Investigation of measurement invariance according to home resources

Periodical
International Journal of Assessment Tools in Education
Volume
8
Year
2021
Issue number
3
Page range
633-648
Relates to study/studies
TIMSS 2015

Investigation of measurement invariance according to home resources

TIMSS 2015 mathematical affective characteristics questionnaire

Abstract

This study aimed to examine the measurement invariance of the mathematical affective characteristics model obtained from TIMSS 2015 4th grade Turkey administration according to home resources. For this purpose, firstly, the factor structure of the mathematical affective characteristics questionnaire was examined by explanatory factor analysis and Velicer's maximum average partial (MAP) test. It was revealed that the questionnaire had three factors. Then the structure was validated by confirmatory factor analysis. In the next stage, multi-group confirmatory factor analysis was employed with a purpose to examine whether the model displayed measurement invariance across the variables of home resources such as internet connection, heating system, cooling system, and dishwasher. The results showed that the strict measurement invariance of the mathematical affective characteristics model was achieved among the subgroups of each of the internet connection, heating system, cooling system, and dishwasher variables. Accordingly, means, variance, covariances, and item residual variances in the subgroups were found to be similar. According to the results of the study, the comparison of the mathematical affective characteristics model based on the home resources is found to be significant and comparisons made show that possible differences arise from the relevant home resource.