Efficacy of Item Response Theory in the Validation and Score Ranking

ISBN-13: 9786204209142 / Angielski / Miękka / 300 str.

Musa Adekunle Ayanwale

Opis

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).

O autorze

Ayanwale, Musa AdekunleThe author holds PhD (Educational Evaluation/Psychometrics & Assessment), M.Ed (Educational Evaluation/Test & Measurement), Post Graduate Diploma in Education, and B.Sc. (Statistics) respectively. Senior lecturer at the Dept. of Education Foundations, Kampala Internat. Univ., and a postdoctoral Fellow at the Univ. of Johannesburg, South Africa.