This report
presents the findings of an evaluation of the item pools used in the computer
adaptive test (CAT) versions of the Maryland Functional Tests (MFTs). The item
pools were evaluated with regard to: (1) pool size; (2) adequacy of the test
information provided by the pool; and (3) balance of content domains and item
difficulty levels. The evaluation found that the mathematics pool should be
augmented with additional items, with the new items being substantially more
difficult than the current items. Increasing pool size was also recommended for
the reading test, but the item pool for citizenship was found to be a good
size. It is suggested that the three item pools can be used more effectively through
changes in the MicroCAT testing programs to adopt a CAT procedure that combines
the adaptive mastery testing (AMT) procedure (D. Weiss and G. Kingsbury, 1984)
with the use of testlets. The AMT procedure could then be used to adaptively
administer testlets until the termination criterion was reached. This procedure
would permit efficient, variable length testing while maintaining the content
balance. Using the testlet based AMT procedure would use the item pool more
efficiently, and items would consequently be exposed at a lower rate. Because
all the items are accessible in a CAT, care must be taken to protect the item
pool. Any judgments about changing the present CAT tests must take into account
the likelihood that they will be discontinued after the Maryland High School
Assessment is implemented.