Abstract

The current research aimed to identify the effect of selecting item response theory models and estimation methods and the interaction between these two variables on the accuracy of estimating item parameters of the Cattell intelligence test and the abilities of individuals. The research instruments used were Cattell's third-scale intelligence test (form A). The participants were (1104) male and female participants from Northern Border University in the Kingdom of Saudi Arabia; the mean for their ages was (20.80) with a standard deviation of (1.31). Using the statistical programs (SPSS- BILOG-MG) the statistical techniques (Two-way ANOVA, independent samples t-test, Scheffe test), the results revealed that there were statistically significant differences in the accuracy of difficulty parameter estimation due to the difference in the model used in favor of one model. There were no statistically significant differences in the accuracy of difficulty parameter estimation due to the difference in the estimation method or the interaction between the model and the estimation method. There were statistically significant differences in the accuracy of discrimination parameter estimation due to the difference in the model used in favor of the two models. There were no statistically significant differences in the accuracy of discrimination parameter estimation due to the difference in the estimation method or the interaction between the model and the estimation method. There were statistically significant differences in the abilities of individual estimation accuracy due to the difference in the model, estimation method, the interaction between the model used and the estimation method.

Keywords

Item response theory, estimation method, estimation accuracy, cattell intelligence test.

How to Cite