs and (b) a clear delineation of the substantive link between the two self terms (Hattie, 1992; Hughes, 1984; Wylie, 1979). The seminal work of Shavelson, Hubner, and Stanton (1976) and Shavelson and Bolus (1982) resulted in the development of a model that emphasized the multidimensional, hierarchical structure of self-concept. However, the model did not explicitly address the issue of delineating self-esteem from self-concept. What Shavelson and colleagues did address was the issue of the relationship between self-description and self-evaluation, noting that the distinction between the two had not been clarified either conceptually or empirically. Consequently, the position that self-description and selfevaluation are empirically the same has been outlined (Marsh & Shavelson, 1985; Shavelson & Bolus, 1982; Shavelson et al., 1976).
Much of the literature that distinguishes between the two constructs equates selfconcept with self-description and self-esteem with self-evaluation. Watkins and Dhawan (1989) distinguished the two constructs theoretically on the basis of this conceptualization. They noted that the items in the Self-Description Questionnaire 1 (SDQ1; Marsh, 1990). which was based on Shavelson et al.'s (1976) model, had self-descriptive items (I enjoy sports and games) and self-evaluative items (I am good at sports) and therefore measured both self-concept and self-esteem. Watkins and Dhawan (1989) argued that self-concept and self-esteem can be distinguished and cited three studies whose findings were suggestive of an empirical distinction between self-description (self-concept) and self-evaluation (self-esteem). Of specific interest was the Bogan (1988) study of 162 college students, which found that 43% of subjects expressed significant differences between corresponding descriptive and evaluative judgments across the six dimensions of self.
In summary, Shavelson et al. (1976) maintained that descriptive and evaluative statements about one's characteristics are indistinguishable and empirically related, whereas Watkins and Dhawan (1989) argued that descriptive and evaluative perspectives (self-concept and self-esteem) can be differentiated, particularly from a cross-cultural perspective.