ABSTRACT

The study examined the differential item functioning of National Examination Council (NECO) and National Business-Technical Examinations Board (NABTEB) Agricultural Science multiple choice test items (2015-2017) in South East Zone. The study adopted Descriptive Survey research design and Ex-Post-Facto research design. Eight research questions and eleven null hypotheses were raised and formulated respectively that guided the study. The population of the study is 10, 194 Agricultural Science students which includes 5043 males and 5151 females and also 3007 private and 7187 public secondary schools Agricultural Science students. Based on Krejcie and Morgan's (1971) formula for determining sample size from a known population, a total sample size of 728 Agricultural Science Senior Secondary Sc1ool students (SSS III) were proportionately drawn from the 72 selected secondary schools that constituted the sample size of the study. This consists of 325 private secondary school Agricultural Science students and 403 public secondary schools Agricultural Science students from Abia, Enugu and Imo state. A multi-stage sampling technique was used for the study. The two instruments used for data collection were NECO Agricultural Science paper III and NABTEB Agricultural Science paper 1 Section A used in 2015, 2016 and 2017 examinations. The second instrument was a socio-economic status questionnaire constructed by Adelusi (1982) modified by Odili (2014) and adapted for the study. The instruments were re-validated by three experts, two from Measurement and Evaluation and one from the Department of Agricultural Extension Education; all from Michael Okpara University of Agriculture, Umudike. The instruments were administered once in the area that was not part of the sample. The data obtained was analysed using Kuder Richardson Formular 20 with reliability indices of 0.76 and 0.77 respectively. The reliability of questionnaire was determined using Cronbach Alpha Coefficient with a reliability index of 0.86. Data obtained for the study were analyzed using the Scheuneman modified Chi-square statistic to analyse all the research questions and chisquare test statistic was employed in testing the null hypotheses 1-8 while two sample t-test of independent was used to test hypotheses 9-1 1 at 0.05 level of significance. Results showed that Agricultural science multiple-choice test items used in NECO and NABTEB 2015-2017, contain test items that significantly functioned differentially for testees on the basis of gender, school location, school type and socio-economic status. That there is no significant difference in the percentage of the number of items that functioned differentially by gender in favour of males and those in favour of females in the 2015-2017 NECO and NABTEB Agricultural Science multiple choice test examination. It was thus recommended among others that test item writers for public examinations should be encouraged to use simple English language. This will reduce the tendency of introducing additional task into the measurement of Agricultural Science ability.