Differential item functioning analysis of senior school mathematics multiple-choice items in Nigeria’s national examinations council

Abstract

Concerns about persistent underperformance in the National Examinations Council (NECO) Senior School Certificate Examination (SSCE) Mathematics in Nigeria have raised questions about the fairness of multiple-choice test items across different student groups. This study examined the Differential Item Functioning (DIF) of the 2024 NECO SSCE Mathematics multiple-choice items to determine whether certain items favored particular categories of students. Specifically, the study investigated DIF across gender, school type (public and private), and school location (urban and rural). A descriptive survey design was adopted, involving 294 Senior Secondary School students drawn from twelve schools in Ilorin, Kwara State, Nigeria. Stratified sampling was used to ensure fair representation of the identified subgroups. The instrument consisted of 60 standardized multiple-choice Mathematics items from the June/July 2024 NECO SSCE. Responses were analyzed using logistic regression to detect DIF and determine the direction and magnitude of item bias. The findings showed that 56 items exhibited negligible gender DIF, while four items (4, 10, 55, and 57) displayed significant DIF. Items 4, 10, and 57 favored male students, whereas Item 55 favored female students. School type produced the largest DIF effect, with 19 items favoring students from private schools. Additionally, five items favored students from urban schools. Overall, 28 items demonstrated moderate to large DIF, indicating the presence of item characteristics that may threaten test fairness and validity. The study recommends revising biased items, developing context-sensitive questions, and ensuring that Mathematics assessments reflect diverse learning environments to promote fairness and equity in high-stakes examinations.