One of my friends scored a 40 Verbal and 50 Math and received a 730 overall. Two months later my other friend got the same raw score yet received a 750. How is that possible?

One of my friend scored 40V & 50 Q and got 730. Two months later my another friend scored same split: 40V & 50 Q and got 750. How is it possible?

This is certainly not one of the FAQs. Some questions, however can be answered, are beyond the scope of this FAQ!

I heard that percentiles change over the time. How are GMAT percentiles calculated?

I heard percentile rank changes over the time. How percentile rank is calculated?

Your percentile rank may change till you get the official GMAT score report with AWA. The percentile rank is calculated by comparing your score with all GMAT scores reported in last three years. The percentile rank indicates how you stack up among all recent GMAT takers.

What does a GMAT score look like?

How GMAT score look like?

Please explain its terminology.
Your GMAT score consists of four marks.
1. Your total GMAT score: between 200 & 800. (& percentile rank in % figure)
2. quant & verbal score: between 0 & 60. ( & their percentile rank)
3. AWA score: between 0 & 6.0
So, next time you see someone’s GMAT score as: 770(Q51/V42) 5.5 you know the splits!

My raw score in the verbal section was lower than my raw score in the math section. Yet, my verbal percentile is higher than my math percentile. Why did this happen?

I have obtained less mark in verbal section still its percentile is higher than quantitative percentile; why is it so?

It is observed that GMAT students score higher in quant than verbal. Thus 40 marks in verbal is something “extra ordinary” where as in quant 40 marks is just an average.

Explain to me what a Computer-Adaptive Test (CAT) is. Why do CAT tests have complicated scoring?

Could you please explain me what’s this CAT all about & why it’s so difficult to estimate the score?

The GMAT (CAT) adapts itself according to your performance to the test. Thus there’s no standard measure for GMAT as being difficult or easy. For some students it may throw easy questions & for some it may throw very difficult questions. When you answer a question correctly then next question will be harder. If you answer the question incorrectly the next question will be easier. Thus the exam tries to gauge your ability level by checking your performance (ie how well you perform at various levels of difficulty). Thus to conclude, it is observed that your score shoots higher if you answer more ‘harder questions’. (i.e. the more you roam at the high level during GMAT CAT higher score you’ll get ;) ) Some people argue that to gauge the test taker’s ability level, the initial questions are sensitive and most critical; they have valid points, however these experimental questions (un-scored items placed randomly throughout the test) makes GMAT scoring algorithm so complex and difficult to comprehend that it becomes almost impossible to estimate the real GMAT score.
One more additional point in CAT, (from the official source) is last questions plays very significant role in GMAT score & if left blank due to lack of time hampers the score severely.

Tell me more about experimental questions.

This is scary! You guys have kept me in dark! Tell me more about experimental questions.
There are around one quarter experimental questions in real GMAT. That means out 37 questions in verbal section 8-9 questions & in quant section out of 41, 9-11 questions are experimental. These questions are randomly placed. Very expert GMAT gurus can guess whether a question was experimental or not, as these questions tend to be somewhat different in nature as well as in difficulty level. They add disturbance to the adaptive-nature of the test. The experimental questions (also known as un-scored items), are not adaptive and the response to these questions are not counted in adaptive test. Thus the question coming next to un-scored item doesn’t depend upon the answer of un-scored item. All this makes GMAT CAT (computer adaptive test) difficult to predict.

What are experimental questions? Why are they in GMAT?

So, what is the significance of these “experimental questions”? Why they are in GMAT in first place?

Each real GMAT question is tested out for some period as an experimental (un-scored) question in actual GMAT. After enough testing these questions are classified into difficulty levels based on the responses it received in the GMAT.