استخدام نموذج راش في تدريج مقياس التفكير الناقد لدراسة بعض العوامل المؤثرة على دقة القياس وفق نظرية الاستجابة للمفردة

Abstract EN

This study aims to reveal some of effective factors such as: the size of used sample in grading, best way to join between different test images, grading one calibration gradation (common vocabularies/common individuals), best number of contributed vocabularies or contributed individuals (who used to join the different test images).

In addition to the impact of used sample in grading vocabularies that effect on the accuracy of measures criterion.

In order to reach to best gradation of Rasch model to measure criticizing intellectuality ability.

It also aims to study the calibration performance (Percentile Ranks and T-scores) to interpret individuals capacity on calibration and using Winsteps program in grading the criticizing measurement of intellectual capacity according to Rasch model.

The study was made on 829 female and male students (4th and 5th primary class students) and(1st and 2nd intermedi The results showed that the best size between used sample sizes in grading measurement of criticizing intellectual capacity by using Rasch model is 800 according to measurement quality.

The best number of contributed and used vocabularies in joining different test images is 11 according to the measurement accuracy, the number of contributed and used individuals in joining test images is 50 according to the measurement accuracy criterion, and the best way to join between test images is joint vocabularies according to measurement accuracy criteria.

In accordance to measurement accuracy criterion, the measurement accuracy was different according to sample difference for the sample benefit with low level.

Measurement independence in measurement of capability was achieved (equivalence accuracy).

The vocabularies difficulty and estimation were graded of capability capacity on intellectual criticizing according to Rasch model after the availability of accuracy criterion.

Also, estimating all interviewed individual capability for each total raw score on capability measurement for criticizing intellectuality according to Rasch model after criterion accuracy existent.

It was succeeded to reach Percentile Ranks and T-scores that explain individuals capability levels on measurement.