This study investigated the structure of social self-concept in preadolescent children at school. Subjects were 227 male and female grade three students from eight schools in a metropolitan area. A multidimensional, hierarchical model of social self-concept was proposed and tested using confirmatory factor analytic procedures within the framework of the analysis of covariance structures. The hypothesized model was based upon the theoretical model of self-concept first proposed by Shavelson, Hubner, and Stanton (1976). Subjects were given two self-report instruments designed to measure self-concept in the areas of social relations with classmates, other children at school, and teachers. Additionally, a teacher rating scale and a peer sociometric instrument were used as measures of the child's actual social behaviour. Data were collected at two time points, in the fall and the spring of the same academic year. The hypothesized and counterhypothesized models provided a poor fit to the data. Analyses then continued in an exploratory mode. Of all models tested, the best-fitting model of social self-concept was found to be a four-factor model with general self-concept as the first factor, self-concept of popularity as the second factor, social self-concept in general and in regard to school and classmates as the third factor, and social self-concept in regard to teachers as the fourth factor. Because the facets of social self-concept were found to be less differentiated than hypothesized it was not possible to investigate the proposed hierarchical structure of the construct. Partial support was found for the hypothesis that social self-concept can be differentiated from social behaviour as measured by peers and teachers. Social behaviour with teachers, in particular, was found to be very different from self-reports of social behaviour. Social behaviour with peers and self-reports of social behaviour with peers were found to be associated moderately, indicating that self and other ratings were referring to the same, or similar, behaviours, but each from their own unique perspective. Analysis of the second time point data did not support the stability of social self-concept over a six month period, indicating the necessity of replicating these results before conclusions can be more than tentative. This study presented an important possible revision to the Shavelson et al. (1976) model of self-concept. The study demonstrated the salience of popularity in the child's conception of his or her own social relations. It demonstrated, further, that the child's perspective regarding social relations differs from that of peers or of teachers.
| Identifer | oai:union.ndltd.org:uottawa.ca/oai:ruor.uottawa.ca:10393/10003 |
| Date | January 1995 |
| Creators | Wilson, Patricia. |
| Contributors | Byrne, Barbara, |
| Publisher | University of Ottawa (Canada) |
| Source Sets | Université d’Ottawa |
| Detected Language | English |
| Type | Thesis |
| Format | 225 p. |
Page generated in 0.0149 seconds