标题：Effect of Q-Matrix Misspecification on Parameter Estimation in Differing Sample Sizes and Test Length for DINA DINA Modelde Q Matrisin Hatalı Belirlenmesinin Farklı Örneklem Büyüklükleri ve Test Uzunluklarında Madde Parametrelerine Etkisi DOI Number: http://dx.doi.org/10.22521/jesr.2017.71.4 Hatice FIRAT & Mustafa
摘要：In Cognitive Diagnosis Models, every item in the measurement tool has a different effect (whichis determined based on the attribute tested) on the classification of individuals in terms ofattributes tested. One of the most effective factors that affects the quality of implications andthe accuracy of classification, is to develop proper item‐attribute relationships, in other words,the correctness of Q‐matrix Misspecification of the Q‐matrix leads to incorrect decisions aboutthe individuals. The present study, serving as a fundamental research, investigates the effect ofthe Q‐matrix misspecification in the DINA model on parameter estimations in the datasets,which are designed as a simulation and have differing sample sizes (50, 100, 250, 500, and 1,000participants) and test length (15 and 30 items). The parameter estimations were made by usingMarkov Chain Monte Carlo method based on Bayesian estimation. The estimations formisspecified Q‐matrix have been compared to item parameters regarding the correct Q‐matrixappropriate to dataset. In the case of underspecification in Q‐matrix, slipping parameters fordeficiently specified items and standard error values related to these; in the case ofoverspecification, guessing parameters related to overestimated items and standard errorvalues related to these were overestimated. The parameter estimation is affected by the Qmatrixmisspecification in all of the conditions discussed. Nevertheless, the amount of error inestimation does not show a regular differentiation in accordance with the sample size.
