Barrada, Abad e Veldkamp (2009) compared both methods and found a slightly better performance for the eligibility method. Besides, Barrada, Olea e Veldkamp (2009) observed that it would pay off to randomize item selection at the beginning of CAT and to save the more informative items towards the end when the estimated ability is close to the true ability of the candidates. Besides, towards the end of the adaptive test, the candidates are spread over the whole ability continuum and over exposure is less of an issue. Because of this, multiple maximum exposure rates can be applied, depending on the position of the item in the test. They were able to demonstrate that a more balanced usage of the item bank could be obtained this way, with hardly any decrease in measurement precision.