المقارنة بين طرق تقدير القدرة باستخدام النموذج المناسب
- Details
- Category: Dept of Psychology
- Published: Tuesday, 01 May 2012 03:00
- Written by Aryaf Ahmad AlTrawneh
- Hits: 2347
Abstract
Comparison Between methods used for estimating ability using the appropriate model according
to standard error of estimation.
Aryaf Ahmad AlTrawneh
Mu’tah University, 2011
This study aimed at comparing between the ability estimation methods: (Maximum likelihood (ML), Maximum A posteriori (MAP) and Expected a Posteriori (EAP)) using the appropriate logistic model depending on the estimation of their standard error. The results of the national exam in Math and Science for the eighth grade which was prepared by the Jordan Ministry of education for the year 2010 were used to fulfill the purpose of this study. Software BILOG-MG was used to analyze the data.
The findings were as follows:
The Expected a Posteriori (EAP) method was the most accurate in estimating the ability followed by Maximum A posteriori (MAP) method and the Maximum likelihood (ML) method was the least accurate.
The Maximum likelihood (ML) method has the drawback of increasing the standard error of ability estimation especially at the two extremes of distribution.
The study recommends using the Expected A Posteriori (EAP) method for ability estimation which was the most accurate