Multidimensional computer adaptive testing (MCAT) is capable of measuring multiple dimensions efficiently by using multidimensional IRT (MIRT) applications. There have been several research studies about MCAT item selection methods to improve the overall ability score estimations accuracy. According to the literature review it has been found that most studies focused on comparing item selection methods in many conditions except for the structure of test design. In contrast with the previous studies, this study employed various test design (simple and complex) which allows the evaluation of the overall ability score estimations across multiple real test conditions. In this study, four factors were manipulated, namely the test design, number of items per dimension, correlation between dimensions and item selection methods. Using the generated item and ability parameters, dichotomous item responses were generated in by using M3PL compensatory multidimensional IRT model with specified correlations. MCAT composite ability score accuracy was evaluated using absolute bias (ABSBIAS), correlation and the root mean square error (RMSE) between true and estimated ability scores. The results suggest that the multidimensional test structure, number of item per dimension and correlation between dimensions had significant effect on item selection methods for the overall score estimations. For simple structure test design it was found that V1 item selection has the lowest absolute bias estimations for both long and short test while estimating overall scores. As the model gets complex KL item selection method performed better than other two item selection method.