دراسة مقارنة في حساب الثبات بطريقة الاحتمال المنوالي لاختبار القدرة العقلية ثلاثية و رباعية البدائل

الملخص EN

The current research aims to investigate the effect of the number of multiple choice (tripartite and quadrant) to test mental ability in assessing each parameter of difficulty and discrimination, the variance of items, the estimation of validity parameters, and the extracted validity for items by probability mode.

To achieve the aims of the current research, the mental ability test was approved with its four multiple choice from (thorpeseducation, 2018).

After verifying the validity of the translation for the test, a second version of this test was worked out, but with tripartite multiple choice after eliminating one alternative.

After presenting the two test versions to a group of referees in the psychological and educational sciences, approval was obtained for the version with the quadrant multiple choice through the Kai square in terms of logical validity between the item and the concept, as well as their approval by 100% on eliminating alternative from the quadrant multiple choice to the original version.

Then it was applied to a sample of (65) to ensure the integrity and clarity of the items and instructions.

The test was applied in its two forms (tripartite and quadrant multiple choice) through two links of (Google form) on two samples each of which reached (400) students for each test version from universities in Baghdad Governorate.

The moderation of distribution was verified through some statistical indicators, and then the psychometric characteristics of the items of the second version of the test were extracted, as the difficulty parameter was extracted for each item of the mental ability test with tripartite multiple choice.

It was found ranging between (0.26-0.53), while the quadrant multiple choice for the second version of mental ability test ranged between (0.30-0.56).

The discrimination parameter for the items of tripartite multiple choice version ranged between (0.31-0.50) and quadrant multiple choice version between (0.30-0.56).

The variance of the test items for test version with tripartite multiple choice ranged between (0.20-0.25), and thus the variance reached its extreme end between the two test versions.

The value of the reliability parameter for the items by probability mode of tripartite multiple choice ranged between (0.33-0.60) and it was found that all the items had a good reliability parameter except for five items, which constitute (0.13)% of the total items.

The quadrant multiple choice of the mental ability test, calculated in the same way as the previous one, was good and the values of its reliability parameter ranged between (0.44-0.68) except for two items, and it constituted (0.05)% of the total items.

The reliability parameter was also calculated by the Kuder-Richardson method (21) for the test version of tripartite multiple choice and its parameter was (0.698), while the reliability parameter of quadrant multiple choice was (0, 2).

The current research has reached the superiority of the mental ability test with its quadrant multiple choice in light of its psychometric characteristics (difficulty - discrimination - variance – validity, reliability) in the version of mental ability test with its tripartite multiple choice, but the characteristics of the latter were not weak as it had good characteristics.

The research came out with a set of recommendations and suggestions.