A Guide for More Accurate and Precise Estimations in Simulative Unidimensional IRT Models

There is a great deal of research about item response theory (IRT) conducted by simulations. Item and ability parameters are estimated with varying numbers replications under different test conditions. However, it not clear what the appropriate number should be. The aim current study to develop guidelines for adequate in conducting Monte Carlo simulation studies involving unidimensional IRT models. For this aim, 192 conditions which included four sample sizes, two lengths, eight replication numbers, models were generated. Accuracy precision parameter estimations model fit values evaluated considering replications. In context, parameters; mean error, root square standard error estimates, fit; M_2, 〖RMSEA〗_2, Type I rates considered. did seem influence fit, was decisive inflation prediction accuracy all It concluded that get more accurate results, be at least 625 terms rate estimation Also, 156 above can recommended. biases examined, largest bias obtained from 3PL model. increase resulted biased estimates.