Effectiveness of Multiple Indicators and Multiple Causes Model (MIMIC) in Detecting the Differential Item Functioning of a Test in Terms of Sample Size and Test Length
Reem Elyan, Yousef Sawalmeh · Jordan Journal of Applied Science-Humanities Series · 2024
This study aimed to examine the effectiveness of the Multiple Indicators Multiple Causes Model (MIMIC) in detecting differential item functioning of a test in relation to sample size and test length. Three random samples were used in the study, consisting of 10th-grade students in Jordan who took the PISA 2018 mathematics test, with varying sizes of 2000, 3000, and 4000 students. Additionally, three test lengths were used: 10 items, 20 items, and 30 items. The Mplus program was utilized for data analysis according to the MIMIC method to detect differential item functioning for gender. The results indicated that there was no statistically significant difference (α=0.05) in the percentage of items with differential item functioning detected due to different sample sizes across the different test lengths. Similarly, there was no statistically significant difference (α=0.05) in the percentage of items with differential item functioning detected due to different test lengths across the different sample sizes.