Computer Adaptive Testing on the GMAT

August 10, 2016

What is an adaptive test? How should you change your study strategy to prepare for the GMAT Computer Adaptive Test (CAT)? Computer adaptive testing definitely puts pressure on your testing experience and forces you to adapt your test-taking strategy in small ways. It’s especially important to understand how the GMAT calculates your score and what this means for you as a test-taker.

The GMAT uses Computer Adaptive Testing (CAT)

This means, first of all, that each question you answer correctly or incorrectly determines what questions you will see later in the GMAT. It also means that any two people, even two people of nearly identical abilities and preparedness, will not see identical questions when they take their respective GMATs. BUT, because of the magic of psychometrics, two people who perform with comparable skill & strategy & focus will have comparable GMAT scores. The magic of how computer adaptive testing is different for everyone but the score is fair for everyone—if you don’t have a Ph.D. in Psychometrics or Statistics or something like that, then just take that magic as an article of faith.

It’s also very important to point out: only the GMAT Quantitative and Verbal sections are computer adaptive. The Integrated Reasoning is NOT computer adaptive, and of course, the AWA essay is not. Everything I say in this blog about the CAT concerns only the Q & V sections.

What is an adaptive test and how does the GMAT CAT work?

The GMAT is trying to figure out objectively your Quantitative Ability and your Verbal Abilities in a relatively short time. Think of computer adaptive testing as a big “twenty questions” game. Suppose your “opponent” picks a US city, and you are allowed to ask “horizontal yes/no questions” (e.g. “Is your city east/west of X?”) and “vertical yes/no questions” (e.g. “Is your city north/south of X?).

You might ask a bunch of horizontal questions. It is west of Albuquerque? No. Is it east of Atlanta? No. Is it east of Denver? No. Is it east of Santa Fe? No. OK, that narrows things down to a relatively thin vertical band.

Then a bunch of vertical questions. It is North of Wichita? No. Is it north of Birmingham, AL? No. It is north of New Orleans? Yes. Is it north of Tucson? No.

Among major cities, those answers are enough to home in on El Paso, TX. Much in the same way, the GMAT asks you two question types, Math and Verbal. By giving you easy and hard questions of each kind, the computer adaptive testing homes in on what is most appropriate for your level.

How accurate is the Computer Adaptive Test?

The “twenty questions” analogy is helpful for understanding computer adaptive testing, but the latter is more complex: after all, what the CAT is measuring is not cut-and-dry exact.

Your score is a composite result that takes into account the difficulty of each question you got right and the difficulty of each question you got wrong. The exact details of the algorithm that the computer uses to do this are (a) probably incomprehensible if you don’t have a Ph.D. in Psychometrics, and (b) the secret proprietary information of GMAC. Legally, we can’t get access to that algorithm, and even if we could, we probably wouldn’t understand it anyway. Rest assured, though: it is accurate. 🙂

What does it imply about the accuracy of the computer adaptive testing if a person takes and retakes the GMAT multiple times, getting a different score each time? This empirically observable fact doesn’t compromise the accuracy, the statistical robustness, of the computer adaptive testing itself.

Does the GMAT only measure how well you can take the GMAT? Well, yes and no. It is not a pure intelligence test, insofar as one can take a course and improve: for example, with Magoosh you can increase your score by up to 70 points or get your money back guaranteed! See our score guarantee page for details. Nevertheless, the folks at GMAC, the place that produces the GMAT, will tell us that, as a statistical tool, the GMAT CAT is not only robust but valid. To be statistically valid means that the test has predictive power: in particular, the GMAT is very good at predicting how one will handle the academic demands of business school. If by studying you make yourself more prepared for the GMAT, unwittingly you have also made yourself more prepared for your future business school classes.

Getting the GMAT Computer Adaptive Testing Facts Straight

False: You can outsmart the CAT.

The algorithm is far too complex. There’s no sense stressing about “how did I do on those questions?” or “why is it asking this kind of question now?” Again just do your best on the question in front of you at any moment, submit it, and then forget about that question entirely.

True: Systematically reviewing math and verbal content, as well as strategies specific to each question type, can vastly enhance your GMAT score.

That is most certainly true, and that’s why Magoosh can give you such an advantage. With a couple hundred lesson videos discussing both content and strategy, and over 800 practice questions, each with its own video explanation, you will get top-notch preparation for the GMAT at only a fraction of what you would pay for a comparable course.

True: You can get several questions wrong and still get a good score.

The computer adaptive testing has to give you several questions well above your ability, questions that you almost invariably will get wrong, in order for it to zero in on your actual ability. You are not penalized for that: that’s just what the computer adaptive testing must do as part of its algorithm.

In fact, here’s the irony: the person who gets a 460 and the person who gets a 780 may get approximately the same number of questions right & wrong. The number of questions correct, by itself, indicates nothing: all that matters is the difficulty of the questions. The person who got a 460 probably got several relatively easy questions wrong but even easier question right. The person who got a 780 got many super-hard questions correct but even harder ones wrong. The computer adaptive testing has an extremely sophisticated way of finding your level and customizing the analysis to someone at your level.

True: Not finishing all the questions in a section hurts your score.

That is quite true. It’s exceedingly important not only to learn content and strategy, but also to practice working efficiently so that you don’t run out of time. Ideally, you want to hone your time management skills on GMAT practice tests and questions so that you have abundant time on even the last questions on a section.

If you are running out of time, you should understand the strategies concerning solution behavior and guessing.

Want to learn more about what’s fact and fiction on the GMAT? Check out these common GMAT myths and realities explained by a GMAC representative!

Author

Mike MᶜGarry

Mike served as a GMAT Expert at Magoosh, helping create hundreds of lesson videos and practice questions to help guide GMAT students to success. He was also featured as “member of the month” for over two years at GMAT Club. Mike holds an A.B. in Physics (graduating magna cum laude) and an M.T.S. in Religions of the World, both from Harvard. Beyond standardized testing, Mike has over 20 years of both private and public high school teaching experience specializing in math and physics. In his free time, Mike likes smashing foosballs into orbit, and despite having no obvious cranial deficiency, he insists on rooting for the NY Mets. Learn more about the GMAT through Mike’s Youtube video explanations and resources like What is a Good GMAT Score? and the GMAT Diagnostic Test.

View all posts

Comments

25 responses to “Computer Adaptive Testing on the GMAT”

DeepikaYadav

February 2, 2018

Hii mike..Is their any negative marking for wrongly answered questions…or only the difficulty level will be adjusted by CAT…

In total how many questions of each section u have to attempt ..n in what duration??

Reply
1. Magoosh Test Prep Expert
  
  February 2, 2018
  
  Hi Deepika,
  
  Good question. 🙂 Strictly speaking, no, there is not any negative marking. But as the test adapts based on every answer you give, some might argue that this is sort of negative marking because you may lose points according to the algorithm’s assessment of your performance. You must attempt as many questions as you can during the test! You have 75 minutes to answer 37 quant questions and you have 75 minutes to answer 41 verbal questions, so that is your duration.
  
  Reply
Ashutosh

March 28, 2017

Hi..
If I am unable to answer a question like if it is very lengthy then can I leave the question unanswered??

Reply
1. Magoosh Test Prep Expert
  
  March 29, 2017
  
  You can chose to arbitrarily answer a question, but you cannot skip questions on the GMAT.
  
  Reply
Nick

May 20, 2016

Hi Mike,

I wanted to understand that is it true that GMAT is not for people who are so good at Quant? I read the chain of messages from above, my understanding is that if a person is struggling with hard questions in a set then that doesn’t appear in the exam. But then to get a good score in GMAT one needs to come across medium level sets.

Pls correct if else.

Reply
1. Magoosh Test Prep Expert
  
  May 24, 2016
  
  Hi Nick,
  
  The GMAT can be manageable for people who aren’t strong in quant, but all students will want to brush up on math basics and GMAT quant question style. The primary hurdle for the GMAT is being able to apply mathematical knowledge to their data sufficiency style of questions. Many people are not trained to think about math concepts in this way, so it poses a challenge, but it certainly is not impossible!
  
  The GMAT is adaptive, so if someone is struggling with harder questions, yes, the test will primarily shift to providing easier questions at the expense of a higher test score. The best score comes from advanced problems being answered correctly, of course, but there is some serious statistical power behind the scenes (and secret) that GMAC applies to the test to generate the final score you get. Not all medium questions, for example, are likely created equal. The best approach is, to be cliche, simply to “do your best.”
  
  I hope that addresses your concern! 🙂
  
  Reply
Q

March 12, 2015

Hi Mike,

I have a question about how the difficulty of the exam increases, this pertains mostly to verbal. Given that the verbal section includes 3 sections, SC, CR, and RC, how does the difficulty of these sections increase? Does it uniformly get more difficult i.e you get first couple correct and then all 3 question types become more difficult or does the difficulty increase by questions type? i.e start with medium level SC questions then as you get more SC correct, then you are fed harder SC questions, then you get a medium CR and then fed more difficult CR questions.

Thanks!

Reply
1. Mike
  
  March 13, 2015
  
  Dear Q,
  The short answer is no. 🙂
  First of all, as you may know, the RC questions come in “chunks” — a single passage has 3-4 questions, all in a row. Between these RC chunks, the SC & CR are distributed more or less randomly. Each time the GMAT gives you a new question, it is performing two separate tasks: it is choosing a question format (RC, CR, or SC) and it is also choosing a difficulty based on the CAT algorithm. So far as I understand, those two tasks are 100% independent. In other words, suppose Q #1-2 on your GMAT are SC, and Q#3 is a CR, and suppose you get all three right. At this point, the algorithm is going to be upping the ante, reaching for something a little harder, but whatever difficulty level the algorithm has in mind for the next question is entirely independent of which question type you see on the next question. The difficulty of the next question will be determined by the CAT algorithm, and it could be any of the three question types. Even if Q#4 begins a RC passage with 3-4 questions, this process continues. They have multiple questions of different difficulty levels for each passage: on any test, you will only see 3-4 of the questions for that passage, but they have many more up their sleeve. Even if you are in the middle of a RC chunk, the algorithm keeps adjusting the difficulty question by question.
  The goal of the question-format-choosing task is to make sure that, by the section’s end, you see a more or less equal mix of the three question formats. The goal of the CAT algorithm, quite separate, is to determine with a high degree of psychometric precision your Verbal score. The CAT algorithm doesn’t care at all about measuring your “SC ability” vs. your “CR ability.” Students care about that sort of thing, but for the purposes of the official algorithm, that is entirely irrelevant. The section is psychometrically designed to measure an aggregate Verbal score, and that’s all it is measuring.
  Does all this make sense?
  Mike 🙂
  
  Reply
Adi

June 20, 2013

Hi Mike – Is it true that the verbal section plays a major role in calculating the GMAT overall score than the quantitative section?

Reply
1. Mike
  
  June 21, 2013
  
  Adi,
  I’m not sure I understand your question. Both V & Q sections are both important — in fact, the combination of them is most important in determining overall GMAT score.
  Have I answered your question?
  Mike 🙂
  
  Reply
  1. Adi
    
    June 21, 2013
    
    I read somewhere if you get a 94 percentile in V and 83 percentile in Q, you tend to get a better overall score (> 700) because people with this combination are rare compared to the people who’ve had a vice versa situation, see them more often. Therefore, higher the V score, better your overall GMAT score. Hope that makes sense.
    
    Reply
    1. Mike
      
      June 21, 2013
      
      Adi,
      That’s precisely what I meant when I said the *combination* is important. I would be hesitant to say either section is “more important”, because it’s very important to do your best on both. I wouldn’t want a student, thinking “V is more important”, to neglect their Q studies.
      Mike 🙂
      
      Reply
      1. Adi
        
        June 21, 2013
        
        Absolutely, I second you on that. I was just intrigued the way GMAT scoring actually worked. Thanks, Mike. 🙂
        
        Reply
        
        Mike
        
        June 21, 2013
        
        You are quite welcome.
        Mike 🙂
        
        Reply
bill murphy

March 3, 2013

Sorry to nitpick, Mike, but it’s “home in on,” not “hone.”

Reply
1. Mike
  
  March 3, 2013
  
  Bill,
  Thanks for pointing this out. I believe both are acceptable at this point, but “home in” is definitely the more traditional construction. I changed it in the text.
  Mike 🙂
  
  Reply
Eran

October 29, 2012

“Myth: The first question is super-important, because that determines the course of easy/hard questions from there”

This is a fact and not a myth, at least in the quantitative section.
It is almost impossible to get 49-50 if you make a mistake in the first three questions.
I saw it happen many times when I made a mistake in a very early stage of the GMAT CAT Test.

Reply
1. Mike
  
  October 29, 2012
  
  Dear Eran,
  Let’s be clear. Nobody knows how GMAT’s question-difficulty rating scale works. Let’s pretend it’s a 1-100 scale, and let’s pretend that the first question is, say, a difficulty of 50. If you make a mistake on a question of difficulty of 50, that is to say, a medium difficulty question, *anywhere* in the test, it will have the same impact. The question has no special leverage simply because it’s one of the first questions of the test. And yes, it will be very hard to get an elite score if you get medium level questions incorrect. Does all this make sense?
  Mike 🙂
  
  Reply
  1. Eran
    
    October 31, 2012
    
    Thanks. Yes, It does
    
    Reply
    1. Mike
      
      October 31, 2012
      
      You are quite welcome.
      Mike 🙂
      
      Reply
Faruk

July 27, 2012

Mike,Is it true that if anyone start answering all the questions right,he/she will not face any low level questions?difficulty level will increase constantly after answering each question right ?

Reply
1. Mike
  
  July 29, 2012
  
  No, that is absolutely NOT true. What is true is — if you keep answering questions right, then *most* of the questions you see will be of higher difficulty question. Nevertheless, the algorithm is not linear, not constant — it doesn’t just increase after each right answer. It’s doing something much much more sophisticated. As I said in this article, you cannot outthink the algorithm.
  Mike 🙂
  
  Reply
  1. Faruk
    
    July 30, 2012
    
    Thanks Mike..:)
    In OG,are the questions are arranged according to difficulty level or are they ordered randomly ?
    
    Reply
    1. Faruk
      
      July 30, 2012
      
      And Mike I also heard about one thing,don’t know it’s true or not..In GMAT questions appear as a cluster( a cluster contains 5-6 questions based on difficulty level)..Once you are done with one cluster,then the new cluster starts with a easy question and so on…is this true or a myth ?
      
      Reply
      1. Mike
        
        July 30, 2012
        
        Faruk: In the OG, the general pattern is from easy to hard, although this pattern does not appear strict. Let’s say easy-to-hard is a general overall approximate trend, though not necessarily true from each problem to the very next one.
        As far as the clustering theory — I have never heard anything of the sort about that from GMAC. I think I am going to say: that is a rumor with no factual basis. The algorithm is much more sophisticated than that. I would say: trying to figure out the CAT algorithm is a much a waste of time in GMAT prep as studying verb tenses in Croatian: neither one is going to advance your GMAT understanding one iota.
        Mike 🙂
        
        Reply