What Is an Adaptive Computer?

Computerized Adaptive Testing (CAT) is a new form of testing developed in recent years. This kind of quiz is based on item response theory and computer technology. It forms a set of theories and methods in the construction of the test bank and the strategies for selecting topics. This article will briefly introduce the principles and methods of computer adaptive testing.

Computer adaptive test

The computer-adaptive test is different from the traditional paper-and-pencil test. The presentation of the test questions and the answers to the test questions are completed by the computer. The computer adaptive test is also different from the general computerized test. During the test, the computer not only presents the questions, enters the answers, automatically scores, and obtains the results. It also automatically selects the most suitable ones based on the different answers to the test questions. The test questions allow the participants to answer, and finally reach the most appropriate estimation of the ability of the participants. Therefore, the computer adaptive test is a test that varies from person to person.

Computer-adaptive test is a test developed on the basis of item response theory. It is a test for analysis at the item level. The author of this test believes that to measure a person's ability, the most ideal item is a moderately difficult item, that is, the probability that he answers correctly or incorrectly is about 0.5. At the beginning of the test, the computer generally gives a moderately difficult problem. If the test is done correctly, the computer will estimate his ability to be higher than the intermediate level, and then give him a more difficult problem. If he does wrong, the computer will Will estimate his ability to be lower than the middle level, and then give him a less difficult problem. The computer then responds to the second question. Estimate the ability again. On the basis of the second estimation, the computer selects the item closest to the estimated value of his ability in the question bank, and then estimates the ability according to the response of the participant. In this way, as the subject does more questions, the accuracy of the computer's estimation of his ability is getting higher and higher, and finally his estimated value will converge to one point, which is the more accurate estimate of the subject's ability.

Computer-adaptive tests have the following advantages that other tests cannot compare:

1. It can select topics differently from person to person, the topic is highly targeted, and it is possible to estimate the ability of the subject more accurately with fewer questions.

2. It is not necessary to specify the time of the test, and the participants can choose their own ideal time for the test. This is because computer-adaptive tests are different from person to person, and there is no need to set a uniform test time for fear of leaking test questions.

3 It can use multimedia technology to create a variety of vivid and vivid scenarios, so that the test presentation method can meet a variety of test purposes, and at the same time can stimulate test motivation.

4 Be able to understand the test results in time, and easily transfer the test results to the required department or individual through the network.

PS: The GRE General Test and GMAT computer test currently used by ETS belong to this form.