How important are the first 10 questions?
Experiment: We will try to disprove the new myth about the Official Guide and GMAC positions. The first 10 questions are not important.
Method: We will attempt to incorrectly answer the first 10 questions. Not guess them, but just answer incorrectly. We will then attempt to answer the remaining questions correctly.
Results: Q38. 48th percentile with just 10 wrong answers.
Analysis: The myth about missing the first 10 questions is busted. They appear to be pretty important on the Quant. If you miss the first 10, you literally have no way to fix it. Most of the questions were easy to medium. On the next myth we will miss the last 10 questions.
Note: When simulated in the prior GMAT Prep version, this experiment placed us in Q41. So there have been changes in the scoring algorithm.
Total score: Each question was answered as “A” and gave us a score of V8, 1st percentile. Total score of 380 in the 10th percentile.
Questions: 17 DS and 20 PS
How important are the last 10 questions?
Experiment: We will analyze the last 10 questions to see if they are just as important as the first 10.
Method: We will attempt to answer the first 27 questions correctly, and will answer the remaining 10 questions incorrectly.
Results: 92nd percentile. Q50. It would be possible to get a Q51, if the 10 questions were answered correctly.
Analysis: The questions tended to be a lot harder with some new types and topics thrown in like coordinate geometry DS. The truth is that the last 10 didn’t count as much as the first 10 did. Take note that I didn’t guess at any of these questions, but just answered them incorrectly. If you tend to guess incorrectly then you may end up with a higher score.
Total Score: Each question was answered “E”, and it gave us a final score of V8, 1st Percentile. Total score was 470, 90 points more than when we missed the first questions.
Questions: 17 DS and 20 PS
How important are the middle 10 questions?
Experiment: We will try to find the importance of the middle 10 questions and see how the results are compared to the other 2 experiments.
Method: We will answer the first 14 questions correctly, answer the middle 10 incorrectly, and the last 13 questions correctly.
Results: 85th percentile. Q49.
Analysis: The questions weren’t easier than the 2nd experiment, even though I had made 10 mistakes. It has better value when it comes to effort, but not by much.
Total Score: Each question was answered “D” and gave us a V6 score, 1st percentile. Total score of 450.
What is the penalty if I miss every 3 questions?
Experiment: This time, we will be testing to see what the score would be if every third question is answered incorrectly, such as 3, 6, 9, etc. This will simulate someone who gets 3 minutes instead of 2 minutes for each question and then guesses on the third one.
Method: Every 3 questions was answered incorrectly. So 24 correct answers and 13 incorrect ones.
Results: Q49. 85th percentile.
Analysis: A bit of a lower score, but much less effort getting there.
Total Score: Each incorrect answer in the verbal section was “C”, and gave us a V8, 1st percentile. Total score of 460, 23rd percentile.
Best Guessing Strategy:
What if a bunch of questions in the middle were guessed?
Experiment: I will attempt to examine guessing strategies to find out where time should be invested.
Method: First 10 questions will be answered, then the next 17 questions will be guessed, and I will answer “C” to them and manage to get 4 out or 17 right, and then the last 10 answered correctly.
Results: 73rd percentile. Q46.
Analysis: This was similar to the above experiment. Missing 3 more resulted in a lower score by 3 points.
Total Score: Each question in the verbal section was answered “B” and it gave us a V6, 0 percentile. Total score 420, 16th percentile.
Questions: 16 DS and 21 PS
Best Guessing Strategy:
Every other question guessed
Experiment: I will try to guess every other question. I answered “C” to all the even numbered questions and managed to get 3 out of 18 correct.
Results: 55th Percentile. Q40. I had 15 incorrect and 22 correct answers.
Analysis: Not really sure what was to be expected only guessing every other question. I managed to get the 55th percentile. I don’t believe this is the best strategy. This score is too low to be considered by anyone.
Total Score: Each question in the verbal section was answered “A”. It gave us V6, 0 Percentile. Total score was 380, 10th percentile.
Best Guessing Strategy:
Variable Guessing
Experiment: The first 10 correctly answered, I will guess the next 9 questions by answering C to them (got 2 out of 9 right), then answer the next 9 correctly, and then guess the remaining 9 questions by answering “C” to them (got 2 out of 9 right). 14 incorrect and 23 correct answers.
Results: 85th percentile. Q49.
Analysis: I was actually amazed to see my score. This is the best guessing strategy so far, considering that I only answered half the questions.
Total score: Every question in verbal was answered “E” and gave us a V9, 2nd percentile. Total score of 470, 26th percentile.
DS -vs- PS: Why are the DS questions important?
Experiment: The first 10 questions were answered correctly. After that I incorrectly answered the DS questions. 12 incorrect answers and 25 correct answers.
Results: 92nd percentile. Q50.
Total score: Each question in the verbal section was answered “E” and it gave us a score of V8, 1st percentile. Total score of 470, 26th percentile.
DS -vs- PS: Why are the PS questions important?
Experiment: The first 10 questions were answered correctly and after that I answered the 12 PS questions incorrectly. 12 incorrect answers and 25 correct answers.
Results: Q49, 85th Percentile.
Total Score: Each question was answered “E” in the verbal section and it gave us a score of V11, 3rd percentile. Total score being 490, 31st percentile.
The importance of answering all the questions in time.
Experiment:
- Answer the first 30 questions correctly and the last 7 questions incorrectly.
Result: 92nd percentile. Q50.
- Answer the first 30 questions correctly and the last 7 questions weren’t answered at all. (out of time)
Result: 71st percentile. Q45.
Analysis: As you can tell, it is very important to finis the test and answer the questions in time.
If you have any ideas or questions that you want to try, let us know.
There will be more experiments in the future, so feel free to throw out more suggestions that would be interesting to test out.
NOTE: Even though the GMATPrep is the closest algorithm to the actual GMAT, it isn’t likely that it would be identical. Some of the experiments could give you a different outcome if you were to attempt it on the real test. So far, there hasn’t been any reports to deny or confirm that these results based on the test day experience.