البنية العاملية لمقياس الإخفاقات المعرفية باستخدام نموذج سلم التقدير والتحليل الشبكي العصبي
The research aimed to explore the factorial structure of the Cognitive Failure Questionnaire (CFQ) through the application of Exploratory Graph Analysis (EGA), which is based on psychometric neural network analysis. This approach was utilized to estimate the dimensional structure of the CFQ scores and to determine the stability of the extracted dimensions and the items within each dimension. Furthermore, the study assessed the importance of individual items using four centrality indices: betweenness, closeness, strength, and expected influence. The Rating Scale Model (RSM), a type of item response theory model, was employed to evaluate the fit of the items to the model's assumptions, assess item quality, and estimate item parameters. The research sample considered of 929 students in general diploma program and bachelor program at faculty of Education, Port Said University, with an average age of 21.772years and a standard deviation of 4.152 years. Questionnaire developed by Broadbent et al. (1982) was administered. Data analysis included the use of means, standard deviations, exploratory factor analysis, fit indices, and parameter estimates derived from the RSM. The analysis was conducted using SPSS 25, JASP 18.1.0, and R software. Results revealed that 17 items (68%) demonstrated good fit to the Rating Scale Model, with item reliability at 0.885 and person reliability at 0.883. The findings also indicated that the best factorial structure for the CFQ was a three-factor model. Moreover, centrality indices highlighted that items A22, A17, A15, and A21 were the most important and reliable within the psychometric neural network of the scale. (Published abstract)