Verification of Dichotomous Scored Junior Secondary Schools Certificate Examination in Mathematics Achievement Test Using Classical Test Theory

Abstract

In educational assessment there are two major frameworks by which items contained in a test can be examined. These are classical test theory and item response theory. In this study, the author prove the applicability of the classical test theory to the assessment of the quality of multiple- choice junior secondary mathematics test items. Fifty multiple –choice general mathematics test items were administered to four hundred (400) students randomly selected from twenty (20) Junior Secondary schools three (JSS III) students. Ten (10) schools each were purposively selected from private and public schools in Ado local government area of Ekiti state, Nigeria. Three general questions were generated and answered, while two hypotheses were formulated and tested using t-test statistic at 0.05 level significant. Results showed that, out of the 50 items in the test on the basis of difficulty parameter, five (5) items (10%) failed to meet the set criteria for item qualify while on the basis of item discriminating index five items were of poor quality. However, 45 items based on the established standards were considered ‘good’ items. Finally on the basic of performance, there was no significant differences between students in private and public schools, and also between male and female students from the twenty schools considered for this study which led to the upholding of the two hypotheses formulated for the study at p < 0.0.5. It is recommended that, general mathematics papers, that are to be used to examine students’ learning outcomes in Junior School Certificate Examination should be properly validated.