I know Manhattan GMAT preaches that the first 10 questions of the exam does not carry any more weight than those of the remaining portion of the exam; however, a number of people, who I have spoken with, used the approach highlighted. Is it possible that there are more experimental questions in the middle portion of the exam than there are in the beginning and ending of the exam? If GMAC does so, the more questions one gets correct in the first half of the exam, the better you will score because those are the questions that get counted toward your final grade. In essence, how does GMAT distribute its experimental questions? Are the questions more in the middle rather than at the beginning or end of the exam?

That is a much-publicized strategy out there, but our position is based on careful research. If by "focusing" on the first 10 questions you mean spending extra time on them, then this will very likely lead to being rushed on the later questions in each section. Being forced to guess on later questions, particularly ones that you could actually solve if you had 2 minutes, is not a good test-taking strategy. Whatever gains you theoretically reap on the first questions would be eroded by a string of wrong answers at the end.

For most people spending extra time on problems does not improve accuracy on the average. The best strategy is the one that we recommend: be disciplined with your time management throughout the exam, staying on target with the timing guidelines and employing educated guessing when you need to.

Lastly, it's just not a good idea to try to determine which problems are experimental. One or more of those first 10 questions could very well be experimental questions -- yet another reason to maintain good time management throughout.

Rey

Straight from the mouth of the guy in charge of the algorithm (I talked to him myself):

1) all questions are worth the same amount
2) the early questions are NOT worth more than the rest of the questions (he emphasized this point)
3) the experimental questions are randomly distributed

The only partial piece of info I have about placement of experimental questions (other than the random distribution) is that the last 5 questions in a row will not all be experimental questions. He wasn't even supposed to disclose that much info - it kind of slipped out - but there is some kind of restriction at the END of the test on the number of experimental questions you can get. That restriction might be everywhere (eg, maybe you can't get 5 experimentals in a row anywhere) but all I know for sure is that the statement is valid for the last 5 questions in the section.

In other words, the only place we know for sure that there is some limit on the number of experimentals is at the end, not the beginning. :)

"The only partial piece of info I have about placement of experimental questions (other than the random distribution) is that the last 5 questions in a row will not all be experimental questions"

Stacey, Does it mean there will not be *any* experimental questions in the last 5 questions or ALL of the last 5 questions can not be experimental?

Any idea how many questions are experimental in a test? I hope number of experimental questions will be same for all the candidate.

Thanks
Path

There may be SOME but the last 5 will not ALL be experimental. And they don't officially tell us how many, but we know it's about 10 questions per section.