7 reasons why your practice test scores don’t match your actual GMAT scores

 

One of the most painful things in the GMAT world is a massive test-day letdown. If you spend time on any of the GMAT forums, you’ll see tons of anguished posts that share a similar trait: a huge discrepancy between test-takers’ practice test scores and their actual GMAT scores.

In the geeky spirit of GMAT CR, our goal in this article is to help you resolve that discrepancy. So here are seven reasons why your test-day scores might be lower than your practice test scores:

Reason #1: you’ve been taking inaccurate, non-official tests

If you’re a regular reader of our little GMAT blog, you’ve heard this story before: the GMAT spends somewhere between $1500 and $3000 developing every official test question, and even the best test-prep companies can’t possibly compete with that.

Of course, it’s even harder for test-prep companies to combine those (inevitably somewhat flawed) questions into a realistic practice test. For example, test-prep companies struggle to mimic the GMAT’s use of experimental questions, or the exact mix of, say, geometry and probability questions.

To make things worse, if you’ve relied heavily on the materials written by a particular test-prep company, then you’ll probably do disproportionately well on that company’s practice exams. It stands to reason that the methods taught by Company X would be more effective on the questions written by that same company.

Sure, some of the higher-quality “knockoff” tests can still be good practice, at least for quant. But you’ll always want to take the scores with a huge grain of salt, and if you’re relying particularly heavily on one company’s GMAT practice tests, then you might want to be especially skeptical of those results.

Reason #2: you’re repeating the GMATPrep practice tests

In a perfect world, we’d all rely exclusively on official GMATPrep tests. The problem, of course, is that the GMATPrep software only offers six adaptive exams, and that might not be enough for you. (Check out this article for an inexpensive way to stretch your supply of semi-official tests.)

Plenty of people choose to retake the GMATPrep exams, and that’s not a bad idea: you’ll always learn something from the GMATPrep tests, and you’ll always see at least some new questions. The only problem is that you’ll also see some familiar questions, and that will bias your score upwards.

Whenever I say that, I hear the same objection: “Yeah, but I don’t really remember the questions, so the scores are probably accurate, right?” Sorry, but no. Even if you don’t consciously remember the questions, you’ll be able to answer them much, much faster if you’ve seen them before. Try reading a novel that you read 10 years ago, but don’t consciously remember. I promise that you’ll read it much faster – and you’ll absorb much more detail – than you did the first time around.

The bottom line: even a few repeated questions can completely change your experience on the GMATPrep tests, because you’ll feel less time-pressured, and your score will certainly inflate at least a little bit.

So please be really, really thoughtful about how you use those GMATPrep tests. If you need to repeat them, that’s OK – but don’t trick yourself into thinking that your scores on repeated GMATPrep tests are accurate. Because they probably aren’t.

Reason #3: you’re repeating the GMATPrep questions

This one is sneaky: if you’re a regular consumer of questions from the GMAT forums or other free sources, there’s a very good chance that you’re inflating your GMAT scores every day, without even noticing.

The internet is filled with official GMAT questions from all sorts of sources – including the GMATPrep exams. And even if you don’t repeat the GMATPrep tests themselves, your scores will also be inflated if you’ve seen the questions somewhere online.

So if you’re using one of the GMAT forums such as GMAT Club (which features two stunningly beautiful competent verbal forum experts), keep a close eye on the question tags. If you notice that a question comes from the GMATPrep tests, you might want to skip it, unless you’ve already thoroughly exhausted the GMATPrep exams.

Reason #4: all test-prep materials are based on GMATPrep questions

You’re not going to like this one.

When test-prep companies develop their own resources – guides, practice tests, practice questions – we have to draw inspiration from official GMAT sources. And since the GMATPrep tests are the closest thing to actual GMAT exams, we have to rely particularly heavily on the GMATPrep tests.

So there’s no way around it: every test-prep company writes their own “knockoffs” of GMATPrep questions. The best GMAT test-prep companies will artfully make their versions look drastically different from the originals; the lazier firms will just swap out a few details.

So if you’re ever had the feeling that the actual GMAT exam “feels weird” compared to the GMATPrep tests, this is one potential reason: you’ve seen tons of questions that resemble the GMATPrep questions, before you ever even touch the GMATPrep software. If this is the case for you, the GMAT questions on your actual exam might feel strange – and your score might drop as a result.

Reason #5: you’re fundamentally inconsistent

You won’t like this story very much, either.

I recently received an inquiry from a long-suffering GMAT test-taker who couldn’t figure out why his actual GMAT scores — ranging from 580 to 640 on six attempts — were so much lower than his “best GMATPrep test score of 720.”

Well, here are the scores from his first attempt at each of the GMATPrep exams:

  • GMATPrep #1: 620
  • GMATPrep #2: 720
  • GMATPrep #3: 580
  • GMATPrep #4: 660
  • GMATPrep #5: 590
  • GMATPrep #6: 600

When you average these six scores, you get 630 – and that’s really close to his top score on the actual exam.

So there are two takeaways here. First, don’t trick yourself into thinking that your best score is somehow your “real score.” Based on his GMATPrep scores, this test-taker should have expected to score in the low 600s – and that’s pretty much what happened. Praying for “a good test day” is never a winning strategy on the GMAT, and that’s basically what he was doing, without realizing it.

More importantly, these test scores clearly belong to a wildly inconsistent test-taker, and that’s a huge problem. The key to the GMAT isn’t cramming tons of knowledge into your head; it’s figuring out how to apply CONSISTENT techniques and reasoning habits every single time you see a particular type of question.

If your GMATPrep test scores are all over the place, they’re trying to tell you something: you fundamentally lack consistency and you’re applying different techniques at different moments, depending on your mood or the direction of the wind. Until your performance is consistent, there’s no reason to expect a great GMAT score – and it shouldn’t surprise you if your test-day performance is much lower than your best GMATPrep practice tests.

So be honest with yourself: until your practice results are CONSISTENTLY in your target range, it’s unlikely that you’ll earn the score you want on the actual GMAT.

Reason #6: you’re fatigued on test day

In a perfect world, you’ll feel exactly the same in the GMAT exam room as you do when you’re taking a GMATPrep exam in the comfort of your own home. But that’s not realistic: you’ll likely be amped on your test day, and your adrenaline will probably be flowing.

And maybe you’ll run out of gas as a result of all of that excitement. Every once in a while, I hear from somebody who simply got fatigued at the end of their exam – particularly if they chose the “conventional” section order, with verbal at the very end.

If fatigue was an issue for you on test day, maybe you need to do more full practice tests, or it might help to practice with some extra-hard LSAT materials, just to build up your stamina. Or maybe more sleep or a better test-day diet would help. Or you could always choose a different section order, so that the least-important sections come at the end of your exam. But test-day fatigue is definitely a very fixable problem.

Reason #7: you’re nervous

This is the elephant in the room: most people get a little bit nervous when they’re taking a high-stakes test like the GMAT. And some people – perhaps around 20%, according to studies conducted here in the United States – get so nervous that their cognitive functions are impaired.

A full discussion of test anxiety could easily fill an entire book, so I won’t say much about it here. But if your GMAT scores suddenly drop on test day, there’s a pretty good chance that nerves are playing a role – even if you aren’t consciously jittery.

If this is the case for you, you might consider trying some of the techniques mentioned in this article or this article or this article. But whatever you do, don’t sweep your test-day anxiety under the rug, and pretend that it didn’t happen. Be honest with yourself – or else you’ll have zero chance of overcoming the problem. 

Still not sure what happened? Get your ESR.

I’m a little bit cynical about the motivations behind the GMAT’s ever-increasing menu of products, but if you’re not sure why your real GMAT score was so low, the GMAT Enhanced Score Report (ESR) is probably worth the price ($24.95). And even if you’ve already canceled the score, GMAC will happily sell you an ESR.

The ESR can’t tell you everything, but it can give you some idea of what happened. For example, the ESR might tell you that you made a bunch of mistakes on easy questions, or that your time management was erratic. Or maybe it’ll tell you that you had a meltdown on one specific question type.

Of course, the ESR can’t tell you why you made those mistakes. But it can at least point you in the right direction if you’re not sure which of our seven reasons explains why your score dropped on test day.

And if you need advice, feel free to leave a comment below, and we’ll do our best to help.

 

Why Adderall, Ritalin, and other PEDs won’t help your GMAT score

 

If you attended an undergraduate program in the United States, you probably know that academic performance-enhancing drugs (let’s call them academic PEDs for short) are a mainstay among university students. I know dozens of students who have popped a few pills – typically ADD medications such as Adderall or Ritalin, obtained through a friend, a law-bending physician, or a black-market dealer – to help them focus while studying.

These drugs are basically legal versions of amphetamines (though their “off-label” use without a prescription is, of course, illegal) that can provide energy and help you hyper-focus while you cram for a test or write a paper. Sounds great, right?

(And just to be clear: I’m NOT talking about anybody who has an actual prescription because of a diagnosed medical condition. I’m referring to off-label use only – and there are plenty of reasons why such use is dangerous, but that’s another story for a different, non-GMAT blog.)

Anyway, I’m not here to moralize about drug use – after all, I guzzled many gallons of performance-enhancing bourbon during my years as a bartender. I’m just here to offer an honest answer to a frequently asked GMAT-related question: can academic PEDs like Adderall and Ritalin help you conquer the GMAT and GRE exams, assuming that you’re willing to ignore the potential health and legal consequences?

The short answer: I don’t think so.

First of all, the whole idea behind academic PEDs is that they help you hyper-focus. Imagine, for example, that you need to spend 14 hours memorizing organic chemistry formulas. Stimulants can help, at least in the short run. Gotta grind your way through some repetitive calculus problems? Well, that takes some short-term energy and focus, and academic PEDs might help with that, too.

But hyper-focusing is actually a bad thing if you’re trying to improve your GMAT score. The GMAT – especially the quant section – requires you to think logically through different solution paths, and then choose the best option. In general, if you’re hopped up on amphetamines, you’ll speedily pick the first solution path that comes to mind – even if it’s an inefficient path or a complete dead-end. Basically, academic PEDs cause cognitive tunnel-vision – and that’s a good way to ruin your GMAT score.

The second reason why I’m skeptical of academic PEDs is that a few of my students have used them while studying for the GMAT, and they haven’t had much success. In every case, my GMAT students were accustomed to taking Ritalin, Adderall, or a similar drug during their undergraduate studies – and in every case, the academic PEDs didn’t seem to help their GMAT prep. They would stubbornly obsess over unnecessary details of GMAT RC passages, crunch through dozens of lines of algebra when quicker solutions were available, and make score-destroying careless errors because they were moving too quickly.

In a different context, maybe the academic PEDs would have helped these very same students. But the GMAT requires a flexible mindset and a solid approach to time management. Drugs that cause you to energetically develop tunnel vision are exactly the wrong prescription for success on the GMAT.

So if you’re looking for quick fixes – well, you probably already know that you’ve come to the wrong GMAT tutor’s blog if you’re looking for gimmicky shortcuts. To be honest, if I thought that academic PEDs could help your GMAT score, I would (probably very quietly) admit it. But the truth is that they don’t seem to do much to improve your GMAT or GRE scores.

I do, however, have plenty to say about another GMAT performance-enhancing substance: food.

Go ahead, roll your eyes… but it’s GMAT quantitative reasoning, not GMAT math

 

If you’ve struggled with the GMAT quant section more than you think you should, this blog post is for you. If you’ve ever said, “I’ve always been a really good math student, but I can’t understand why the GMAT quant is so hard for me!” – then this post is definitely for you.

In my former life as a PhD student, I spent far too much time learning about the statistical science underneath standardized testing, known as psychometrics. My nearly three-year PhD odyssey didn’t result in much other than some grey hairs and a “thank you for playing!” Master’s degree in education, but I did experience a few things that probably helped me become a better GMAT and GRE tutor.

This is a story about one of those things – and at the time, I never would have guessed that it would be useful for my GMAT and GRE students.

In my first year as a PhD student, I went to a psychometrics conference, populated by academics and employees of major standardized testing organizations, including ETS (makers of the GRE and TOEFL) and GMAC (creators of your beloved GMAT). I attended a presentation by a high-ranking GMAT psychometrician, who discussed… well, nevermind that part, I’ll put you right to sleep if I start talking about it.

Anyway, here’s the useful bit: somebody in the audience asked a question about the “math section of the GMAT.” The GMAT psychometrician interrupted him politely: “Excuse me,” he said, “there is no math section on the GMAT. There’s only quantitative reasoning.”

I probably rolled my eyes. “What a dick,” I thought, “why would he make a big deal out of that? It’s math. S#!t, I’ve been teaching it for a decade. Whatever, dude.”

Sure, maybe the GMAT psychometrician wasn’t picking the best moment to make a big deal out of it, but he absolutely had a point. In the few years since I attended that conference, I’ve realized that my students – particularly Americans – actually perform better on the quant section of the GMAT when they stop thinking of it as “math” and start thinking of it as “quantitative reasoning.”

Here’s the thing: in the United States, “math” knowledge – at least through the high school level – is typically taught as sequences of mechanical steps that you need to memorize and follow. Throughout much of my public school education, our daily homework would consist of 10 or 20 nearly identical math problems. The problems were usually so similar that there was no reason to think about what any of it meant. If you could follow instructions, you’d get an A – even if you had absolutely zero understanding of the underlying mathematical concepts.

As a result, most Americans think that the word “math” just refers to a boring series of steps that you follow. Sadly, we don’t think of mathematics as a way of thinking, or as a set of useful tools for reasoning our way through useful problems. There are, of course, plenty of exceptions, but the overwhelming majority of Americans have learned math in a way that strips it of its logic, meaning, and intuition.

So it’s no surprise that I hear this over and over from GMAT test-takers, especially Americans: “I’m a really good math student, but I can’t understand why the GMAT is so hard for me!”

Obviously, there are a ton of reasons why somebody might struggle with the GMAT quant section, but plenty of GMAT test-takers make the subtle mistake of trying to learn too many formulas, memorize too many steps, and drill too many mechanical aspects of mathematics. The GMAT, for all of its flaws, brilliantly twists 10th-grade math into a hard-to-penetrate – or at least a hard-to-quickly-penetrate – tangle of logic.

In other words: if you’re trying to blindly apply mechanical techniques to GMAT quant questions above the 500 or 600 level, the exam will eat you for breakfast.

Let’s look at an example (with apologies for the blurry fractions):

Which of the following is greatest?

GMAT blog example quantitative reasoning

If you think of this as a mechanical “math question,” you’ll follow some well-worn steps here: find some common denominators, add the fractions, and THEN compare the sums.

Go ahead and try it if you’d like. If you can correctly solve the question that way in two minutes or less, I’ll give you a cookie.

But if you’re thinking of the GMAT as “quantitative reasoning” – with or without the eye-roll – then maybe you’ll try something quicker, smarter, and less arithmetic-intensive. In this case, we’re just looking for the greatest value – and we don’t care what that value actually is, as long as we know that it’s larger than the other four answer choices.

So since the question is just asking for the greatest of the five answer choices, you can just compare pairs of answer choices, and knock off anything that’s the smaller of the two. Let’s start with D and E. It’s easy to see why E is larger than D once you notice that 1 – ½ = ½, so D is gone.

Similarly, B looks a lot like E, except that the denominators in B are squared – and since larger denominators mean that the fractions must be smaller, we can cross off B. The same argument holds for C – it’s clearly smaller than E as well. And then A has smaller denominators than E – so A is your answer.

No computation required, right? If you’re approaching this wisely, you barely need to lift your pen.

So if you’re thinking of the GMAT quant section as a set of narrow mathematical tasks – formulas that need to be memorized, or boring-ass steps that need to be followed – then you’re barking up the wrong tree, at least if you want an elite GMAT quant score. Once you start looking for opportunities to apply flexible logic and identify multiple solution paths, then you’re on the right track.

If any of this strikes a nerve, then it might not be a bad idea to stop yourself whenever you start thinking about the GMAT “math section.” Roll your eyes at yourself if you’d like, but thinking of the GMAT quant section as “quantitative reasoning” might help you embrace the flexibility and logic you’ll need for a top GMAT quant score.

An unfortunate story about GMAT testing irregularities

 

I’m reasonably confident that the GMAT does its honest best to ensure that the GMAT testing experience is the same for every GMAT test-taker. But unfortunately, stuff happens in life – even in the tightly controlled environment of a GMAT testing center. I documented a few minor GMAT testing center horror stories on my GMAT blog a bunch of years ago: basically, a few GMAT test-takers encountered computer glitches that torpedoed their tests, or had shorter breaks because the proctor failed to release them promptly from the testing room – that sort of thing.

Since then, my GMAT students have occasionally experienced computer failures, GMAT test-center power outages, and even a haunting or two — though the latter might have happened only in a GMAT student’s (literal) nightmares. In each of the real-world cases, the folks at the GMAT have done the right thing: they’ve at least offered a free retake of the GMAT.

Sadly, it sounds like the GMAT has modified its behavior toward testing irregularities lately, and that’s pretty frustrating. The circumstances of the latest GMAT testing center horror story aren’t terribly interesting: as one of my GMAT students returned from his post-Integrated Reasoning break, the proctor botched the password when he tried to log back into the computer. He apparently botched it several times, and my poor GMAT student – let’s call him Mr. B, since he apparently makes a mean bhindi masala – lost about two minutes from his GMAT quant section.

In theory, losing two minutes shouldn’t have affected Mr. B’s quant score much – after all, that’s only the equivalent of one GMAT question, right? But let’s be realistic: since we all know that you need every possible advantage on the GMAT these days, wouldn’t you freak out just a little bit if you lost two minutes from the test? And wouldn’t that affect your performance?

Unfortunately, Mr. B badly underperformed on his GMAT that day – particularly on his slightly shortened quant section – and he chose to cancel the score. We can’t completely blame his GMAT quant score on the proctor’s error, but it certainly didn’t help.

Regardless of the effects of the proctor’s error, it’s still true that the GMAT is supposed to be a standardized test. “Standardized” means that every test-taker should be given the same, fair testing experience. The GMAT quant section is 75 minutes – not 73 minutes. What happened to Mr. B is pretty much the textbook definition of a testing irregularity.

In this situation, the GMAT should have apologized, wiped the score off Mr. B’s score report, and offered him a free retake. After all, that’s what they’ve done in the past – plus, it just seems like the right thing to do. Mr. B didn’t have the same testing experience as his competitors for coveted MBA slots, and that’s not fair – even if the disadvantage was relatively small.

Mr. B is not a particularly shy man, and he pushed the GMAT for a fair resolution. There was a paper trail – the proctor filed a report indicating that there had been a testing irregularity. What did the GMAT do after several phone calls and a few emails? Nothing. They told Mr. B that he “didn’t have to cancel his score” and refused to offer a retake.

Again, it’s not the end of the world. Mr. B earns a good living, and the $250 he paid for his next GMAT exam did not bankrupt him. But it seems unfair – and it definitely represents a departure from the GMAT’s behavior in the past.

Here’s my rather cynical theory about the GMAT’s decision: perhaps due to increasing competition from the GRE exam, the number of GMAT exams has fallen recently, and the GMAT’s revenue has presumably fallen accordingly. But if each GMAT question costs somewhere between $1500 and $3000 to develop, the GMAT can’t really afford a loss of revenue.

So it looks like they’re making up that revenue however they can. Since 2012 – when the number of GMAT tests dropped – the GMAT has begun to offer plenty of pricey (and often very useful) products, as I mentioned in an earlier GRE vs. GMAT blog post: enhanced score reports ($24.95), the ability to cancel ($25) or “un-cancel” ($50) your score after test day, the worth-every-penny GMAT Exam Packs ($49.99 each), and new GMAT OGs, released every year instead of every few years ($85 for the latest bundle on the GMAC website).

And we can apparently add a new revenue source to the list: a refusal to compensate GMAT test-takers when minor testing irregularities occur.

I’m obviously not impressed by the GMAT’s behavior in this case, but if you’re just a reader of this little GMAT blog, the good news is that these sorts of things don’t happen very often. But if you’re truly paranoid about the GMAT’s increasing miserliness, then maybe you have another reason to think about taking the GRE instead of the GMAT?

How to get a perfect 800 score on the GMAT… sort of

 

More than five years ago, I earned a perfect 800 on the GMAT. I don’t really think of it as much of an achievement, to be honest – and I definitely don’t think that it’s an important qualification for the best GMAT tutors.

But I’ve been asked about it literally hundreds of times over the years – by GMAT students, prospective GMAT students, random people in the GMAT forums, strangers who’ve encountered my little GMAT blog, and plenty of others. So here are a few answers to the 800-related questions that keep coming my way.

Did you get every question right? 

No, I’m 99% sure that I didn’t. GMAT enhanced score reports didn’t exist back in 2011, so I can’t actually see if I missed any questions, but as soon as I finished the exam, I worked through every quant question I could remember – and I’m pretty certain that I missed at least one. Maybe more.

The bottom line, for whatever it’s worth: you can miss a bunch of questions and still get a “perfect” GMAT quant score of 51.

But you got every verbal question right? 

I think so. But it was mostly due to dumb luck.

In all of my previous GMAT exams – including the GMATPrep practice tests the first time I took each of them – I always fell short of a perfect verbal score. I inevitably missed a few questions, partly because I tend to lose focus at the end of the GMAT, but mostly because I screw stuff up sometimes, just like every other human.

And if you’ve ever taken the GMAT or the GRE or the LSAT, I’m sure that this part will sound familiar: I inevitably face a few “coin tosses” on verbal – questions where I’m down to two answer choices, but I’m not terribly confident in the final selection. (Incidentally, if you’re down to two choices on a GMAT verbal question and you select the wrong one, that’s not necessarily a sign that you were “close” – it’s usually a sign that you misread something in the passage.)

But when I finally scored an 800 on the GMAT, that didn’t happen at all – I was pretty much 100% confident on every verbal question. That has never happened to me on any GMAT, LSAT, or GRE exam before or since – including the day when I got a perfect GRE score. That test felt freaking brutal, and I was shocked that my GRE score wasn’t lower in the end.

Most importantly, all four of my GMAT reading comprehension passages were bizarrely interesting when I got that 800. I still remember one of them – it was about a type of plant called dodder that apparently has a sense of smell. Amazing. Again, that’s never happened before or since: when have you ever had four interesting GMAT reading comprehension passages on the same exam?

So there you go: yes, I’m pretty good at the GMAT, but those last 10 or 20 points were dumb luck – or measurement error, if you prefer the technical term.

Were you banned from taking the GMAT ever again? 

Yup. I received a nice letter from the GMAT Office of Test Security, informing me that I would need a damned good reason if I ever wanted to take it again. And I don’t have a damned good reason. “I want to help my GMAT students beat your stinking exam” isn’t going to fly with the GMAT test security folks.

My MBA.com account is suspended too, though the GMAT will still happily accept my money whenever I want to buy GMAT practice tests or the GMATPrep Question Pack from them.

Did you study for the GMAT before you got the perfect score?

Well, I earned the perfect score in 2011. I started working as a GRE and GMAT test-prep tutor in 2001, starting with a gig at a large test-prep company before I became an independent tutor a few years later. So in some sense, I “studied” for 10 years before I got a perfect GMAT score – and I’m still “studying,” since I work with GMAT students almost every day.

You probably don’t want to do that. Unless you want to become a GMAT tutor yourself, “studying” for more than a decade is an epic waste of your time.

Are there certain GMAT test-prep materials that would help somebody get a perfect GMAT score?

It’s funny, I read a GMAT blog post from another test-prep company that recommended its own materials for anybody who wants a perfect GMAT score. Frankly, that’s ridiculous, partly because knockoff GMAT materials can never be as good as official GMAT materials – and partly because that particular GMAT test-prep firm writes especially lousy materials, in my opinion.

So, no: other than the official GMAT materials, there aren’t magical GMAT test-prep resources that will get you to a perfect 800 on the GMAT.

And more importantly: there’s absolutely no practical reason for you to want a perfect GMAT score, anyway. An 800 will not help you get into a great business school, and I wouldn’t even argue that it’s a necessary quality for the best GMAT tutors.

So if you’re wondering how to get an 800 on the GMAT, don’t waste your time thinking about that. Go write an interesting MBA essay instead, or better still: go eat a tasty snack.