Analysis of my ESR (Enhanced Score Report)
In this post, I’ll share some observations from my ESR that I found interesting. The order of the observations below is not according to their importance but according to the order things are presented in the ESR.
1. In the IR section, there are 9 scored and 3 experimental questions.
How can I say so? Because the percentage of correct answers is always a multiple of 11% (I have checked this with ESRs of my students too). This is possible only if the total number of scored questions is a multiple of 9.
2. In Verbal, I got around 30 scored questions and 11 experimental questions.
I got two questions wrong in Verbal, one in the first and one in the second of the four sets.
How can I say so? Because neither 12% nor 14% can stand for 2 questions. Since if 12% stands for 2 questions, 88% will stand for around 15 questions, meaning a total of 17 questions in the first of the four sets. Since there are only 41 questions in all, including experimental questions, it doesn’t make sense to have 17 scored questions in one of the four sets.
Similarly, we can rule out that 14% stands for 2 questions.
So, both 12% and 14% stand for one question each. It means that the first set has 8 questions and that the second set has 7 questions. (Well, someone may wonder “why would GMAC create sets with different numbers of questions?”. I think for the simple reason that they don’t want a guy like me to figure out the exact number of experimental questions in the test!)
It means that there were 15 scored questions for me in the first two tests. If I extrapolate that to the next two sets, it would mean that I had 30 scored and the remaining 11 experimental questions. That’s a large number of experimental questions, I believe! Now, of course, I have extrapolated the data of the first two sets to the last two sets. There is indeed a possibility that the last two sets may not have the same mix of scored and experimental questions as the first two sets. In such a case, the total number of scored questions in my test might be higher than 30.
However, it could be lower than 30.
And I believe it makes sense for it to be lower than 30. Why? Because I was consistently getting the questions right. Rather, I didn’t get any question wrong in the last two sets. So, the system didn’t need to doubt my capability, and thus, could throw even more experimental questions at me to figure out the quality of the questions.
It also means that you may not face the same number of scored or experimental questions as I did since you might perform differently than me and the system may need more or fewer questions to figure out your exact capability level and score.
3. In the first half of the test, the system had decided that I didn’t deserve V51.
As can be seen from the graphic in the point 2 above, I didn’t get any question wrong in the last two sets. Thus, it means that I didn’t even have a chance to score V51 after I got a couple of questions wrong.
As can be seen from the above graphic, the average difficulty level actually dropped in the last set of questions. Probably, the system had made its mind on my score and didn’t want to waste more high difficulty level questions on me. Probably, they wanted to save them to be used in my next attempt! Also, since the system had already made its mind, it seems likely that the system might have thrown more experimental questions at me in the last set. They were using me to assess the quality of their questions!
4. No matter what you do, you cannot get more than 51 on Quant.
I think a lot of people already know this, but my ESR confirms this. I didn’t get a question wrong and got Q51.
5. The system behaved quite differently in quant than in verbal, in terms of the difficulty level of the questions.
While in the verbal section, the system seems to have had made up its mind by the third set, in the quant section, the questions became more and more difficult even till the last set. In the last set, the difficulty level seems to be almost breaking through the roof!
6. The strategy that you should be spending more time per question in the first ten questions doesn’t work all the time.
As can be seen above, I completed the first 10 questions in about 13 minutes i.e. 1.3 minutes per question, way below the required average of 2 minutes per question. However, as can be seen from the graphic in the point above, the average difficulty level of the questions in the first set was way lower than that of the questions in the remaining sets. If I had unnecessarily spent more time on these questions, I would not have had the luxury to spend an average of two and a half minutes per questions in the last two sets.
So, I’d suggest that you attempt the first 10 questions with care but not with over-caution. Otherwise, you may end up unnecessarily wasting your time that you could use up in the latter harder questions.
This is all I had!
All the best!