Does this GMATPREP Score make sense?

Question

Hi All,

After few months of studying, I took GMATPREP Exam 1 today, and even though I scored fairly well as per the algorithm, I wanted to get the community's thoughts on my Quant score. Specifically, I am very surprised to see that I managed to score  Q50 despite getting 7 questions wrong . I know the difficulty level of the missed questions matters just as much, so here is a high level breakdown of my answers. Around question 20, I noticed that i was running behind and decided to "throw away" some questions (I think Questions 25 and 27 were pure guesses) to catch up with the clock. I had heard that to get to Q50, you can only afford to get 3-5 questions wrong, so I was expecting a Q49 score or so. The reason I am trying to get sense of whether algorithm is working fine is that I don't want to read too much into my scores if they aren't correct/accurate.

Overall: 24 correct, 7 incorrect
Breakdown of Incorrect answers along with the stats from GMATClub Timer in brackets
 #3: incorrect  
#21: incorrect  
#24: incorrect  
#25: incorrect  
#27: incorrect  
#28: incorrect  
#30: incorrect

Any help is appreciated!

unraveled · Answer

It is said that you can get GMAT score in +/- 30 range to what you get in GMAT prep tests.
With 7 incorrect you got Q50. This suggests that you did well in higher level questions unlike what i got, Q48. However, nowadays, test takers are saying that actual test is harder(may be test pressure/stress) but more important is Q50 is possible with 7 mistakes. Q51 is possible with 1-2 incorrect answer and Q50 with 3 but the possibility is much on the lower side.

Also, you were almost perfect until 20th question though may have taken more time. This suggests of starting well in test. The third one might have been harder one.

Toffelfabriken · Answer

Hi,

I've had similar experiences with the official mock tests, but considering that they use the official scoring algorithm, it should be nothing to worry about. I too got a Q50 on my first official mock, with a 770 total score. You can see breakdown below.

It seems to simply be the case that you are not punished very hard for messing up on a few difficult questions.

https://i.imgur.com/cRLqexN.png

IanStewart · Answer

There's nothing at all unusual about getting a Q50 with 7 wrong answers. I recall seeing a Q50 on the old GMATPrep tests (when the Quant section had 37 questions) when someone had 13 wrong answers, so you can score almost at the top of the scale even when you get roughly 1/3 of your questions wrong. But the only way to do that is if you answer all (or almost all) of the easy, medium, and harder questions right, and only (or almost only) get the absolute hardest questions wrong. If the test is persuaded you're very strong, based on your early performance, it will ordinarily try to feed you a lot very hard questions to decide if you're a Q49, Q50 or Q51 level test taker. That's what happened in your case, and really the only thing your wrong answers established (to the algorithm) was that you probably weren't a Q51 level test taker. The one unusual thing about your score is that it seems you got one very easy question wrong, and that can be very hard to recover from, but your performance on the rest of your test was so strong, you were able to compensate for that. You should pay careful attention to that mistake in particular (moreso than the others), because had you done that twice on your test, there'd be almost no chance you could get a Q50.

It is true on some of the official practice exams (the ones you buy) that the question bank is small, and you can run out of very hard questions. So on those tests, because most of the questions can be on the easier side, you have much less latitude to make mistakes. So if you read somewhere that you can't afford more than 3-5 mistakes to get a Q50, that might be true on those specific tests. But it's not at all likely to be true on your real GMAT (or on the free online tests) which have bigger question banks and a bigger supply of very high-level questions.

Q50 is a great score by the way!

It is true on some of the official practice exams (the ones you buy) that the question bank is small, and you can run out of very hard questions. So on those tests, because most of the questions can be on the easier side, you have much less latitude to make mistakes. So if you read somewhere that you can't afford more than 3-5 mistakes to get a Q50, that might be true on those specific tests. But it's not at all likely to be true on your real GMAT (or on the free online tests) which have bigger question banks and a bigger supply of very high-level questions.

Q50 is a great score by the way!

lemonmelon · Answer

Thanks. Interesting that you got a different score with as many incorrect answers. Does the breakdown of your incorrect answers look much different than mine?

Posted from my mobile device

lemonmelon · Answer

Indeed, this is almost similar to mine. In fact, you actually got few more incorrect than I did which does surprise me a little bit. Like you said, maybe it doesn't seem to punish as hard for missing difficult questions towards the end, but i still find it a little perplexing

lemonmelon · Answer

Thanks IanStewart for such a thorough post. It is encouraging to see that other people have had this scenario in which # incorrect questions was larger than typical despite getting a solid score. You are right that I should make sure to avoid getting easy questions wrong - I should devise a better strategy of "throwing away" questions when i am running behind the clock.

Re: Quant on actual GMAT being higher, I have seen that theme in a lot of debriefs that I have read. Have you heard something similar from your students?

IanStewart · Answer

In general, I'd recommend giving each question a chance - as you can see on your test, there are sometimes easy questions late in the test that do count. You don't want to be guessing randomly at those, because the consequences can be severe. So I'd suggest you read each question at a minimum, and if you know you can answer it right away, it's probably not one of the hardest questions, so you'll want to answer correctly - solve those questions, even when you're behind on the clock. If you truly need to catch up on time, and you don't know what to do soon after reading a problem, then it's probably a very high-level question, and the consequences of a guess will not be severe at all. So if you must guess, guess at those questions. Probabilistically speaking, the last questions on your test are quite likely to be near your level, so if you did end up obligated to guess at those, it's not too bad (and that would be better than not finishing the test, at your scoring level). But it's better to pick good spots to guess, and those are the questions you can be nearly certain are very hard.

It's generally also always worthwhile at least reading DS questions, because you can usually do better than a random guess -- often it's immediately clear one statement is insufficient alone, for example, and then you can guess from three choices instead of five, which is a meaningful difference.

Yes, but I've been hearing that for years. It's the nature of adaptive testing -- an adaptive test should seem hard to almost everybody, no matter what their level. An adaptive test gets the most information about you by giving you questions near your level. Questions at your level you only know how to answer half the time (you'll get them right about 60% of the time, because you still sometimes guess correctly when you don't know how to answer) -- that's essentially the definition of question difficulty the algorithm uses. So if the test thinks you're a Q50 level test taker, after a while it's going to start giving you a lot of questions around the Q50 level, and you're only going to see how to solve about half of them. If you could solve more, you wouldn't be a Q50 level test taker -- you'd be a Q51 (technically you might be a Q50.4 or something and still get a Q50 because of rounding, so this isn't strictly right for Q50-level test takers, but it would be for anyone in the middle of the scoring range). But any test where you only see how to solve half the questions is probably going to seem difficult!

There are some exceptions to the above -- the test might seem easy or medium level rather than hard in these cases:

- someone above the Q51 level might not find too much of the math hard;
- someone who gets off to a bad start on the test (say because of some careless mistakes on questions they'd normally get right) will start to see questions well below their level. Then the test might seem easy until the person answers a lot of those questions correctly and the difficulty level rises again (it's a long enough test that most of the time you can recover from a bad start). Notice though in this case, the fact that the test doesn't seem hard is a bad sign, not a good one -- it happens because you're doing badly and therefore getting easy questions;
- a test may not seem hard, no matter how well you're doing, if the underlying database of questions from which the test is drawing does not contain very many hard questions. That is true of some of the official practice tests (some of the ones you can buy), but should only rarely be true on the real test.

Whether an adaptive test is easy or hard (i.e. has a lot of easy questions, or a lot of hard questions) does not affect your score at all, incidentally (so it's not worth even thinking about mid-test). When a test is easy, you need to get a lot of questions right to get a good score. When a test is hard, you don't need to get nearly as many questions right to get the same score. Either test will generally produce a Q50 score (with some variance) though for a Q50-level test taker, even though that test taker will have a dramatically different 'hit rate' on them. More often than not, it's actually a very good sign when your test seems hard, because it usually means you haven't made any careless mistakes on easier problems, and the test thinks you're very good, so if that does happen to you on test day, it's something to be happy about, and not something to be concerned about.

I should add as well that it will hurt you as much to get an easy question wrong if that is question #5 on your test as when it's question #30, everything else being equal. There are a lot of myths about the importance of early questions and later questions, and while that's a bit of a long topic to discuss properly (I've done that elsewhere so won't again here), you shouldn't take seriously anyone making claims about that kind of thing. If you know how to answer a question, you should pretty much always answer it, and if you don't, you normally shouldn't invest too much time before deciding to move on.

Toffelfabriken · Answer

For what it's worth, here's the strategy I used for recovering from being low on time. I assumed every question would take ~2 minutes on average to complete. Every two or three questions, I'd glance at the number of questions remaining, multiply it by 2 and compare it to the time remaining. If the time remaining was significantly lower than the time required, I would do the following.

I would try to find a question that I was fairly certain I had the correct answer to, and then immediately guess on the next question and move on. This way, I would save 2 minutes, allowing me to catch up. Also, after having given a correct answer, there would be a lower probability of getting an easy question, so most likely I would be guessing on a more difficult question, leading to a lower penalty.

unraveled · Answer

See how silly mistakes i made. #3 65% difficulty, #8 45% difficulty, #11 15% difficulty, #14 35% difficulty ... 25,26 and 27. I had only one mistake of 85% difficulty wherein you had 3 of that difficulty or more.
You can imagine how dumb i get in actual test and that's i said you may have got harder ones wrong - Q50; I got easier ones, as you can see, wrong - Q48.

Another difference is i got three consecutive wrong which, looks like, punishes more harshly than getting only two consecutive wrong. All i remember that i felt test is throwing easy questions which later on reviewing i understood i was getting wrong too.

Now, i understand why such a penalty. Btw can you share how many 700 level(as per Gmatclub) questions you got overall.

lemonmelon · Answer

Thanks Ian! Lot of good nuggets in this extremely detailed and thorough post. Exactly what I was hoping to get from knowledgeable folks on this forum.

I agree with y’all’s suggestion around guessing strategy. I definitely look to improve it for the upcoming mocks. Even if I am not running behind the clocks on a practice test, I am planning to use this method as if it were a real exam and I was forced to make such choices. I think that’s definitely one way to be more realistic in the prep

Posted from my mobile device

lemonmelon · Answer

Yeah this makes sense and consistent with other posters above. I don’t have that info handy with me but when I do, I will post it here.

Thanks

Posted from my mobile device

ScottTargetTestPrep · Answer

Hi lemonmelon,

I see there are many great answers here. At the end of the day, I think the good news from all of this is that you pulled off an awesome quant score on your first practice exam. I'd be interested on how you perform on the rest of the GMAC practice exams. When do you plan to take your next one?

lemonmelon · Answer

Thanks  ScottTargetTestPrep

I took my Practice Exam 2 today, and achieved a similar score but with slightly different distribution of errors. I will be doing a more detailed analysis of my correct/incorrect answers, but below is the screenshot of the performance. I am pretty pissed at myself for making the mistake on q#5. Just as I clicked on the confirm answer button, I realized the logic mistake i made in that DS question. I need to be better at avoiding such mistakes during the real exam.

Re: pacing, even though I had to rush through 2-3 questions towards the end, I did feel slightly better than last time. After I realized my mistake on q#5, I also caught myself nearly making similar silly mistakes on couple other questions and at that point, I started spending time double checking every answer. My goal was to ensure as few mistakes as I possibly could in the first 20 questions, and to work faster on the remaining ones.

I do realize it is a good score but I am still taking all of it with a grain of salt and assuming that I may be at Q48-49 in real exam especially if it is true that the Quant on real test has evolved to be more challenging. We shall see....

ScottTargetTestPrep · Answer

I think this is a great sign! Let me know how the other practice tests go.

	Magoosh increases your score, guaranteed - 15% off ALL Magoosh GMAT plans with code GMATCLUB15 + Free GMAT Club Tests.
	Manhattan Prep: $79 Bonus Score Higher on Your GMAT Exam. Guaranteed.
	Target Test Prep - Save up to $325 $200 Discount + $79 Bonus GMAT Club Tests
	e-GMAT: 60% Off on GMAT Focus Get to 735+ from any starting point
	Experts' Global - ~~$1720~~ $960 End-to-End GMAT Prep + App Support: 150+ Videos, 2000+ Questions, 15 FLTs +
	GMATWhiz - $299 ~~$499~~ Experience the course through a free trial.
	GMAT Club Tests - Free Included with every course purchase of $149+ View all deals

Does this GMATPREP Score make sense?

Prep Toolkit

Top 5 GMAT Debrief Videos

615 to 715 in 15 Days (Ria's Strategies)

More than Quant/Verbal Abilities: Speed vs. Accuracy

745 with Only Self-Prep and GMAT Club

From 555 to 765: 210-point Improvement

Perfect 805 Score Debrief by Julia

My Rewards