Giorgio wrote:Ok i get it now ...
What about GMAT hidden test questions? I've read that 9 of quant and 11 of verbal are not counted in the score ....which means , let's assume you can answer 28 out of 37 correctly and still get 51 in Q?
These are normally called 'diagnostic' or 'experimental' questions, and yes, they do not count at all towards your score. Before you answer a single question on your test, the computer decides which experimental questions you will see; the questions are organized into groups, and the computer will choose one group to insert into your test. I'm sure the questions are grouped so that they will span all difficulty levels; each test taker should have the same approximate range of difficulty on their experimental questions, the majority likely around the average level, with one or two very easy questions and one or two very hard questions. So your performance in no way influences which experimental questions you will need to answer.
The computer also randomly determines where in your test you will see the experimental questions - so it might decide, when you sit down for your test, that questions 2, 5, 11, 13, etc will be experimental on your Quant section. These selections are different for everyone, and because the question numbers are randomly selected, there's no way to predict where in the test you will see your experimental questions.
So you could, in theory, answer all of your experimental questions incorrectly, and all of your 'real' questions correctly, and get a perfect score with about 10 mistakes on the Quant section. In reality, there is no way that will ever happen. If you answer the first few 'real' questions correctly, every real question you see after that will be near the 800-level. If you're capable of answering *every* 800-level question correctly, you won't have any difficulty at all answering the experimental questions correctly, because pretty much all of your experimental questions will be well below the 800-level.
I'd add that, on test day, it is not a good idea to try to determine which questions count, and which do not. It's impossible to judge even when you are familiar with the test, particularly since there are reasons why you can see questions that actually do count, but which are not at your ability level. It's best to treat every question as though it's important, and not even think about whether a question is experimental or not - you have more important things to worry about during the test!
