Computer Adaptive Tests (CATs) are Replacing Classic Models Fast!

Modern GRE, GMAT, LSAT, SAT and MCAT is called computer adaptive tests (CATs) because computer adapts the questions to the level of your skill and performance. Sometimes before all psychological tools were offered in pencil and paper to measure different aspects of your personality and intelligence. It is not too far when all psychological tests shall adapt to the computer based algorithm. 

The computer is given access to a large number of test items classified in question types like graphs, antonyms, math, quantitative, etc. and arranged in order of difficulty. 

When you face Computer Adaptive Tests, the computer program continuously updates your estimation from the very first question to the last item. Initially, you are supposedly an average person. When you answer the first item correctly, the next question becomes a bit difficult. When you again answer correctly, another harder item is offered. The computer continues increasing the level of difficulty until you start answering wrongly. 

Watch Computer Adaptive Test Article on Video

Computer Adaptive Tests V.S Traditional Aptitude Tests

The CATs environment is different from the typical testing session. You sit in a computer lab or testing center while a monitor is placed in front of you. You may be alone in the examination room, or other candidates may be sitting in nearby cabins. 

You are required to start your test by clicking an icon on the computer screen. The first question appears, and you are required to answer it by choosing from the multiple options. Then you are required to confirm your answer by clicking a box on the same screen to move forward. Then a new question appears on the screen, and you repeat the process until you reach to the last question.

This mode of testing is different from the usual psychological testing session on many accounts. 

1-    In a normal pencil-paper standardized test, you are given specific items in the booklets of questions and answers. Every subject receives the same number and kind of items. Everyone has to face the same level of difficulty or ease. These comfortable and hard questions are like adding constants to someone's score. They provide relatively little information about the subject's ability level. Consequently, large numbers of items and examinees are needed to obtain a modest degree of precision. However, in a CATs, you may see the questions entirely different from the candidate sitting next to you. Your ability level relative to a norm group can be iteratively estimated during the testing process. You can be given the items that maximize the information about your ability level from your responses to different questions. You will receive items which are very easy or very hard depending on your responses to your current answers. 

2-    In a regular standardized psychological test, you may skip from one question to another by leaving many questions unanswered and then come back to make a well-calculated choice. You can answer the questions in any order as you like. However, in these tests, you are not allowed to skip through the problems. When you jump a question, you can’t come back to answer it later on. You are required to answer the questions in the order as they are offered on your monitor.  You need to calculate your estimated time that you can allocate for each item. You should not spend too much time on a question for the correct answer. When you are not sure of your answer, you should choose moving forward. Keep in mind that most of the computer adaptive tests don’t penalize you for your wrong choices.

3-    In a regular psychological test, you may have a second thought about your choice. You can go back, erase your previous pick and select a different option. However, this luxury is not available in the computer adaptive tests.

4-    In a standardized psychological you are free to make some marks, cross out incorrect options, underline keywords, and highlight questions to consider them later. However, this facility is not available in computer adaptive tests.

Components of CATs

The psychologists build computer adaptive tests with the following constructs:

The items pool

The pool of items must be calibrated with a psychological testing model. Typically item response theory (IRT) model is applied in the computer adaptive testing.

Starting Point

Computer adaptive tests assume that you are average and the first question is always an ordinary question.

Item Selection procedure

The IRT places you and the items on the same metric known as theta. So, when you answer your first question, it starts measuring your level of capability, and the process continues until completion of the test.

Item scoring procedure

When you answer a question correctly, your ability is scored higher, and when your answer is wrong, your capability level falls. Using the procedure, each item is composed of the already set value of the question.

Termination criteria

The computer adaptive tests (CATs) need a termination criterion otherwise you shall continue answering the questions until the whole item pool exhausts. Often computer adaptive tests are terminated when your standard error of measurement falls below a certain level.

A Primer on CATs

GMAT Test Preparation

Importance of GRE Exams

IELTS and GRE Comparison

Graduate Aptitude Test in Engineering

GRE Exam

GMAT test preparation 

ShareFacebookTwitterPinterestTumblrRedditWhatsApp