The study aims to reveal the effect of test items orderin estimating persons and items parameters using the five parameter logistic model, using the descriptive survey methodology. The study population consisted of all secondary students in the scientific stream in Ein Albash district, approximately their number (1940) male and female students. The study sample consisted of (580), (290) male and (290) female by using intentional sample. The study instrument consisted of three experimental models. The validity of the content, construction, criterion-related and reliability was verified by internal consistency. Results were obtained for matching individuals and items to the five parameter logistic model using matching indicators. persons, items and standard errors were parameter estimated in the classical test theory (difficulty, discrimination) and item response theory (difficulty, discrimination, guessing, Upper Asymptote and asymmetrical).Assumptions were achieved (unidimesionality, monotone homogeneity, item characteristic curve, test and item information function).The results in the classical test theory have found differences between the mean parameters (discrimination) of the two ways order logical difficulty and random difficulty, in the favor for logical difficulty. Also found differences between the mean parameters (difficulty in the classical test theory and discrimination in item response theory) of the two ways order (sequence of subject content and logical difficulty) on one hand and random on the other, in the favor for (sequence of subject content and logical difficulty).
"The effect of items order in estimation of persons and items parameters using the five parameter logistic model,"
An-Najah University Journal for Research - B (Humanities): Vol. 36:
2, Article 5.
Available at: https://digitalcommons.aaru.edu.jo/anujr_b/vol36/iss2/5