Examining the Construct Comparability of the English and French Versions of TIMSS

Abstract

The objective of this research was to examine the comparability of constructs assessed by English and French versions of the Third International Mathematics and Science Study (TIMSS). The differences in constructs observed in our analyses indicate serious limitations of using TIMSS results for making comparisons that use overall performance in mathematics and science. In particular, large differences between constructs were observed between the U.S. and French scales. Such limitations include rank ordering of performance of countries such as the United States and France, as well as conducting research using TIMSS data to compare factors associated with performance. The results from this study point to differences in constructs assessed by TIMSS in different countries and the importance of empirical evidence to support construct comparability before TIMSS results can be meaningfully used for research.