Analysis of Psychometric Qualities of National Examinations Council (NECO) Mathematics Essay Test Using Generalized Partial Credit Model

Abstract

The main purpose of this study therefore was to analyze the psychometric qualities of NECO mathematics essay test using generalized partial credit model. The study adopted an instrumentation research design and was carried out in Benue state of Nigeria. The population of the study was 41,836 SS 3 students who registered for June/July NECO SSCE Examination in Benue State of Nigeria and a sample size of 650 students were used. The instrument for this study was an adopted NECO 2016 and 2017 June/July mathematics essay questions. The instrument for the study was administered to the respondents by the mathematics teachers in the sampled schools under the supervision of the researcher. Data was analysed using item parameter estimates of R-Package to answer the research questions, while t-test statistic was used to test the null hypotheses at 0.05 level of significant. The result of the study showed that NECO 2016 and 2017 mathematics essay questions were unidimensional in nature. NECO 2016 had lower step difficulties than NECO 2017. NECO mathematics essay test for 2016 discriminated between high and low ability students than the NECO mathematics essay test for 2017. Based on the findings of the study, it was recommended among others that test developers and examination bodies such as NECO should undergo rigorous processes of psychometric properties of essay test before and after administration of test instrument. This will help the test developers to determine the existence of these psychometric properties and the need to establish them in any given test. It was also recommended that Seminars and workshops on Item Response Theory should be organized by the examination bodies for Psychometricians, test developers and teachers involved in test development.