1. The algorithm of item selection in a CAT operates based on a specified dimensionality structure. Since that, what is the main reason of the claim that item responses collected by a CAT should be to reanalyze to investigate the dimensionality structure again?
2. The covariance matrix was not manipulated in this study. How about the performance when higher or lower values are given to the off-diagonal elements?
3. The simulated conditions are too restricted. Consequently, employing other IRT models may yield different conclusions.