Comparative Study of Classical Test Theory and Item Response Theory Using Diagnostic Quantitative Economics Skill Test Item Analysis Results

In this study a comparison was made on DQEST item parameters and test statistics results estimated using Classical Test Theory (CTT) approach and Item Response Theory (IRT) three parameter logistic model (3PLM) to find out the similarities and differences in the two frameworks. 517 randomly selected senior secondary three (SS3) economics students comprised the sample. Three research questions guided the study. Responses obtained from SS3 economics students in 50 multiple choice items of Diagnostic Quantitative Economics Skill Test (DQEST) were used for the analysis. DQEST items certified the unidimensionality, local independence and model-data fit assumptions. Then results from CTT and IRT analyses were compared. In terms of very difficult item and item that discriminate poorly, CTT were found not to be comparable with the 3PLM the most appropriate model for DQEST data. The calculated reliability value for CTT was found to be low when compared to that generated by 3PLM. Therefore, it could be concluded that there was disparity between CTT approach and 3 parameter IRT model in terms of item parameters and test statistics. Thus IRT model with the best data fit should be employed for an enhanced test validity and reliability.