The Self Description Questionnaire III: Testing for Equivalent Factorial Validity across Ability
- 1 April 1988
- journal article
- Published by SAGE Publications in Educational and Psychological Measurement
- Vol. 48 (2) , 397-406
- https://doi.org/10.1177/0013164488482012
Abstract
Confirmatory factor analysis (CFA; LISREL) was used to (a) validate the Self Description Questionnaire III (SDQ III) subscales measuring general, school, English, and mathematics self-concepts for a sample of 898 (285 low track, 613 high track) grades 11 and 12 students, and (b) test the equivalency of the factor structure across academic track. The results confirmed a 4-factor structure and revealed all factor loadings and covariances to be invariant across groups; 11 of 21 uniquenesses were noninvariant. Although, in a strict statistical sense, the differentially reliable items bore implications of bias in favor of the high track, these discrepancies, on the basis of absolute values, were judged to be of little practical significance. Overall, in light of the stringency of LISREL CFA procedures in general, and those used in this study in particular, the SDQ III demonstrated exceptionally superior psychometric properties.Keywords
This publication has 19 references indexed in Scilit:
- Perceived Competence Scale for Children: Testing for Factorial Validity and Invariance Across Age and AbilityApplied Measurement in Education, 1988
- Adolescent Self-Concept: Testing the Assumption of Equivalent Structure Across GenderAmerican Educational Research Journal, 1987
- Detecting Item BIAS in Affective ScalesEducational and Psychological Measurement, 1987
- On the structure of adolescent self-concept.Journal of Educational Psychology, 1986
- Multitrait-Multimethod Comparisons Across Populations: A Confirmatory Factor Analytic ApproachMultivariate Behavioral Research, 1985
- On the Meaning of within-Factor Correlated Measurement ErrorsJournal of Consumer Research, 1984
- Confirmatory Factor AnalysisPublished by SAGE Publications ,1983
- Significance tests and goodness of fit in the analysis of covariance structures.Psychological Bulletin, 1980
- What Does it Mean to Say a Test is Biased?Education and Urban Society, 1975
- Simultaneous factor analysis in several populationsPsychometrika, 1971