Evaluation of the Reliability Coefficient for the Achievement Test of Selected Defensive and Offensive Basketball Skills Using the Classical Theory and Rasch Model.

The aim of this research is to evaluate the reliability coefficient of the achievement test for selected defensive and offensive basketball skills using both the Classical Test Theory (CTT) and the Rasch Model (Item Response Theory - IRT) as part of modern measurement theory. This research gains its significance by highlighting the effectiveness and accuracy of tools and methods used in measuring athletic performance, contributing to the improvement and development of achievement tests. The research population was determined to include students from the College of Physical Education and Sport Sciences at the Universities of Duhok and Zakho, with a total study sample of 213 students. Data were collected using scientifically designed measurement tools, and reliability was analyzed using two methods: the classical theory and the Rasch Model. The results showed that the reliability coefficient based on the classical theory reached 0.85 using the Spearman-Brown method, reflecting a high level of consistency. On the other hand, the Rasch Model's Person-Item Reliability coefficient was 0.712, with fit indices (Infit and Outfit) ranging between 0.71 and 0.87, indicating good item fit to the model. The comparison between the two methods revealed that while the classical theory provides high overall reliability values, it lacks detailed analysis of item performance. Conversely, the Rasch Model offers a more comprehensive and precise evaluation of item quality and the overall test. The study recommends revising items with lower performance for improvement and combining both methods for a holistic assessment. It also calls for expanding the sample size and applying the research to other sports fields to enhance the generalizability of the results and the reliability of achievement tests. [ABSTRACT FROM AUTHOR]