January 21, 2025

Corlyn Sophia Tablante Salazar 

9th Grade

John F. Kennedy High School

Have you ever taken a standardized test, like the SAT and have thought “How does this work?” Many standardized tests function differently, but self-adaptive or adaptive tests are one of the most common types of standardized tests.

But what is self-adaptive standardized test-taking and how does it function?  Computerized Adaptive Testing (CAT) is a method of testing, where the test varies in difficulty depending on the examinee’s ability and physiological condition. These tests vary depending on the mental state of the examinee. The questions on a test are selected from a pool of computed items by a computer algorithm or AI, while the examinee is still taking the test. Self-adaptive tests function the same way as CAT, but they allow the examinee to choose the type of questions they want for a more stress-free and accurate process. 

The Item Response Theory (IRT) grades each question, the IRT can give more precise results on the examiner's performance and can select the difficulty of each question. Usually, the questions are 3x more than the intended amount. Due to the algorithm of the test, some go by quickly and some take ages. The algorithm is supposed to max out the process. After each question that the IRT grades, it has to be evaluated termination criterion. Here, if the algorithm stops for a specific reason, it checks to see if some precision was made by the examinee. For example, say that you’d got the question “ the square root of 90” wrong because you’ve put “9.4” instead of “9.48.” Through the process, it can say if there would need to be changes within the algorithm, question, or the examinee’s answers. In this case, the software can say that “9.4” is accurate, and the algorithm can be reevaluated to say that it’s correct depending on if some issues or concerns were made with the test itself. Another way to see the precision of the examinee is through the shadow test. The Shadow test is created to be able to adapt to maximize the ability of the examinee. After an examinee is done with each question, the shadow test looks over it and compares the examinee’s answer to the overall estimated score. This happens before the test scores. After all this, the score’s report is made, based on the ability and physiological condition of the examinee. Through the results, we can see the mental state or intelligence of the examinee. 

Adaptive standardized testing began in 1905 by a man named Alfred Binet. His adaptive tests were paper-based and used to study the examinee’s ability at the start of the test. However, it was canceled in the 1950s’. In the early 1970s’ it started to gain more popularity, due to its accuracy, fairness, and overall better experience. When computers started coming around, adaptive tests went from paper to computers. Thus, the name CAT appeared. Since the tests are made to fit each individual’s physiological condition, they could be personalized. That’s how the tests became self-adaptive.

Over time, self-adaptive and adaptive standardized tests became more common, due to how quick, flexible, and accurate they are. Some of the tests you know, like the SAT are adaptive as well. Standardized tests are created in different algorithms, depending on the objective.