This works in theory, but posters on the scoretop.com website claim to have seen anywhere from a few (1 or 2) to as many as a dozen or so. If the claims on the scoretop website are true, then that's not merely a small advantage. I actually believe it to be nothing more than false reports. They might have seen 1 or 2, but I don't believe for a second any of them saw 12 questions.
On the other hand, if GMAC has 5000 questions, and each question costs GMAC $2,400 (as reported in the news articles covering the ScoreTop.com fiasco), and if they refresh their question bank every 30 days, as has been reported is true Even if GMAC refreshes 80% of their question bank, at $2,400 a question, this is ~$1,000,000 per month in overhead just to create new questions, and 20% of the questions were used last month and could be seen by test takers that know of the question (although this "old" bank is still 1,000 questions and the test taker sees only 37 of these questions - also see debate on size of question bank for 700+ test takers). This would be 4,000 people taking the GMAT each month just to pay for the new questions (which is not unlikely). Furthermore, how are the new quesitons tested? Is each question put on someone's real exam as the test question(s)? How would they go about testing 4,000 question per month, 48,000 questions per year.
Is it just me, or do these figures seem rather high? $1,000,000 a month in just test question overhead seems rather high to me. Thoughts on this anyone?
StartupAddict wrote:
The impact is very minimal. GMATPrep has about 5000 questions or more, and I imagine the 'live' GMAT has more. If you see maybe 50 questions, every 3rd time you rewrite the GMAT you might see 1 repeat. And that wouldn't influence your score that much.