Determination and analysis of intersectional differential itemfunctioning: TIMMS 2019 example

The aim of this study is to determine whether the items in the science subtest at the 8th grade level in the TIMSS 2019 test show an intersectional Differential Item Function (DMF) at the level of intersectional groups obtained using traditional and intersectional groups obtained using these variables according to gender and socioeconomic status. The study group of the research consisted of the 8th grade students in the Turkey sample of the 2019 TIMSS study who took the common science subtest in the first and second booklets. The DIF levels of the items in the TIMSS 2019 test were determined by Mantel-Haenszel (MH) and SIBTEST methods using R software, which includes the difR package. Before the DIF determination analyses, the data were prepared for analysis with SPSS and Excel programs. As a result of the traditional DIF analyses performed in the study; 6 items were flagged as DIF with MH analyses and 3 items were flagged as DIF with SIBTEST analyses out of 26 items in the science subtest on the basis of gender variable. On the basis of socioeconomic level, 1 item was found to contain DIF as a result of MH and SIBTEST analyses. In this study, intersectional groups were formed by using gender and socioeconomic level variables other than traditional approaches. As a result of the intersectional DIF analyses performed on the basis of these groups; 10 DIF were flagged in 7 items in 78 comparisons made with 26 items. In SIBTEST analyses, 8 DIF were flagged in 7 items.