I just Beat The GMAT!

I took the GMAT on 12/23 and got a 710, 44q/44v/6 AWA. A friend of mine happened to take the test 6 days later and get the exact same quant/verbal scaled scores but he got a 720. I've been searching everywhere to understand how this could happen and haven't been able to figure it out. It seems to me that the same scaled scores, within a week of each other, would result in the same overall score.

Has anyone seen this before? Can anyone shed any light on how this might happen? I have seen a few other people on beatthegmat who have received 720s with the 44/44 scaled breakdown back in October. I have also seen 44/45 scores that translated to 720. The percentile breakdowns for each of our scores (70th % Q, 97th% V) were exactly the same for both us us as well. I asked my MGMAT instructor (great course by the way) and she thought it was very strange as well. If we took the test a few years apart I would think that the range had changed as more people took the test, but I can't imagine that 6 days over the Christmas holiday would change the 3 year averages significantly enough to result in a different overall score.

I have called the GMAT customer service line twice now and both times, after explaining my situation and waiting on hold for about 20 minutes, I got the same response "we can't discuss the scoring, please email [email protected]. To me this sounds like a blackhole@ email address and I expect the response via email to be the same.

Someone suggested that the GMAT must have an ombudsperson to oversee this kind of thing but I don't know how to breach the customer service shield to get a response. I know they can't give me the scoring algorithm, but they don't even say "this can happen, it's normal, it is because xyz, etc" which makes it seem all the more fishy.

I can't complain too much but who wouldn't want the extra 10 points right?

Thanks for your help!

Here's a recap of my test prep:

Took Manhattan GMAT course, studied the MGMAT, OG, OG verbal, and OG quant books that came with the course. I did a few practice questions from a kaplan book as well but not many.

My diagnostic score was 530, mostly due to a terrible quant score.
I took 6 MGMAT Cat exams ranging from 650 to 700 on the last one about 2 weeks before the test. I knew that quant was the area I needed the most work on so I focused my studying almost entirely on re-learning the necessary math and familiarizing myself with the data sufficiency question type (which gave me lots of trouble). Strangely, as my quant scores increased quite a bit, my verbal scores actually went down through the first couple of exams.

With about 3 weeks to go I started to add verbal concepts to my study regimen and my scores began to improve.

I took the two practice tests from mba.com leading up to the test, getting 690 (44Q,42V I think this was the breakdown but don't remember exactly) on the first and 720 (48Q,40V) on the second. Throughout my preparation I always ran out of time on the quant section until the last practice test. I was hoping it was due to my warm-up regimen of quant problems and that I could recreate the same on the actual exam but unfortunately I still ended up moving more quickly through the last couple of questions than I would have liked.

I did feel more confident, and seemed to do better, on the quant sections when I had done a number of practice problems before hand. For this reason, I scheduled an afternoon test and did some problems in the morning ahead of the test. The sense of confidence may have been more of the cause than the actual problems, but either way that was my approach.

Hopefully that helps people who are getting ready for the test. If anyone knows anything about how my friend and I could have gotten different scores I would love to hear about it.

Thanks and good luck to everyone.

Based on my experience with the paper based tests, I could see how two people with the same scaled, section scores, could end up with different overall scores.

The scale score is obtained from a chart using your raw score. Raw scores of 42 and 43 on the quant section could convert to the same scaled score of say 44. Similarly, raw scores of 45 and 46 on the verbal could convert to a scaled of say 44. When it comes to calculating the overall score, the person getting raw scores of Q43,V46 could end up scoring as much as 20 pts higher than the person getting raw scores of Q42,V45, even though they both had the same scaled verbal and quant scores (Q44, V44).

I'm not sure exactly how pearson's algorithm determines overall scores, but I think the rationale should not be too different from what ETS used on the paper tests.

Hope this helps.

Thanks Ian,

I had no idea there was yet a third invisible number being used to calculate the scores. Your explanation certainly makes sense, although I don't see why GMAC wouldn't include this information in their explanation of the scoring. I can't see how it would betray any vital secrets to share this with test takers.

Anyway, thanks again.

Is anyone else familiar with what Ian described?

Confused,

I, too, am confused as I have just encountered a similar situation. Were you ever able to receive a sufficient answer?

Ian's response does make sense, but that doesn't make it right in the eyes of the person who got the shorter end of the stick. Ultimately you're right--10 points isn't going to help or hurt you much but in my case, I would love to hit the 700 as opposed to the 690.

Thanks for starting this post!