Abstract:
The study examined the application of differential item functioning (DIF) procedure in analyzing Economics students’ achievement in Senior Secondary School Certificate Examination in Nsukka education zone of Enugu state. Two research questions were posed to guide the conduct of the study. Causal comparative or ex-post facto research design was used for the study. The population for the study comprised 2563 SS3 Economics students comprising 1422 males and 1141 females in the 46 public co-educational senior secondary schools in Nsukka Education Zone of Enugu State. The sample size of 444 Economics students was used for the study. The instrument for the study was 2018 SSCE 50-item multiple choice Economics questions developed by West African Examinations Council (WAEC). To determine the reliability of the instrument, the instrument was administered to 25 Economics students in Nsukka Education Zone of Enugu State who were not part of the study. The responses obtained from the students were subjected to Kudder-Richardson 20 (KR-20) formula to determine the internal consistency of the instrument. A reliability coefficient of 0.89 was obtained. The data collected from the study were analyzed using Logistic Regression procedure. The findings revealed that out of 50 items in 2018 WAEC SSCE Economics questions, 14 items which represents 28% of the test items that displayed significant gender DIF at 0.05 level of significance. Among the 14 items displayed significant gender DIF, only 1 item representing 2% was identified to exhibit significant gender DIF against male students. Out of 50 items in 2018 WAEC SSCE Economics questions, 31 items which represent 62% of the test items displayed significant location DIF at 0.05 level of significance. Among the 31 items, 27 items representing 54% were identified to exhibit significant location DIF against urban students. Based on the findings of the study, it was recommended that DIF analysis should be conducted by test experts and test developers on their test items. This will help to reveal items that exhibit Differential item functioning which could be revised or eliminated in order to produce items that do not exhibit DIF only.
