Analytical Comparison of Item Statistics Employing IRT 3PLM in WASSCE and NECO SSCE 2023 English Language Multiple Choice Tests

Abstract

This assessment encompassed the analytical comparison of item statistics employing the Item Response Theory Three Parameter Logistic Model (IRT 3PLM) in WASSCE and NECO SSCE 2023 English Language multiple choice tests. Three hypothesised questions were articulated. In this study, a survey research design was utilised. This investigation’s population comprised 72,400 English Language SS 3 students in 384 public senior secondary schools in Ogun State. A multistage sampling procedure was applied to pick a sample size of 8,000 English Language SS 3 students. The assessment instruments used were WASSCE and NECO SSCE 2023 English Language multiple choice tests. The data collected were subjected to IRT 3PLM using jMetrik psychometric software to generate the item parameters. All hypothesised questions were tested for significance utilising independent samples t-test statistics. This assessment’s outcomes divulged a significant dissimilarity in the guessing parameter between WASSCE and NECO SSCE 2023 English Language multiple-choice items. However, no significant dissimilarities existed in the discrimination and difficulty parameters between the two. The pragmatic extrapolation of this analytical comparison accentuated in the findings alluded that employing IRT 3PLM to establish the item statistics before administering the tests could further enhance the quality of English Language multiple choice items developed by WAEC and NECO.