Abstract
This study examined the degree of comparability between two versions of the Trends in International Mathematics and Science Study’s 1999 mathematics test results from the United States of America and Turkey. Measurement invariance was assessed between the two language versions of tests using differential item functioning analyses and exploratory factor analyses. The impact of the differences on the score scale comparability was also examined by comparing the test characteristic curves. Approximately 23% of the items were identified as differentially functioning for the two countries. The factor analyses indicated differences in the structure of the two tests. However, the effect of these differences on score scale comparability was minimal as was demonstrated by very similar test characteristic curves for the two language versions.