Why Practice Test Scores Vary : General GMAT Questions and Strategies

I recently took a practice test and it was much lower than my other practice tests. I’ve been studying for awhile now, and I am freaking out a little bit. Shouldn’t I be improving. Here are the scores from my practice tests.

Practice test 1: 580
Practice test 2: 620
Practice test 3: 590
Practice test 4: 550

What happened? Can you help me improve? I don’t want this to happen when I take the test.

This is a common question that we encounter at Magoosh—a student is taking practice tests and their scores are all over the place. Shouldn’t there be some trend, like going up, after lots of practice. But this is not the case. Why?

You may be surprised that statistical variation is common and happens for a number of reasons. Let’s explore some of the reasons for this variation.

Different Practice Tests
The most obvious reason for a score difference is using practice tests from a different companies. Each test prep company creates questions of varying quality. And this quality affects students scores. Poorly created questions, like you might see in a Princeton Review, will not give you an accurate prediction of your score. Whereas, a Manhattan practice test will have better questions with less mistakes, making it a more accurate assessment, but not a perfect assessment (I hear grumblings in forums about inflated verbal scores on Manhattan tests).

But it doesn’t end here. We also have to think about how test prep companies calculate your estimated score at the end of the test. No company knows exactly what the GAMC does to convert a raw score into a scaled score. As such, each company has to approximate this conversion, and each company does something slightly different. Since this is not standardized across practice tests, you can definitely expect to see variations in the scaled score from test to test.

If you want the best prediction of your score on the test, I recommend taking a practice test made by the same people who make the GMAT—GMAT Prep.

Different Testing Environment
Environments exert a powerful influence over us. If you take the same test twice, but change the testing environment, you can expect to see a difference in scores. The difference between sitting in your room and going to a private study room at the local library to take a practice test is dramatic. One spot is obviously more ideal than the other. And by “ideal,” I mean that it is more like the testing environment for the test. Did you vary your testing environment? Was your low score from the test you took while in a cafe? Look no further. You found the culprit then. Make sure that you emulate the testing environment that you will see on test day.

Different Diet
What you put, or don’t put, into your body matters. Think back to what you ate in the 24 hours leading up to your practice test. Was your belly full of sweets or junk food? Did you skip a meal? Did you indulge in delectable southern BBQ? That might have been the difference. To optimize your score and your ability to succeed, you need to eat well. And not just before the test, but all the time. You need to give your brain the food it needs. Eat fish, almonds, avocados, garlic, and blueberries.

Different Sleep Patterns
Erratic and irregular sleep patterns are detrimental to learning and performance. Researchers at UC Berkeley have shown that a consistent eight hours of sleep is an important part of learning. Not only will it boost your focus and performance, but it will also keep you positive and foster emotional health. Make it a point to sleep enough before your test. And if you want all those daytime hours of studying to really make a difference, you need to get plenty of sleep in the weeks leading up to the test.

Takeaway
Variation in practice test scores is normal and expected. Many factors affect the predicted score you receive on a practice test. Ultimately, don’t invest too much in the estimated scores. Remember that it is only a predication and not necessarily accurate. Use practice tests as a way to simulate the test, not as a untainted measure of your GMAT score.

More from Magoosh:
How to Calculate GMAT Scores
How Hard is the GMAT
What's a Good GMAT Score
How Long Should I Study for the GMAT
_________________

Hi Kevin,

I completely agree with you and thank you for sharing your insights. This is a common question we too receive.

We see that a score can dip from one exam to another even with the same amount of knowledge or the same percentage of correct/incorrect. The GMAT scoring algorithm takes many factors into account including where you make errors, how many mistakes you commit in a row, whether you finish both sections or leave questions out, whether the last few questions of either section seemed like they were guessed at, how you felt on test day, etc. These factors can all affect your score.

From one test to the next you can easily jump +/- 30 points (and even up to 80 or so) depending on where the mistakes are made. *Where* you make mistakes can often be more important than how many. Overall, when a student wants to know where he/she stands, the best way to do so is to take the average after a few sim tests. This should get you to a fairly close estimation of where you are at.

As for Economist GMAT Tutor, in general, our students tend to score within about +/- 30 points or so from what they get on our tests and the GMAT test preps.

Regards,

Elizabeth
_________________

Hi Elizabeth,

Thanks for adding your thoughts. Those are excellent points to consider that I didn't mention in my post! The GMAT scoring algorithm is way more complex than students often think. So it is important to get this information out there.

Cheers,

Kevin
_________________

Hi Kevin

I was wondering why you feel the Princeton Review questions are poorly created. I have take a few Princeton Review tests and thought the questions were fairly similar to the actual GMAT. I have also taken the MGMAT tests and found the quant questions were much harder than what was actually tested. I didn't find the GMATPREP practice tests nearly as hard as the MGMAT. My scores on the MGMAT tests were almost 100 points lower than what I scored on the Princeton Review and GMATPREP tests.

I'm going to be retaking the GMATs next month and I'm trying to find the best strategies that can help me boost my score. I'm looking to score in the 650-700 range.
I'd appreciate any advice/insight that you may have.

Thank you.

Elizabeth and Kevin, what an informative thread you two have started here! Students place a ton of emphasis on CAT scores, so it’s important to flesh out their knowledge about how scores are calculated and why they might vary. I'd like to add some thoughts:

Stick to MGMAT CATs
It’s helpful to avoid taking a variety of third party CATs. Their scoring algorithms and problem quality vary, so it is hard to compare the scores and gauge improvement across them. Obviously, the GMATPrep CATs are the best, and those should be used by all GMAT students. For third party CATs, MGMAT is a great choice as Kevin mentioned; they are widely regarded as the best. If a student takes only those CATs, then sub-scores and composite scores can be reasonably compared across them.

CAT Readiness
The concept of CAT readiness - how ready the student is to take a CAT - is crucial! As Kevin said, sleep, energy levels, test environment, diet, etc. really matter. Here is a key additional point: a student controls those variables. For every CAT, the student should focus on optimizing her CAT readiness. A belly full of junk food or testing in a noisy environment should NEVER happen. If the student takes the CAT seriously, then the variance in scores can be attributed to skill/ability changes, which is crucial for measuring how well the prep method is working overall.

Finish Each Section
That idea relates to some of your great points too, Elizabeth. Random variations in performance - such as missing several problems in a row - will affect scores, but that type of thing should average out over the course of a section. Not finishing a section definitely causes major score variance due to the penalty. This is why it is crucial for the student to prioritize effective time management and completing the Quant and Verbal sections.

Score Variation Takeaways
Having sufficient time and practice in between each CAT - along with good time management, managing CAT readiness, and consistent third party CATs - ensures that a student's score on a CAT reflects her current ability levels and doesn't vary due to variables under her control. That is when CATs become exceptional useful as measurements of prep quality. If scores go down when the important variables are controlled, then the student knows that prep adjustments or improvements must be made.

Thoughts for Amy
Amy, the Princeton Review tests are widely regarded as poor CATs. It is likely that your particular skill set is what made those tests feel more accurate for you than the MGMAT CATs.
As far as advice/insight on improving your score, I recommend GMAT Clarity. It’s an innovative guide to GMAT self-study I just published that has a lot of value for retakers. I'd be happy to help you use it to its fullest. It’s currently available for only $7 on amazon (soon to be $40)!

Hi Amy!

I think that Thomas nailed it! The Princeton Review has subpar questions that do not reflect the range of difficulty levels on the GMAT. If you are aiming for the 650-700 level then you really need to become comfortable with those hard and difficult questions on the MGMAT.

As for retaking the test and looking for improvement, I recommend reading this article that I wrote on how to study for a retake:

https://magoosh.com/gmat/2014/how-to-stu ... at-retake/

Also, I think that Thomas gave some great advice like being ready for the timed sections and aiming to finish each section. Much of what I have learned from the scoring algorithm on the GMAT is that you should aim to finish each section and aim to not miss a lot of questions in a row at the end of a section. You'd be better off guessing on really hard questions in the middle of the test and missing them than doing the same thing at the end of a section. You want to finish each section strong.

I hope this helps! You have a lot of great advice here! Let me know if you need anything else!
_________________

KevinRocci wrote:I recently took a practice test and it was much lower than my other practice tests. I’ve been studying for awhile now, and I am freaking out a little bit. Shouldn’t I be improving. Here are the scores from my practice tests.

Practice test 1: 580
Practice test 2: 620
Practice test 3: 590
Practice test 4: 550

What happened? Can you help me improve? I don’t want this to happen when I take the test.

This is a common question that we encounter at Magoosh—a student is taking practice tests and their scores are all over the place. Shouldn’t there be some trend, like going up, after lots of practice. But this is not the case. Why?

You may be surprised that statistical variation is common and happens for a number of reasons. Let’s explore some of the reasons for this variation.

Different Practice Tests
The most obvious reason for a score difference is using practice tests from a different companies. Each test prep company creates questions of varying quality. And this quality affects students scores. Poorly created questions, like you might see in a Princeton Review, will not give you an accurate prediction of your score. Whereas, a Manhattan practice test will have better questions with less mistakes, making it a more accurate assessment, but not a perfect assessment (I hear grumblings in forums about inflated verbal scores on Manhattan tests).

But it doesn’t end here. We also have to think about how test prep companies calculate your estimated score at the end of the test. No company knows exactly what the GAMC does to convert a raw score into a scaled score. As such, each company has to approximate this conversion, and each company does something slightly different. Since this is not standardized across practice tests, you can definitely expect to see variations in the scaled score from test to test.

If you want the best prediction of your score on the test, I recommend taking a practice test made by the same people who make the GMAT—GMAT Prep.

Different Testing Environment
Environments exert a powerful influence over us. If you take the same test twice, but change the testing environment, you can expect to see a difference in scores. The difference between sitting in your room and going to a private study room at the local library to take a practice test is dramatic. One spot is obviously more ideal than the other. And by “ideal,” I mean that it is more like the testing environment for the test. Did you vary your testing environment? Was your low score from the test you took while in a cafe? Look no further. You found the culprit then. Make sure that you emulate the testing environment that you will see on test day.

Different Diet
What you put, or don’t put, into your body matters. Think back to what you ate in the 24 hours leading up to your practice test. Was your belly full of sweets or junk food? Did you skip a meal? Did you indulge in delectable southern BBQ? That might have been the difference. To optimize your score and your ability to succeed, you need to eat well. And not just before the test, but all the time. You need to give your brain the food it needs. Eat fish, almonds, avocados, garlic, and blueberries.

Different Sleep Patterns
Erratic and irregular sleep patterns are detrimental to learning and performance. Researchers at UC Berkeley have shown that a consistent eight hours of sleep is an important part of learning. Not only will it boost your focus and performance, but it will also keep you positive and foster emotional health. Make it a point to sleep enough before your test. And if you want all those daytime hours of studying to really make a difference, you need to get plenty of sleep in the weeks leading up to the test.

Takeaway
Variation in practice test scores is normal and expected. Many factors affect the predicted score you receive on a practice test. Ultimately, don’t invest too much in the estimated scores. Remember that it is only a predication and not necessarily accurate. Use practice tests as a way to simulate the test, not as a untainted measure of your GMAT score.

More from Magoosh:
How to Calculate GMAT Scores
How Hard is the GMAT
What's a Good GMAT Score
How Long Should I Study for the GMAT

Hi Kevin,

I took a GMAT test yesterday from GMAT prep and my score was 530 and i was shocked

Because last week i took another test from GMAT prep and i got a 610

I dunno what i must gauge of this.

Im really disappointed because I have put in a lot of effort and its disappointing to see my score down by 80 points

Please help if u can

Thanks