1) The paper provided solid theoretical analyses and designed the study well (e.g., section 5.5) to clearly and logically form and support its hypotheses. Also, the authors used many “real situations” in CAT to emphasize the contribution of this paper to the CAT field. Such writing helps readers understand the paper easily and this paper is considered as an excellent writing example for journal articles.
2) Why should an item with a higher chance of guessing the correct answer be used for more capable test takes?
3) Page 288, the authors stated that A-optimality prefers an item with high discrimination parameter on a single ability, but this statement is not very consistent with Figure 5. From my perspective, it seems that both D- and A-optimality tend to choose an item with high discrimination parameters on a single ability. Also, D-optimality prefers items with high discrimination parameters on multiple abilities.
4) Page 289. Tables 2 and 3 listed the findings of theta 1 intentional and theta 2 nuisance. It seems that the tables for 5.2 are missing, or did I miss anything?