Time Saving Tip For Your GMAT Exam

Students often ask variations of the question:  “How much time should I spend on each __________ question?”.  Is it 2:00 per question?  Or 1:30 per data sufficiency, 2:15 per problem solving?  How much time should you spend reading an RC passage vs. answering the questions?  Should you spend more than 1:30 per critical reasoning question? 

The answer to all of the above is “it depends”.  That’s perhaps not the answer you want to hear, but it’s the one you need to hear.

Consider this situation:  When you spend 2:15-2:30 per math problem you get most of them right.  But when you spend less than 2 minutes you tend to make careless mistakes, forget to double-check assumptions/units/what-the-question-is-asking, etc.  Should you try to get your pace-per-question down to 2?  No!  Not if doing so means that you’ll probably get 25-30% of those questions wrong.  If you have a month or two until your test, you should work a little on recognizing common algebraic setups, using number properties and doing quick mental math, etc. But if you’re close to test day or you still need to learn about how to attack Data Sufficiency and other big concept areas, “trying to work faster” can be a catastrophic decision.  If, as you approach test day, you know that you typically need 2:15-2:30 per math question to get them right, take that time that you need.  And then guess 3-4 times over the course of the exam – odds are, you’ll get one of those right anyway just based on probability – to save yourself the 5-6 minutes that you need.

Why? With the GMAT’s adaptive engine, it’s important that if you can get a question right, you get it right.  If you invest the extra 20-30 seconds per question to get them right, then when it comes time to guess to get back on pace, you’re probably looking at a question at or around your upper limit of difficulty anyway.  You probably would have missed it even if you spent the time, so the guess doesn’t cost you much if anything, and it gains you back the time you required to earn such a hard question. But on the other side, say you rushed through those previous 6-7 questions to save time. Odds are, you made a careless mistake on 2 or 3 of them. You’re now lower on the ability level scale and you need to get the next couple questions right.  You haven’t allowed yourself to perform to your upper ability limit, so your score is languishing below your potential and you’ve actually raised the stakes – you have to get more questions right.

Now, this is a simplification of a nuanced scoring algorithm, but keep the point in mind – if rushing gives you a 25% chance of making a mistake but taking your time bumps that down to 5% (on questions you *should* get right), you’re much better off guessing once every ten questions to save that time (expected value:  you get ~9 of them right – and the one you miss, theoretically, is hard enough that you would have missed it anyway) than you are trying to cram in all ten (expected value: you get ~7.5 of them right, including 1-2 that you really shouldn’t have missed).

Know this about the GMAT: *most* test takers experience a time crunch on this exam. Nearly everyone has to make that decision of how to allocate time – that’s part of what the test is assessing.  And almost everyone – it can’t be more than a few exceptions per year if that – misses questions.  You’re supposed to get  questions wrong.  So accept that going into your exam – you will probably need to make compromises on time to ensure that you finish on time, and you will definitely get questions wrong.  So do both of those things on your own terms – spend the extra few seconds making sure that you get the questions right that you can get right, and burn a handful of questions as needed to buy yourself that necessary time.

How much time should you spend on each question?  As much time as it takes to answer those that you can answer correctly, correctly; and a few seconds each time you realize that it’s time to save time.