أثر حجم العينة و طول الاختبار على دقة تقدير معالم الفقرة و القدرة باستخدام نماذج نظرية الاستجابة للفقرة اللامعلمية

الملخص EN

The Study aimed at investigating the effect of the sample size and length of test on the accuracy estimation of the item-parameter by using the Non- Parametric item- response theory based on the indicators of measurement accuracy (Bias) and Root Mean Square Error (RMSE),To achieve the purpose of the study, the individuals' abilities were generated, The Sample of those participants consisted of (100, 250, 500, 1000) of natural distribution of (0 , 1) according to the category (20, 40, 60) items.

Under the assumption of the natural distribution of the itemdifficulty (0 , 1) and the distribution for the uniform distribution for the discrimination parameter in a primary value of (0.4) and final value of (1.2) and the uniform distribution for C parameter in a primary value of (0.2) and final value of (0.3) on the assumption of multiple-choice question test that suit the three parameter model using data generating program Win Gen v.3.

To answer the questions of the study ,Test graf programing was used to estimate item- parameter and the ability by using Non-parametric Kernel Smoothing (KS) and then comparing the estimated parameters with the real (generated) ours using Bias and RMSE.

The findings show statistically significant differences at (α = 0.05) in the means indicator of Bias attributed to the variable of the sample size .

The Findings also indicator statistically significant differences at (α = 0.05) in the means of Bias of the estimation in parameter C attributed to the variables of the study (Sample size and test length) and the interaction between them.

In addition, the Findings showed statistically significant differences at (α = 0.05) in the means of Bias in the estimation of the ability parameter θ attributed to the mentioned variables, whereas there were no statistically significant differences at (α = 0.05) in the means of Bias in the estimation of the difficulty parameter b.