A correlational study of self-concept assessments in four and five year olds

Includes bibliographical references (pages 44-46)The purpose of this study was to investigate the\ud relationship among five measures designed to assess the\ud self-concepts of preschool children. The subjects were\ud 40 four-and-five year old children living and attending\ud preschool or child care programs in the San Fernando Valley\ud of the greater Los Angeles area. The study included\ud 21 boys and 19 girls. The mean age of the children was\ud four years, five months.\ud The research was based on a modified multi-trait,\ud multi-variable research design. Three of the five\ud assessments of self-concept were considered to be direct\ud methods: those given directly to the child. Two inferred\ud assessments of the self-concept were utilized: one \ud administered to the child's teacher, the other to the\ud child's parent. These assessments resulted in five scores\ud for each child.\ud A reliability study analysis was completed on the\ud internal consistency assessments which tested for internal\ud consistency using Cronbach's coefficient alpha. The\ud analysis affirmed that the assessments used in this study\ud all had internal consistency. Alpha scores were .67 for the\ud Joseph Preschool Primary Self-Concept Test; .85 for the\ud Preschool Self-Concept Picture Test; .93 for the Behavioral\ud Academic Self-Esteem; .93 Purdue Self-Concept Scale. The\ud JPPSST Inferred Questionnaire had only one question and\ud therefore could not be tested for reliability of the\ud internal consistency variety.\ud The data indicated that no significant correlation was\ud found between the inferred assessments and the direct\ud assessments. However, the three direct tests correlated\ud with one another. For example, the PSCS and the JPPSST\ud correlated at P .001. In addition, the two inferred tests\ud correlated with each other at P .048 level