Weighted Maximum Likelihood Estimator For Latent Trait And Its Efficiency Of The Simulation Study Based On The Four-Parameter Logistic Model

Four-parameter logistic model is a new developed item response theory model in recent years. It has incomparable advantage for two and three parameter model in theory. The research about four-parameter model is still in the initial stage and many problems have not been solved very well. In item response theory, the unbiasedness of ability parameter estimation is very important for the application of the models. This article applies a weighted maximum likelihood latent trait estimated method proposed by Warm to the four-parameter logistic model. Conduct a comprehensive evaluation of the effectiveness of this estimated method by simulation study. Compared with maximum likelihood estimation and expected a posteriori estimation, weighted maximum likelihood estimation corrects the ability parameter estimation of skewness to some extent and improves the accuracy of estimation, with better statistical properties.