In recent decades, the use of large-scale standardized international assessments has increased drastically as a way to evaluate and compare the quality of education across countries. In order to make valid international comparisons, the primary requirement is to ensure the measurement equivalence between the different language versions of these…

Literature in the United States provides many examples of no difference in student achievement when measured against the mode of test administration i.e., paper-pencil and online versions of the test. However, most of these researches centre on "regular" students who do not require differential teaching methods or different evaluation…

As the largest international study ever taken in history, the Trend in Mathematics and Science Study (TIMSS) has been held as a benchmark to measure U.S. student performance in the global context. In-depth analyses of the TIMSS project are conducted in this study to examine key issues of the comparative investigation: (1) item flaws in mathematics…

Intellectual abilities were measured for 217 German and 205 American college students using tests (in the subjects' native languages) in which equivalence was established by an item-response theory-based differential-item-functioning (DIF) analysis. Comparisons between groups were not the same before and after removal of DIF items. (SLD)

The relationship between gender and performance on mathematics items varying in level of cognitive complexity and content was assessed, using 1,789 female and 1,951 male Thai adolescents and 2,040 female and 1,884 male American adolescents. Data suggest that performance relative to both cognitive complexity and content is related to gender. (TJH)

Describes a method for assessing the quality of translations based on item response theory (IRT). Results from the IRT technique with French and Chinese versions of a scale measuring individualism-collectivism for samples of 250 U.S., 357 French, and 290 Chinese undergraduates show how several biased items are detected. (SLD)

Evaluates the measurement equivalence of an English-language version of the Trier Personality Inventory, using statistical methods based on item response theory to identify items displaying differential item functioning (DIF). Results with 295 U.S. and 213 West German undergraduates and 203 U.S. college students indicate significant agreement in…