Kennedy IMASUEN 1 and Iyabo Ayodeji UKANAH 2
1 Institute of Education, University of Benin, Nigeria
2 Department of Guidance and Counselling, Faculty of Education, Adekunle Ajasin University, Nigeria
Abstract
This study compared the Bootstrap, Jackknife, and Item Response Theory (IRT) methods in estimating item parameters of the 2022 National Business and Technical Examinations Board (NABTEB) Mathematics multiple-choice test. A survey research design utilizing an ex-post facto approach was adopted. The target population comprised of students who took the 2022 NABTEB Mathematics test, with 5,500 students selected from the South-South region of Nigeria using a stratified random sampling technique. The NABTEB Mathematics multiple-choice questions served as the study instrument, with its validity established by NABTEB. The test’s internal consistency was assessed using the Kuder-Richardson 20 (KR-20) formula, yielding a reliability coefficient of 0.82. Data analysis was conducted using R statistical software, applying the four-parameter logistic model (4PLM) through direct estimation (IRT), Bootstrap, and Jackknife resampling methods. Bias was evaluated by comparing parameter estimates and constructing 95% confidence intervals. Findings revealed that Jackknife estimates closely aligned with IRT, demonstrating greater stability and reliability, while Bootstrap slightly overestimated difficulty and discrimination parameters. Both resampling methods provided consistent estimates for guessing and upper asymptote parameters. Overall, Jackknife proved to be the more precise and less biased method, reinforcing its suitability for standardized testing applications. It was recommended that the Jackknife method should be prioritized in large-scale educational assessments to enhance precision and reduce bias. Additionally, integrating IRT with resampling techniques can improve test development by optimizing item difficulty and discrimination, ensuring more accurate and fair student assessments.
Keywords: Item Response Theory, Bootstrap, Jackknife, Item Parameter Estimation, Educational Assessment, Measurement and Evaluation
How to Cite this Article:
Imasuen, K. & Ukanah, I. A. (2025). Comparison of the Bootstrap, Jackknife and Item Response Theory Methods in Estimating Item Parameter of 2022 NATEB Mathematics Examination. VillageMath Educational Review (VER), 7(1), 74-83. https://doi.org/10.5281/zenodo.15745234
