أثر نموذج نظرية الاستجابة للفقرة ذات الاستجابة المتعددة التدريج على دقة تقدير القدرات للأفراد والخصائص السيكومترية للفقرات والاختبار
The study aims at exploring the impact of polytomous IRT model used in analysis (NRM, GPCM, grm)* on the accuracy of estimation of individuals’ abilities and the psychometric properties of items and test. To achieve the purpose of the study, Wingen 3 program was used to simulate data for a test which consists of 50 items according to GPCM model. Each item consists of five levels, with four thresholds. Results indicated that the GPCM model is more accurate in estimating abilities than both grm and NRM models. Additionally, it was found that the NRM model is more accurate in estimating discrimination parameters of items, compared with the other two models; grm model was more accurate in estimating the difficulty parameters for thresholds than the other two models; grm model presents maximum amount of information of the test, compared with the other two models, and it was more relevant and accurate in analyzing data. Results also indicated that there were no significant differences in coefficient values due to polytomous IRT model. (Published abstract)