The Impact of Escape Alternative Position Change in Multiple-Choice Test on the Psychometric Properties of a Test and it's Items Parameters

Abstract

This study aimed at investigating the impact changing of escape alternative position in multiple-choice test on the psychometric properties of a test and it's items parameters (difficulty, discrimination & guessing), and estimation of examinee ability. To achieve the study objectives, a 4-alternative multiple choice type achievement test consisting of 39 items in psychological and educational measurement and evaluation was constructed. The test had four different forms according to the escape alternative position. The statistical software Bilog - MG3 was used to analyze the responses of the total sample (1521) examinees according to the three parameter logistic model (3PLM) of item response theory (IRT). The results of the study revealed that there were statistically significant differences in the means of item parameters (difficulty & discrimination) between the first and the fourth forms, due to the changing of escape alternative position; the differences were in favor of the four form. The results also showed there were statistically significant differences in the means of examinee ability between the first and the fourth forms, due to the changing of escape alternative position; the differences were in favor of the fourth form, and there were no statistically significant differences between other forms. The results also revealed there were statistically significant differences among criterion validity coefficients in favor of form four of the test. Finally, significant differences were noticed in the value of empirical reliability coefficients in favor of the fourth form.

Keywords: Multiple-Choice Tests, Escape Alternatives, Item Response Theory.