Akademik Veri Yönetim Sistemi
JOURNAL OF MEASUREMENT AND EVALUATION IN EDUCATION AND PSYCHOLOGY-EPOD, cilt.7, sa.1, ss.170-181, 2016 (ESCI)
On this research, it has been aimed to the examination of consistency of Localized Grey Relational Grade-Problem (LGRG-P) values and accommodation item classifications which are made via Problem Caution Index (PCI) between different samples that are randomly chosen from the same universe. In accordance with this purpose, 20 items and 1000 response patterns based on these items have been obtained by simulation, 10 different dataset (sample) which are randomly chosen among response patterns and involve 30 response patterns have been constructed. In order to examination of consistency of obtained LGRG-P values among different samples, Spearman Rank-Order Correlation Coefficients have been calculated and additionally significance of difference among value averages has been examined by Kruskal Wallis-H Test. Moreover, with the help of percentages at what rate of item classifications made by PCI obtained from samples show accommodation. Findings which are obtained have showed that Sato Theory Parameters which are produced in order to determine the quality of items have given consistent results.