تقدير معلمة أنموذج راش للبيانات المصنفة لنظرية السلوك الكلاسيكية بطريقة المحاكاة

Abstract

This research aims to study the simulation to estimate parameter of the Rasch model categorical data as the difficulty parameter of item on intelligence tests and track the impact of (the estimated parameter, volumes samples) by the mean Square errors (MSE), and The mean absolute percentage error (MAPE) parameter model estimated by the conditional of Maximum likelihood estimation method.

The main conclusions of the research are: that the average potential of the values ​​of the difficulty parameter of individuals in response to the items and the dividend and for all sizes of the samples are negative and positive, and the MSE values ​​of the difficulty parameter of the items are for regular distribution of less than MSE values ​​of normal distribution, and the MSE values ​​of normal distribution be less of MSE for the distribution of beta values, Whatever the estimated parameter is negative or positive signal.