Construction and standardization of An Academic Self-Concept scale

Abstract

This paper presents the findings of the study conducted to construct and validate an Academic Self-Concept Scale for Sri Lankan Junior Secondary level school children. The Academic Self-Concept Scale is a multidimensional self-report Likert type scale which was constructed and validated for the purpose of assessing the academic self-concept of Junior Secondary level school children in Sri Lanka. The construct of academic self-concept is defined as the perceptions and evaluations, one holds true about his or her academic abilities in different school subject areas and in general. This scale can be used for research purposes as well as for assessing the academic self-concept of students for counseling purposes. Items for the scale were adapted from the academic self-description questionnaire (Marsh 1999) and the format and procedures for administration and scoring were adapted from the self-concept scale of Chandradasa (1996). The scale contains five sub scales which were designed to measure academic self-concept of students relevant to specific subject areas of Mathematics, Science and Technology, First Language, Second Language and school subjects in general. Each sub scale has six items and each item comprises of a statement that express one’s perceptions and evaluations regarding one’s academic abilities. Most of these statements are positive while few of the statements are negative. The phrasing of the six statements for each subscale is strictly the same with only the subject being changed. The common phrasing for six statements are as follows. Work in (mathematics) is easy for me I am good at (mathematics) I have always done well in (mathematics) I am hopeless when it comes to (mathematics) I learn things quickly in (mathematics) I get good marks in (mathematics) classes. Students are asked to respond to teach statement by choosing one of the choices: true; mostly true; sometimes true and sometimes false; mostly false, false. Marks are given respectively for the responses as 5, 4, 3, 2, 1 for positive statements and in reverse order for the negative statements.Test-retest reliability of the scale was assessed using a sample of 50 students and the r value is 73. Factor analysis was conducted to assess the factorial validity and it was found that each subscale measures a different factor. Cronbach α was calculated for each subscales and the values were between .63 and .84. Data gave evidence for the construct validity and the reliability of the scale