|
|
|
Welcome to the Adaptive Tests of Ability web site. All tests that are available on this web site contain item types that have shown consistent validity to measure intelligence. Currently, the tests are available for free to researchers who want to use a measure of fluid intelligence in their study (see New Users). See Research for references to supporting data for the tests on this web site. All tests on this web site are administered adaptively, such that the greatest precision is obtained for each examinee. Abilities are estimated using calibrations for the items that the examinee received. Calibrations are obtained from previous studies on the items and on the cognitive design principles. What is Adaptive Testing? Adaptive testing is a test administration process in which test items are selected dynamically from a bank (collection) of test items. The items are selected based on the examinee's ability estimate (responses to previously administered items). For example, an examinee who succeeds with a given item will be presented next with an item of greater difficulty. An examinee who does not do well with the given item would next receive an item of lower difficulty. After each response (good or bad), the test revises a stored estimate of the examinee's ability. Many revisions allow the estimate to become more reliable. Older testing proceedures, such as paper-and-pencil tests are typically "fixed-item" tests in which all of the examinees are given the same items within a test. The tests offer little information about an examinee's ability level due to the set difficulty or ease of the fixed items. Computerized adaptive tests allow an examinee's ability level to be iteratively estimated and new items can be selected based on the current ability estimate. Relatively few items are needed to be administered to maximize the information about their ability levels from the item responses. This leads to better and more accurate testing results. What is IRT (Item Response Theory) Item Response Theory (IRT) is a set of mathematical descriptions of traits that describe what influences or actions are involved in determining the correct response for an item. For example, verbal proficiency or mathematical facility. The 3-parameter IRT model uses: a - item discrimination parameter b - item difficulty parameter c - guessing parameter Each item contains a different values for these parameters. The parameters are used to calculate the probability of a correct response with the examinee's ability.
Summary An examinee is given a test that adjusts to the responses given. If the examinee answers an item correct, then the next item is harder. If the examinee does not answer correctly, the next question can be easier. The system keeps track of the examinees ability on the test.
Contact Information |
|
Send mail to webmaster with
questions or comments about this web site.
|