Assessing difference between classical test theory and item response theory methods in scoring primary four multiple choice objective test items

Abstract

The study assessed the difference between Classical Test Theory (CTT) and Item Response Theory (IRT) methods of scoring multiple-choice objective test items. The sample comprised of pupils from three purposively selected primary schools in Pankshin Local Government Area of Plateau state. The descriptive research design was used for the study. Data were collected through a primary school mathematics diagnostic test (PRISMADAT) developed and calibrated based on IRT. All research participants were ranked on the CTT number correct scores and the corresponding IRT item pattern scores from their performance on the PRISMADAT. Wilcoxon matched-pairs signed-ranks tests were used to test the hypotheses. A statistically significant difference existed between the ranks participants obtained with CTT number correct scores and the ranks with corresponding IRT item pattern scores. It was recommended that teachers should be trained in IRT pattern scoring to use in assessing children's school performance instead of the CTT number-correct method.

Keywords: Classical Test Theory, Item Response Theory, Number correct scores