Evaluating Your Practice Tests, Part 1 of 2

by on September 23rd, 2009

Practice tests are an invaluable component of any test-taker’s study plan, but the most valuable thing is actually not the act of taking the practice test. Just taking a test doesn’t help us to improve all that much. While taking a test, we are concentrating on doing (using everything we’ve learned up to that point); as a result, we’re not really learning much.

The most valuable thing is actually the data that you can extract when you’re done with the test; that’s how you learn to get better and know what to study before you take another practice test. There are two main components to that data:

  1. Statistics and metrics based on timing, difficulty level, percentage correct, question category, and so on
  2. A thorough, analytical review of the specific questions that you saw on the test

This week, I’ll take you through how I review the statistic and metrics from my own students’ practice tests. I’ll base my discussion on the metrics that are given in ManhattanGMAT tests, but you can extrapolate to other tests that give you similar performance data (note: you need per-question timing and difficulty level in addition to percentage correct/incorrect data). Next week, we’ll take a look at how to review specific questions.

First, naturally, I look at the score. I also check whether the student did the essays (if she didn’t, I assume the score is a little inflated); I also ask the student to tell me whether she used the pause button, took extra time, or did anything else that wouldn’t be allowed under official testing guidelines.

Next, I look at the problem lists for the quant and verbal sections; the problem lists show each question, in order as it was given to the student, as well as various data about those questions. First, I scan down the “correct / incorrect” column to see whether the student had any strings of 4 or more answers wrong. If so, I also look at the time spent; perhaps the student was running out of time and had to rush. I also look at the difficulty levels because sometimes I’ll see this: the difficulty level is high for the first problem or two, and the timing is also way too long. On the later questions, the difficulty level is lower, but the timing is also too fast. Essentially, the person had a sense that she spent too long on a couple of hard questions, so she sped up… and then she not only got the hard questions wrong but she also got the easier questions wrong because she was rushing.

I also look at the timing for the last five to ten problems in the section to see whether the person was rushing or had a lot of time left (meaning she rushed earlier in the section).

Next, I count the number of questions that fall into the “way too slow” category. Too much time is: 3+ minutes on quant or CR, and 2+ minutes on SC. RC is a bit trickier, because the timing for the first question includes the time spent to read the passage. If it’s a first question, “too long” is above 5 minutes. If it’s not a first question, “too long” is above 2.5 minutes. If there are more than a few, then the student has a timing problem. My next question: how significant is the problem?

For these “too slow” questions, I count how many there are, how much time was spent total, and how many were correct vs. incorrect. For quant, I also count how many were Problem Solving vs. Data Sufficiency. Finally, I see whether there are any patterns in terms of the content area (for example, perhaps three of the “too long” quant problems were geometry problems or two of the “too long” SC problems were modifier problems).

Next, I count how many “way too fast” incorrect questions there are. “Way too fast” is anything done in less than half the time it was supposed to be done (for example, “way too fast” would be less than 1 minute for a quant question). I do not, however, count incorrect “too fast” questions that are rated 700+ unless that student is scoring 700+; I assume the student (wisely) realized the problem was too hard, made a guess, and moved on. That’s the only good reason to get something wrong in a “too fast” timeframe. Otherwise, these lower-level, too fast, incorrect questions represent missed opportunities – careless mistakes – and they were caused by the “too slow” questions from above.

All of the above is to quantify for the student just how bad the timing problem is. Literally just seeing the data can help students start to get over that mental hurdle (“I can get this right if I just spend some more time!”) and start balancing their time better. And the stats on question type and content area will help the student to be more aware of where she tends to get sucked in.

Next, I run the assessment reports and look at the Assessment Summary. This tells me percentages correct for the five main question types, as well as average timing and difficulty levels. Problem areas are indicated by:

  • Percentages correct below approximately 50%, especially when coupled with lower average difficulty levels (though I’m not worried if I see, say, 48% correct with an average difficulty level of 730 – that’s a good result)
  • Average timing that is 30 seconds (or more) higher or lower than it should be on average
  • A big discrepancy (more than 20-30 seconds) in average time for correct vs. incorrect answers of the same type

I then go through the other reports (showing all of the problems divided into various sub-categories) while categorizing things into five buckets (though I may have to adjust my assessment if there are only one or two questions in a category):

  1. 50% + correct plus timing within the expected timeframe (otherwise known as strengths)
  2. Less than 50% correct plus timing in the expected timeframe (possible weaknesses in content area, methodology, etc, BUT check the difficulty levels; maybe this category just happened to be really hard on this test!)
  3. Less than 50% correct plus timing way too fast (an average more than 30 seconds faster than it should be); are these really weaknesses or was the student just going too fast (and, naturally, making more careless mistakes)? Why was the student going too fast on these?
  4. 50%+ correct plus timing way too slow (an average more than 30 seconds slower than it should be); these are still weaknesses even though the percentage correct is high! Figure out why the timing is higher and how you can do these more efficiently.
  5. Less than 50% correct plus timing way too slow (an average more than 30 seconds slower than it should be); these are the biggest weaknesses, obviously. Get them wrong faster. Seriously – you’re getting them wrong anyway, so start by just taking less time to get them wrong! That will improve your performance on all those ones on which you’re currently rushing and making careless mistakes!

Click here to see the second article in this series, where we’ll talk about how to review the specific questions, topics, and content areas that were tested on the exam.

Read other articles in this series:


  • As always, great post Stacey.


  • Great article Stacy, I'm really looking forward next week topic,
    how to review the specific questions and how to exploit the maximum of them.

  • What an amazing set of articles!
    My name is Oriana Torres, Colombian, I`ve done the GMAT 3 times, 490, 620 and 490. The last 490 was a disaster result of tension and nervoussness, though I still have challenges with accuracy and timing in math ...

    Reflecting, I have decided that my main strategy before I present the test next time is to take as many Practice Tests as possible. I feel "ready" in terms of knowledge and untime rehearsal, I mean, you will never be 100 % ready, but partly the test is about playing with the (solid) basis you have and get the best score out of it.

    I think your approach describing the diverse categories on how to assess the questions answered in practice test is AMAZING, it will help me a lot!!


    • Hi Oriana,

      I am having the same problem like you. Even if I didn't took any ream GMAT exams, I am having stress and tension while taking the mock up tests. Can you tell me how you improved from 490-620. It was a good change, I hope. I will also take the real GMAT exam on July 13, 2010.

      Thank you.

  • Hey Mati,

    I wish I could tell you that when I got the 620 I was very calmed!! I was not - it was not as disastrous as the last time when I falled into the 490, but I still was not a "benchmark" in terms on how to be calmed - imagine, probably I would have got even more than 620!

    I think in general my jump in score was due to a strong preparation,despite the nervousness, my confidence content-wise was very high. And this is what I am trying to gain back. I also got very vert tensioned in the mocks, I see them as the ultimate indicator if all those study times were worth, and in a way they are, but one needs to treat them carefully because no matter how well prepared you are if you can´t manage your attitude and confidence during the test, all the knowledge that surely is in your head is meaningless. That´s why now my stratategy is just sit down and do as many tests as possible under this becomes so normal for me that facing a real test date makes no difference.


    BELIEVE you will do well, and you will!!!

    • Thank you man. To tell you the truth I am not that much afraid of taking the test, because the schools I applied already accepted me and I am waiting for an I-20. But, I need the GMAT for financial aid purpose so that I will save money for other expenses.

      The thing that stress me is, not answering a question or not understanding what the passages says. And when I get angry I loose focus, which is the worst of all. But from now on I will keep checking my self.


  • Great Post Stacy!!

  • Dear Stacy

    I will be appearing for the exam in the last week of october 2011 but i am worried about the score as i am getting a score of 520-540 but i need to score more than 675 please help me that how do i need to improve in SC ,RC and CR.The most problem i face in is SC which let me down and so as the score.please advice on how to improve it as i have already read Manhattan sentce correction twice as i did improved the score but not scoring upto that extent.


    • Those are good questions to ask - but the forums are the place to ask, not here. :)

      Post in the forums and give ALL of the details of your case. In order to advise you, the experts need to know your scoring breakdowns and detailed analysis of your strengths and weaknesses.

      I don't participate on the BTG forums, but I do answer these kinds of questions on the MGMAT forums. If you'd like to hear from me, please post @ MGMAT. (But I also encourage you to post here at BTG - it's good to get advice from multiple sources.)

      If you post on MGMAT, please use the below article (an update of the one you just read) to analyze your most recent MGMAT test and include your analysis in your post (Note: NOT just the raw data, but what you think the data means - that's an important skill to develop).


  • I have booked my GMAT test date on May 9 2012. I have taken the first Manhattan GMAT cat code and the score i got was 400. My target is to score in the 600-700 range. Is it possible because I have just got less than a month's time. I am trying to focus on my fundamentals. I want to be strong in the fundamentals so that I avoid making silly mistakes. Could you please help out how my study plan should be ??? How many more practice tests should i take before the real GMAT !  

    • I don't think I have ever heard of a case of someone going from 400 to 600+ in one month. I won't say it's impossible - you never know! - but most people would need several months at least in order to accomplish that kind of jump. I know that's probably not what you want to hear, but it's better to know so that you can make appropriate plans.

      We don't help people develop study plans here in the article comments, but you can post to ask for help on the forums here (note: I don't participate on the BTG forums) or on the MGMAT forums (I do participate on those forums).

  • Great post.. Thanks a ton.
    Could you please tell me how to generate the assessment report for the manhattan tests?
    Thanks in advance.

Ask a Question or Leave a Reply

The author Stacey Koprince gets email notifications for all questions or replies to this post.

Some HTML allowed. Keep your comments above the belt or risk having them deleted. Signup for a Gravatar to have your pictures show up by your comment.