مقارنة بين الطريقة البييزية و طريقة الأرجحية العظمى في دقة تقدير معلمة القدرة و معلمة الصعوبة وفق نموذج راش باستخادم بيانات مولدة (محاكاة)‎

Abstract EN

This study aimed at comparing estimation accuracy of ability and difficulty parameters when using Bayesian and maximum likelihood according to rush model using simulated data.

data was generated for two tests (50) and (25) items for two samples of (500) and (250) examinee using (Wingen) program , Bilog-MG program was used to estimate ability and difficulty parameter.

results showed statistically significant differences between means values of RMSE attributed to sample size and test length on favor of the Bayesian method, also no statistically significant differences between the mean values of RMSE in estimating difficulty parameter attributed to sample size and test length.

results showed statistical significance differences according to method of estimation on favor of Bayesian method, a statistically significant differences between means values of RMSE in estimating ability parameter according to sample size on favor of (500) examinee.