ITEM PARAMETERS, SCORING MODELS AND ABILITY ESTIMATES OF DISTANCE EDUCATION STUDENTS: IMPLICATIONS FOR THE NATIONAL OPEN UNIVERSITY OF NIGERIA

Abstract

This study investigates item parameters, scoring models, and ability estimates of distance education students, with implications for the National Open University of Nigeria (NOUN). As the institution continues to expand its distance education offerings, understanding the effectiveness of assessment tools becomes increasingly crucial. The study employs both Classical Test Theory (CTT) and Item Response Theory (IRT) to analyze item parameters. Survey data were collected from a sample of 126 second-year Mathematics students at NOUN. Instruments used for the analysis include 1-Parameter Logistic Model (1PLM-MAT), 2-Parameter Logistic Model (2PLM-MAT), and 3-Parameter Logistic Model (3PLM-MAT). Data were analyzed using R and SPSS software packages to facilitate data interpretation. The findings reveal that polytomous scoring models provide more reliable estimates of student ability compared to dichotomous scoring models under the 2-PL and 3-PL models. The study recommends that NOUN consider integrating a variety of scoring models, including those incorporating partial credit scoring and polytomous item responses, to enable a more detailed assessment of student performance, particularly in disciplines where binary scoring may not fully reflect learning outcomes.