ABSTRACT

This study analysed the psychometric properties of Basic Science multiple choice test of National Examination Council and State Basic Education Certificate Examination (2016-2017) in Abia State. Six research questions were raised and six null hypotheses were formulated for the study. It adopted instrumentation research design. The population of the study was made up of 300 examination questions drawn from NECO BECE and State BECE question papers 2016 - 2017. Out of 30,160 JSS III Basic science students in Abia State, a total of 248 students in 2016-2017 academic sessions served as respondents to the questions that were sampled and used for the study. Two instruments used for data collection were NECO Basic Science multiple choice test and State BECE Basic Science multiple choice test used during 2016-2017 examinations.  The instruments were re-validated by three experts, two from Measurement and Evaluation and one from Biology Education; from Michael Okpara University of Agriculture, Umudike. The establishment of the reliability coefficient of NECOBECE and State BECE Basic science multiple choice test items were carried out using Kuder Richardson Formular 20. The data obtained was analysed using Kuder Richardson method which yielded a correlation coefficient of 0.75 and 0.76 respectively. The researcher sought for permission from the principals of the selected schools to carry out the research.  The instruments were administered to the students with the help of two research assistants. Data obtained for the study were analyzed using Scheuneman modified chi-square and logistic regression statistics for the research questions while t test of independence was used to test the null hypotheses at 0.05 levels of significance. The results of the data analyzed revealed that the reliability indices for the items used by the two bodies were equal in 2016 and considered high. There was no significant difference between the mean values of Standard Error of Measurement of Basic Science multiple choice items administered by NECOBECE and State BECE in 2016. There was no significant difference between the mean values of Standard Error of Measurement of Basic Science multiple choice items administered by NECOBECE and State BECE in 2017. Based on these findings, it was recommended that students should be allowed to seat for any of the tests by the examination bodies, since the quality of the tests were the same. Also, the examination bodies are encouraged to make future reference of the items since the items examined have satisfactory psychometric properties