A validation study of selected self-concept measures for children

No Thumbnail Available
Authors
Treloar, Judith
Advisor
Joy, Kenneth L.
Issue Date
1981
Keyword
Degree
Thesis (D. Ed.)
Department
Department of Educational Psychology
Other Identifiers
Abstract

Although self-concept has been a prominent and highly utilized hypothetical construct in psychological research for many years, studies attempting to establish the validity of self-concept instruments are relatively few in number. The purpose of this study was to determine the cross-structural validity of selected self-concept measures -- to determine if self-reports and behavioral ratings of self-concept measure the same construct.There were 202 subjects who participated in the study. They were selected from a pool of fourth-, fifth-, and sixth grade regular education students whose parents had signed permission to participate. One male and one female were selected from each classroom. Certified psychometrists administered two self-concept instruments and one anxiety measure to the students in their regularly assigned buildings, Order of test administration was counterbalanced among buildings, Each teacher completed two behavioral ratings of self-concept and one behavioral rating of anxiety on the male and female selected from his or her classroom, Order of completion was counterbalanced among teachers.A multitrait-multimethod analysis was conducted to determine cross-structural validity. Four different correlations were analyzed: (1) monotrait-monomethod, (2) heteromethod-monotrait, (3) heterotrait-monomethod, and (4) heterotrait-heteromethod. The .01 level was used to establish statistical significance of the correlations.Results of this study demonstrated that there is little cross-structural validity between the self-report measures and behavioral rating forms used:1. The common factor variance was not sufficient enough to consider the methods comparable in spite of the fact that there was a statistically significant relationship between self-report measures and behavioral ratings of self-concept.2. Method variance contributed to the substantial concurrent validity between the Children's Self-Concept Scale and the Self-Esteem Inventory. Similarly, method variance contributed to the substantial concurrent validity between the Inferred Self-Concept Scale and the Behavior Rating Form.3, High negative correlations between self-concept and anxiety within each method suggested that self-concept and anxiety share enough common method variance to disallow the use of anxiety as a totally distinct trait.4, Self-reports and behavioral ratings may each measure a different aspect of self-concept -- the experienced and the presented. Therefore, a combination of the two methods might yield a more global assessment of self-concept.