Abstract
The aim of this study was to investigate relationships between children's self-processes--- self-concepts and self-esteem --- on the one hand and their academic achievements on the other. The subjects largely comprised intact school classes and were aged eight, eleven and thirteen years. A battery of instruments, some standardized, some specially developed, was employed to obtain measures of selfesteem, academic self-concept, achievement, intelligence, achievement value, academic locus of control, cognitive complexity, field-independence and popularity. Some of these measures were derived from repertory grids, a new technique being devised for eliciting personal constructs from children. In addition, ratings of academic achievement were obtained from both teachers and classmates. The data revealed an age related trend towards increasing differentiation between general and specific self-processes. Thus, the youngest children evaluated themselves globally whereas the two older age groups made independent evaluations of their academic and social functioning. Moreover, with age, children's academic self-concepts became increasingly accurate, corresponding more closely both with actual attainments and with ratings by others. Self-processes were found to predict academic achievement independently of intelligence. General self-esteem was not directly related to achievement, academic self-concept acting as an intervening variable. A comparison of the different approaches to measuring self-processes served to substantiate the construct validity of the repertory grid measures of academic self-concept and academic self-esteem, but not of the grid measure of general self-esteem. Further research with this approach is indicated. The concepts of general self-esteem, academic self-esteem and academic self-concept are re-evaluated in the light of the above findings. Implications for the measurement of self-processes and for educational practice are discussed.