TIMSS is an international study which aims to follow the changes and developments in student achievement and identify the differences between the education systems of the participating countries. In this study, it is aimed to test whether the variables which are effective in mathematics achievement of fourth grade students provide the measurement invariance conditions or not. It is a correlational study and its sample consists of 6456 fourth grade students participated in TIMSS 2015 from Turkey. The Multigroup Confirmatory Factor Analysis (MG-CFA) was used for testing the measurement invariance of the factors related to the student achievement. The configural, metric, scalar and strict invariance tests in terms of genders were done incrementally for liking mathematics, interest in mathematics and confidence in mathematics latent variables in the research. CFI, comparative fit indices, differences between the unconstrained condition (configural invariance) and the conditions in which more restrictions were added incrementally were assessed in the analyses. As a result, it was seen that the three of the latent variables tested provided the configural invariance condition, however only liking mathematics latent variable provided the strict invariance condition in gender groups.
Primary Language | Turkish |
---|---|
Subjects | Studies on Education |
Journal Section | Articles |
Authors | |
Publication Date | December 28, 2018 |
Submission Date | April 4, 2018 |
Published in Issue | Year 2018 Volume: UBEK-2018 |