Bücher von Musa Adekunle Ayanwale

Proper assessment leads to adequate judgement of examinees performance in external examinations. The method employed by any public examining bodies such as National Examinations Council (NECO) might be adduced to one of the reasons examinees¿ performances in Mathematics were fluctuating. Extant literature showed that NECO only established item and person parameters of their dichotomous and polytomous items using Classical Test Theory (CTT) which is sample-dependent, without exploring efficient frameworks by Item Response Theory (IRT) considering its efficacy for accurate calibration and ranking. This study, therefore, developed dichotomous (60) items and polytomous (5) items through IRT based calibration. Concurrent true ability indices and relative efficacy of IRT and CTT in ranking of scores were also estimated.The study was anchored to Item Response Theory (modern psychometrics theory), while instrumentation approach and counterbalance design were adopted. The study was conducted in two phases: Validation and calibration of multiple choice and constructed response items of Mathematics Achievement Test (MAT).

The researchers assessed comparability of item and person statistics of 2017 basic education certificate mathematics examination of National Examinations Council through Classical Test Theory (CTT) and Item Response Theory (IRT) measurement frameworks. The paper suggested that National Examinations Council should jettison the use of Classical Test Theory and embrace utilization of Item Response Theory framework during their test development, item analysis and scoring of basic education certificate mathematics examination test items. This would enable many public examining bodies in sub-Saharan Africa to espouse the invariant advantage of Item Response Theory during estimation of item parameters and person statistics of their candidates, because pivotal decision would be made from such estimation, which need to be valid and reliable.