Application of item response theory in calibrating 2020 neco mathematics multiple-choice questions
Romy O Okoye, Somtoo Victor Nduka
Item analysis is indispensable in the field of measurement and evaluation, as it provides a means of assessing the quality of test items. This study demonstrates the use of the item response theory to calibrate the test item parameters (item difficulty and item discrimination) of the 2020 NECO Mathematics multiple choice questions. The population for the study was made up of all the 26,421 SSIII students of the 2020/2021 academic session who enrolled for the NECO examinations in Anambra State, out of which a sample of 1,828 was selected by the use of purposive sampling technique. The students’ responses were analyzed, and the 4PLM was used to calibrate the IRT-based parameters. The findings from this study revealed that out of the sixty (60) items in the test, twenty eight (28) were moderately easy and twenty nine (29) were moderately difficult as they respectively fell between -3 to 0 and 0 to +3 on the logit scale. These fifty seven (57) items were, therefore, considered as good items. However, three (3) items were found to be too easy as they fell below -3 on the logit scale, and therefore, were considered bad items. In terms of the discriminating powers, eighteen (18) items had moderate discriminating power, nineteen (19) and seventeen (17) items respectively had high and very high discriminating powers. These fifty four (54) items were, therefore, considered as good items. However, two (2) and four (4) items respectively had low and very low discriminating power, and therefore, considered as bad items. The study concluded that the two item statistics (item difficulty and item discrimination) produced almost the same item characteristics, and therefore, recommended that item analysis should be maintained in the test development process because of its importance in determining the overall effectiveness of the test.
Romy O Okoye, Somtoo Victor Nduka. Application of item response theory in calibrating 2020 neco mathematics multiple-choice questions. International Journal of Advanced Education and Research, Volume 7, Issue 1, 2022, Pages 1-5