The Factorial Validity of Three Forms of the Dimensions of Self-Concept Measure

Abstract

For each of three samples of 145 elementary school (ES), 250 junior high school (JHS), and 166 senior high school (SHS) students who were administered one of three essentially parallel forms of a self-report inventory entitled Dimensions of Self-Concept (DOSC), efforts were made (1) to determine the empirical factor structure of each form, (2) to ascertain whether empirical (rotated) factors as interpreted corresponded to the five school-related a priori dimensions of the Aspiration, Anxiety, Academic Interest and Satisfaction, Leadership and Initiative, and Identification versus Alienation scales of the DOSC, and (3) to identify the extent of invariance in the empirical factor structure of the three inventory forms across the three samples. For the ES, JHS, and SHS samples, respectively, the orthogonal solution provided 2, 3, and 3 interpretable unidimensional factors, whereas the oblique solution yielded correspondingly 1, 4, and 3 identifiable unidimensional factors. Only a modest degree of congruence occurred between each of the five a priori or logically derived factor dimensions of the inventory and the identifiable empirical (rotated) factors. Only the single factor dimension of Anxiety was invariant across all three samples in both the orthogonal and oblique factor solutions. Anxiety appears to be a construct underlying self-concept that is associated with school learning at all grade levels studied.