The Effect of the Difference in the Distribution of the Level of Ability that is Skeweded Positive for the Parameters of the Items of the Mental Ability Test According to the Item Response Theory

Abstract

The research aims to study “the effect of the difference in the distribution of the level of ability that is skeweded positive for the parameters of the items of the mental ability test according to the item response theory.” In the light of their observations, no item in the test was excluded because its percentage is (100%). Thus, the descriptive validity of the test was confirmed, and the test was applied to a sample of (60) male and female students chosen at random from middle school students / fourth grade (scientific and literary). It became clear from this experience that the instructions and items of the test are clear and understandable. To conduct the statistical analysis, the test was applied to a sample of (1000) male and female students, depending on the three-way teacher model. On the lower limits of "Gatman", which is a statistically significant factor when the potential root that can be explained is equal to or greater than (1), and the adoption of a saturation percentage (0.30) to accept the saturation of each of the test items with the general factor. The saturation of the test with the general factor showed what except for (6) items of the sequenced items (7, 19, 34, 51, 73, 86). They were deleted so that the test consisted of (88) items. After that, the researcher entered the data and analyzed it again with the ByLock program to judge the suitability of the test items to the model. Depending on the value of the chi-square, where it appeared that (80) of the test items were appropriate to the model and their value of the chi-square (Ka)2 was not statistically significant, while the value of the chi-square for (8) items was statistically significant, which is of the sequence (1), 13, 33, 44, 54, 68, 75, 83. Therefore, they were excluded, and the test consisted of (80) items, and the analysis was re-analyzed for the second time in order to identify the inappropriate items for the purpose of excluding them from the test. Its discrimination coefficient (5.0) to (5.2) and the items in which the guessing coefficient was greater than (30.0), in which two items were deleted from the test and its sequence (19, 43) because its estimate coefficient amounted to (0.408, 0.442) respectively. The analysis was repeated for the third time, and the test in its final form consisted of (78) items, and the validity was verified (descriptive validity, functional itnegraity), and the value of the reliability coefficient was . ( 0.991 ) To achieve the objectives of the research, the data was generated using the Wing program in order to obtain the shape of the positive skeweded distribution. The results showed a statistical significance in the difficulty, discrimination and estimation of items according to the variable form of the positive skewed distribution.