Using Rasch rating scale model to reassess the psychometric properties of the Persian version of the PedsQLTM 4.0 Generic Core Scales in school children

The PedsQLTM 4.0 Generic Core Scales was completed by 938 Iranian school children and their parents. Convergent, discriminant and construct validity of the instrument were assessed by classical test theory (CTT). The RSM was applied to investigate person and item reliability, item statistics and ordering of response categories.The CTT method showed that the scaling success rate for convergent and discriminant validity were 100% in all domains with the exception of physical health in the child self-report. Moreover, confirmatory factor analysis supported a four-factor model similar to its original version. The RSM showed that 22 out of 23 items had acceptable infit and outfit statistics (<1.4, >0.6), person reliabilities were low, item reliabilities were high, and item difficulty ranged from -1.01 to 0.71 and -0.68 to 0.43 for child self-report and parent proxy-report, respectively. Also the RSM showed that successive response categories for all items were not located in the expected order.This study revealed that, in all domains, the five response categories did not perform adequately. It is not known whether this problem is a function of the meaning of the response choices in the Persian language or an artifact of a mostly healthy population that did not use the full range of the response categories. The response categories should be evaluated in further validation studies, especially in large samples of chronically ill patients.There are two different methods for testing psychometric properties of quality of life (QoL) instruments, including classical test theory (CTT) and item response theory (IRT). Despite the popularity of CTT, it is unable to consider important aspects of measurement such as item difficulty, item discrimination and ordering of response categories [1]. Traditional psychometric techniques, called CTT, focus on summated scores, in which the scores on multiple items are added together [2]. In IRT, unlike in CTT, the properties of items can be anal