Is it possible to get more than 50 in quantitative part?

Same wonder as Becoolja

The best quant score I ve ever seen is a 51 and it was on this site a couple of weeks ago. I was thinking the raw score rated is out of 60 but the final score was out of 51... maybe I was wrong.

Did someone encounter any ET with lets say 55 in Q ?

Hi,

It is certainly rare to see Q score above 50, only about 1% of takers are in the 51-60 range. For technical reasons, one is likely to encounter a number of scaling problems in the 51-60 range but I doubt anyone wants to hear Hjort's discourse on the matter.

Hjort

Hi

I got 51 in Q but i am not sure if its possible to get 55 or more. In the powerprep tests even when i got all the 30 questions right my Q score was 51. And i also found the quants scale very sensitive. One question wrong and it goes to 50 which is 95 percentile and so on.....

Same goes with Verbal - they say that the score is from 0 - 60 scale but I have yet to come across a verbal above 48. I came across a 790 with 47 in verbal but never a verbal above 48. There was a 800 GMAT with 51 Q and 47 verbal.

Perhaps Hjort would have statistics of people getting 48+ in verbal and/or 52+ in Quants.

About getting 51 in Quants - I believe that if one practices every test with an aim to get 51 and make a record of all the silly mistakes he/she makes - 51 is gettable. There is a tendency to overlook silly mistakes one commits in practice tests.
The other aspect I found helpful was improving my DS section and that comes only with practice and the strategy of AD vs BCE.

In my actual test the toughest questions were the DS ones...the problem solving ones were fairly straightforward with as many as 5 questions on exponentials which i solved in about 10-15 seconds each...

Just in my idea, though you annswer all correctly, you will not get 60 if you fully spend 75 minutes. Maybe, you should have 10 minutes left to be qualifed for full score. If both A and B answer all questions correctly but A spends 10 minutes less than B, is it fair to give both of them full score?

Hi,

That's an interesting theory but I'm afraid the reason is far more mundane.

Hjort

Hjort..
I am getting curious..what is the mundane reason?

In actual test I finished my quants 5 minutes earlier..in one of the powerprep test i finished my quants about 10 minutes earlier but both time i got 51 in quants..
Unfortunately in the second powerprep test i got 2 questions wrong....

In any case is it that the 51 is a raw score and the actual score is different?

Why is it that verbal 47+ is 99 percentile and one never comes across 50+ in verbal...Is it more to do with statistics than with the actual score...?

Note that GMAC states that "the scores from the verbal and quantitative multiple choice sections . . . are comparable to scores obtained at any previous GMAT administration" and that they are reported on a fixed scale. Thus, the 51 is comparable across administrations and is "higher" than a 48 on another administation.

Hjort

I'd love to hear the reason... I know this thread is extremely old, but I'm really curious why the effective cap for section scores appears to be 51.

My theory was the same as above - I thought maybe the CAT was restricting the difficulty of questions for top test-takers based on their expected finish time. I.e., if a "top test-taker" (able to score 51+) is answering questions correctly but moving too slowly on more difficult questions (implied level 52+), the CAT chooses not to give the harder questions, thus settling at a lower level due to the implied time constraints.

It seems unfair for the test to cause someone to run out of time while answering 52+ level questions if the tester is able to finish on time while answering 51-level questions. Everyone talks about how the CAT knows the actual % of correct/incorrect answer for each question and uses that to determine score (e.g., rhyme's informative post at gmatclub[dot]com/forum/gmat-scoring-algorithm-my-observations-28493-60.html#p204280), but I have not seen discussion about how the CAT accounts for average response times for each question - and it seems likely that the CAT takes this information into account (or, at the least, that it could and should take this information into account). I guess I am reaching the simplified hypothesis that no 52+ level questions exist with average response times <3mins, and the CAT avoids these questions if a test-taker's time remaining is less than 3mins/question (where "3mins" is an approximation and I am sure the coding of this into the CAT's logic is actually quite a bit more complicated than my simplification).

In my case, I remember taking quite a bit of time on a particularly difficult question towards the end of the quant section on my test (I believe it was question #32). I spent ~5mins and believe I got the question right, but I definitely ended up behind the clock, I rushed through the last five questions (where were relatively easy and took ~1.5mins each), and I finished with 3 or 4mins to spare and a quant score of 51. But I also felt like I could have scored higher than a 51 (given a little extra time or one more 52-level question), and I'm curious if anyone has evidence (or suspicions) that the time remaining affects the CAT's decisions about which questions to give you.

Anyways, just curious. I'm no expert in standardized test creation (or item response theory) - just trying to understand why so many bright people appear to score 51 instead of 52+.

Would love to hear from you all - particularly any 99%ers (looking at you NonYankee, kevincan, and everyone else at gmatclub[dot]com/forum/99-ile-clubbers-add-your-name-to-the-list-77041.html) or test experts (rhyme, Hjort, any all others).

Thanks!

The official answer is that 51 is currently the max. They are reserving 52-60 for future tests. For example, if everyone magically got smarter, they want to be able to adjust the test and give people scores commensurate with their performance on a harder test.