Abstract
A structural equation model was developed to examine the relationships between the self-esteem and defensiveness scales of the Coopersmith Self-esteem Inventory. Data were analysed for a sample of over 800 New Zealand children using methods of confirmatory factor analysis. The fitted model suggested that the data were consistent with an hierarchical model in which the observed self-esteem scores were fallible indicators of self-esteem in four specific areas of experience. These areas in turn were subsumed by a higher-order factor of global self-esteem. The model also suggested that defensive responding systematically contaminated the subjects' over-all reporting of self-esteem. Model estimates showed that between 28% and 54% of the variance in the self-esteem indicators arose from random errors of measurement and a further 1% to 3% was due to the effects of defensiveness. It is suggested that confirmatory factor analytic methods should be used in further analyses of the relationship between self-esteem and other measures to take account of possible biases arising from these sources.
Get full access to this article
View all access options for this article.
