MAB에 대한 중다집단 확인적 요인 분석

다국어 초록 (Multilingual Abstract)

This study focuses on the examination of the construct validity of the Multidimensional Aptitude Battery(MAB). The purpose of this study is to factor analyze the data obtained from the 1990 norming study using the maximum likelihood confirmatory factor analysis in order to investigate the extent to which the proposed two factor model explains the variation in the MAB subtests. The total sample size is 1400, approximately evenly divided by gender.
For this study the two factor (Verbal factor and Performance factor) model was specified. All hypotheses were tested by using the covariance matrices from the 10 MAB subtests scores for each age group. LISREL 7 was used for the data analysis. The X² tests of fits of the proposed two factor model, single factor model, and null model to the observed data were conducted. X² difference test, a ratio of X² to df, NFI were used for additional goodness-of-fit index considering the possible impact of sample size on X² test, and X² difference test
The proposed two factor model well explained the subtest variation of the MAB. The two factor model and single factor model better explained the data than the null model with no common variances. From the results of this research, it is implied that if used for group testings, the MAB can be an alternative for the WAIS-R.