COMPARISON OF CLASSICAL TEST THEORY AND ITEM RESPONSE THEORY ABILITY ESTIMATES OF SECONDARY SCHOOL ECONOMICS STUDENTS IN FEDERAL CAPITAL TERRITORY, ABUJA

This study is a comparison of the ability estimation of secondary school Economic students in Federal Capital Territory, Abuja based on the Classical Test Theory (CTT) and Item Response Theory (IRT). Survey research design was employed with 7 research questions raised while 8 research hypotheses were tested at the 0.05 level of significance. The population of the study consists of 18,643 SS II students (10,009 males and 8,634 females) from 206 Senior Secondary Schools (53 rural and 153 urban) that offered Economics at SSCE.The sample of the study comprised 903 SS II students (420 male and 483 female) Economics candidates from 23 Senior Secondary Schools (11 urban and 12 rural) in Federal Capital Territory, Abuja which were selected using stratified random sampling and purposive sampling techniques. A 50 item Economics Students‟ Achievement Test was constructed by the researcher and subjected to experts. Content validity with a logical index of 0.79was obtained for the instrument. The instrument was trail tested and a split-half reliability test conducted yielded 0.86 coefficient of internal consistency. Regular class teachers were used to administer the test, retrieve same for scoring by the researcher and thereafter subjected to CTT and IRT comparison using AcerConQuestand X-Caliber software. The research questions were answered using descriptive statistics (mean and standard deviation) while the hypotheses were tested using t-test statistics. The findings showed that the items have moderate psychometric indices when tested using CTT model while the psychometric properties as depicted by IRT indicated that one items was outside the acceptable range. With the CTT, all the 50 items were accepted and analyzed, while in IRT, one item was rejected as a result of no variance in the scores of the whole sampled students, the mean score for the CTT was higher than that of IRT. Also, there is significant difference in the difficulty indices of Economics test items between CTT and IRT among the testees irrespective of sex of the students. On the whole, the study established that IRT model possess better ability estimation than CTT. It was thus recommended that the examination bodies should embrace the use of IRT model for more accurate estimation of candidates‟ ability