The Relation Between Classical Test Theory and Item Response Theory: Computerized Adaptive Test Perspective

This study aims to transform the calculated item difficulty statistics according Classical Test Theory (CTT) into parameter of Item Response (IRT) by utilizing normal distribution curve and analyze effectiveness this transformation based on Rasch model. In regard, 36 different data sets created with catR package were studied. For each set, parameters transformed correlation coefficients between these analyzed. Then, Computerized Adaptive (CAT) simulations performed using parameters. According simulation results, estimated theta values both methods high. Furthermore, in CAT which used, especially samples over 250, it was found have similar bias, RMSE values, average number administered items.