Comparability of teachers' educational background items in TIMSS: a case from Turkey.

Large-scale international assessment studies such as Trends in International Mathematics and Science Study (TIMSS) or Programme for International Student Assessment (PISA) provide researchers and policy makers the opportunity to conduct secondary analyses to answer questions related to educational outcomes and compare the impact of certain inputs on student outcomes across countries. These comparisons are made under the assumption that the questionnaire items translated to different languages are understood in the same way by its participants. Presenting a case from Turkey, this paper shows that equivalency of questionnaire items is not always achieved. The case explores demographic information related to teacher preparation and the sample is drawn from eighth grade science and mathematics teachers participated in TIMSS 2007, 2011, and 2015 in Turkey. Descriptive analysis of data collected from these teachers and comparisons across subjects and years show that teachers may have misunderstood a question regarding their major, thus limiting potential claims related to teacher preparation in Turkey. Researchers and policy analyst who use secondary data collected by international assessment studies should be aware of such comparability issues in adapted items prior to conducting any secondary analyses. [ABSTRACT FROM AUTHOR]