For mass administration of linear paper-and-pencil tests, subsequent to each administration it is routine to evaluate the test forms delivered. With computerized adaptive tests, in which the number of test-takers receiving a given set of items is kept intentionally small, methods of post-administration evaluation are less obvious. Sets of rules (flags) for screening large numbers of adaptive test records were developed in an attempt to identify potentially problematic adaptive tests. Experiments were performed that involved modifications of adaptive test design variables, as well as the exploration of improvements to the item resources available, in order to attempt to reduce the number of flagged test-taker records. The most successful of these experiments indicated that improving item resources is the most effective means of enhancing the functioning of adaptive tests.