Construction of an item bank using Rasch Analysis

• Main Authors: Bahrom, Suhaila, Mat Nawi, Mohd Nazim, Saidin, Nurhafizah, Mat Yaacob, Siti Nubailah, Happas, A'fifah, Shaharom, Noraslinda
• Format: Proceeding Paper
• Language: English; English; English; English
• Published: 2023
• Subjects: QA Mathematics; QA276 Mathematical Statistics; QA300 Analysis
• Online Access: http://irep.iium.edu.my/105817/1/Construction%20of%20an%20Item%20Bank%20Using%20Rasch%20Analysis.pdf; http://irep.iium.edu.my/105817/2/watch_v%3D68giKNB6-Xw; http://irep.iium.edu.my/105817/3/ICReST%202023%20Abstract%20Book%20v5.pdf; http://irep.iium.edu.my/105817/4/Certificate%20of%20Attendance%20SUHAILA%20BINTI%20BAHROM.pdf; http://irep.iium.edu.my/105817/

An item bank is a large collection of high-quality test items that have been analysed and systematically stored in a computer, making them easily accessible to exam setters. The purpose of this study was to develop a new item bank and to assess the validity and reliability of the final examination paper for a Statistics course taken by 344 students. The course included 7 structured questions, divided into 21 items. Rasch analysis was used to analyse the data, linking all item difficulties and students' measured abilities on the same linear scale. The results showed that the Statistics course was unidimensional, with a Cronbach’s alpha value of 0.90, item and person separation greater than 2, and person and item reliability of 0.87 and 0.99, respectively. Nineteen items were found to fit the measurement model and were stored in the item bank, while two items were found to be misfits and require revision. In summary, a well-developed item bank can be extremely beneficial to exam setters by making test construction easier, faster, and more efficient, resulting in higher quality tests.