ITEM ANALYSIS OF THE 2019 ENGLISH LANGUAGE MOCK EXAMINATION: IMPLICATION FOR TEST DEVELOPMENT IN TARABA STATE, NIGERIA

This study examined the item analysis of the 2019 English Language Mock Examination conducted in Taraba State, Nigeria. Two objectives were raised to guide the study, which addressed two research questions and tested two null hypotheses at 0.05 level of significance. A Multi-stage sampling procedure involving simple random and stratified sampling techniques was used to select 600 marked scripts from a population of 11,268 scripts of those student who sat for the 2019 English Language Mock Examination. Descriptive statistics, specifically percentages were employed to answer research questions, while an independent samples t-test was used to test the hypotheses. The results indicated that the examination items generally demonstrated acceptable technical quality in terms of difficulty, with the majority falling within the moderate difficulty range, despite this balance. The test showed weak discriminatory power, as 74% of the items were classified as poor or marginal in their discrimination indices. This suggest that the examination was limited in the ability to effectively differentiate between high- and-low performing students. The finding also revealed that there were no significance differences in the mean difficulty levels and discrimination power based on gender. Based on the findings, the study recommends that tests items with difficulty indices outside the acceptable range of 40-60% should be reviewed and refined for future use. Furthermore, the Education Resource Centre should strictly adhere to established item writing guidelines to enhance the quality of the items for future mock examinations.