How to Analyze a Practice Problem

by on October 9th, 2009

We spent the last two weeks discussing how to analyze practice tests. Let’s dive a bit deeper into how to analyze each practice problem. (You can find the two-part article on analyzing practice tests here: Part 1 and Part 2.)

When we study practice problems, our overall goal is to master the problem we’re working on right now. What does mastery mean? It means that, when we see a future different problem that tests the same thing as this current problem, we will recognize that the future problem has certain things in common with this current problem, and we will know what steps to take as a result — we will, literally, recognize what to do on the future different problem, a problem we’ve never actually seen before.

It’s necessary to get to this level of mastery because the problems we study will never be the actual problems we’re expected to do on the test. But we will see similar problems — problems that have something in common with problems that we’ve already studied. If we can recognize what to do, then we will be faster (which is always important on this test), and we will be more effective — we’ll be more likely to get it right because we’ll know that the method we’re using actually worked the last time we saw a similar problem.

This mastery we’re talking about — the ability to recognize what to do on a new, different-but-similar problem — comes from the analysis we do after we’ve already finished trying a new problem for the first time. And that’s the focus of today’s article: how do we analyze the practice problems after we’ve tried them?

First, of course, you have to try the problem itself, and you should generally hold yourself to the time constraints given by the test. Pick an answer to that problem within the expected time frame, even if you have no idea what the answer is — every aspect of this test needs to be practiced, including how to guess when you don’t know what to do!

Okay, so you’re done with the problem. Now what? Well, the first thing everybody does is check the answer, right? Interestingly, the analysis doesn’t depend much on whether you got it right or wrong! But we all want to know, so go ahead and check the answer. Just be aware that this doesn’t change your review process much.

Next, answer a series of questions.

1. Did I know WHAT they were trying to test?

  • Was I able to CATEGORIZE this question by topic and subtopic? By process / technique? If I had to look something up in my books, would I know exactly where to go? (If I couldn’t do this, I need to take the time to categorize it now, while I’m analyzing the problem.)
  • Did I COMPREHEND the symbols, text, questions, statements, and answer choices? Can I comprehend it all now, when I have lots of time to think about it? What do I need to do to make sure that I do comprehend everything here? How am I going to remember whatever I’ve just learned for future?
  • Did I understand the actual CONTENT (facts, knowledge) being tested? (Go back to your books and teacher to learn / understand anything that gave you trouble.)

2. How well did I HANDLE what they were trying to test?

  • Did I choose the best APPROACH? Or is there a better way to do the problem? (There’s almost always a better way!) What is that better way? How am I going to remember this better approach the next time I see a similar problem?
  • Did I have the SKILLS to follow through? Or did I fall short on anything? (Again, go back to your books and teacher to learn / understand anything that gave you trouble.)
  • Did I make any careless mistakes? If so, WHY did I make each mistake? What habits could I make or break to minimize the chances of repeating that careless mistake in future?
  • Am I comfortable with OTHER STRATEGIES that would have worked, at least partially? How should I have made an educated guess? (Think about this even for questions you get right — it’s often easier to develop alternative strategies and learn how to make an educated guess on problems that you got right. Then, you use the lessons learned when you see a harder problem of the same type / category.)
  • Do I understand every TRAP & TRICK that the writer built into the question, including wrong answers? (Again, it’s often easier to understand and learn to avoid traps on problems that you got right! Learn how to spot them consciously so that you can still avoid them even on harder problems.)

3. How well did I or could I RECOGNIZE what was going on?

  • Did I make a CONNECTION to previous experience? If so, what problem(s) did this remind me of and what, precisely, was similar? Or did I have to do it all from scratch? If so, see the next bullet.
  • Can I make any CONNECTIONS now, while I’m analyzing the problem? What have I done in the past that is similar to this one? How are they similar? How could that recognition have helped me to do this problem more efficiently or effectively? (This may involve looking up some past problem and making comparisons between the two!)
  • HOW will I recognize similar problems in the future? What can I do now to maximize the chances that I will remember and be able to use lessons learned from this problem the next time I see a new problem that tests something similar?

If you are not doing the above analysis on every GMAT problem you study, then you are not getting the most out of your study. If you have done tons of OG problems but haven’t done the above analysis, then you aren’t done with those questions. Go back and start doing this analysis. It doesn’t matter if you do fewer problems — obviously, this kind of analysis is going to take lots of time! But this kind of in-depth, high-quality review is exactly how you master this test.

I would far rather see my students do 20 problems with this kind of analysis than 50 problems without this kind of analysis. Generally, we should be spending about two to five times as long on analysis as we spent doing the problem in the first place. On a 2-minute problem, that means spending 4 to 10 minutes analyzing!

Okay, so what are you waiting for? Start studying! And good luck! :)

40 comments

  • Really helpful stuff.
    Hope to learn better by applying
    this technique of analysis.
    Thanks!!

  • Hello Stacey ,

    I have read all the 3 articles and the only word I can use for them is EXCELLENT .

    Your articles have completely look and analyze problems and this is really working for me .

    Thanks for the excellent work keep posting .

  • Good one!!

  • Hi Stacey,

    Thanks for the above article! Its a great eye opener on what should be the way to ID the problems and to keep a track of where you are going wrong and how to rectify that. I will be employing this from today onwards!

    Thanks,
    Anirudh

  • Hi..

    Stacey you are GOD..... this article is really an eyeopener for me. Infact
    I have been studying for last 3-4 months & planning to take final test in
    coming mid of Jun.

    I found the only reason that I am continuously getting score of around 580, is not analysing my practice work. But now, I have changed my
    strategy & decided to spend same time for analysing also.

    Thanxs a lot .... U r GR8

    AIMKP
    India

  • That 's good article .. realyy helpful to solve problems..

    Thanks Stacey..

    Anju

  • Hi Stacey,

    I've found your articles to be extremely helpful - thank you so much. I have a quick question for you (sorry if this is not the right place to post this) - I am about two weeks away from taking my GMAT. Pretty happy with my practice test results thus far, but I do want to polish up my CR a bit.

    I have gone through the MGMAT course and the CR strategy guide. I've heard that PowerScore CR Bible is a good resource, but with just two weeks to go I'm not sure if this book is worth the time and investment.

    What is your personal opinion of this book? Will it teach me anything I don't already know from MGMAT guide? I know it breaks down different types of questions in further detail, but I'm not sure how useful that really is.

    Thank you!

    James

    • I have never actually looked through Powerscore's bible. (On purpose - because I do research and write curriculum for MGMAT, I have to be careful not to be too influenced by what other companies are doing.)

      I have heard very good things about it from students on the forums, though. My only concern is that you only have 2 weeks to go. Typically, you should stop trying to learn major new things with about 10 days to go - from that point on, it should really be about reviewing everything that you already know (with perhaps some minor tweaks - but not learning totally new strategies for an entire question type).

      In general, though, for others reading this, if a certain approach isn't working for you, definitely try other approaches. Just be aware that it is typically problematic to try to make major changes within a week or two of the official test. That's typically not enough time to learn an entirely new set of strategies.

    • I appreciate this Stacey - thank you. I'm doing pretty well so far on CR - I guess the potential extra "umph" is not a good return when compared to the risk of having too many strategies to choose from and thereby jeopardizing my score. If I'm already doing well with my CR, there really is no reason to switch it up right now.

      Thanks!

  • Hi Stacey,

    I hope to get your thoughts on my plan of attack for the next 12 days before GMAT. My apologies if this isn't the right place to post such an inquiry - more than happy to move this to the forum if needed be.

    A quick summary of my progress thus far:

    1.) Took MGMAT 9-week Classroom course last summer
    2.) Resumed studying this January, on and off, and finished the core curriculum in June (minus the Green and Purple Supplementary questions, which I'm saving for this last push)
    3.) Thus far I have taken four MGMAT practice tests:

    a.) Overall Score: 720-770
    b.) Quant Score: 48-50
    c.) Verbal Score: 40-45

    4.) Recently gone through all of the OG questions I've done (Math and SC), re-did questions I either gone wrong or unsure of and noted improvement areas. I also redid questions I got wrong on the practice tests.

    If you're thinking "wow what a model student", I am about to break your heart. I have real troubles with timing (even though I do all practice questions timed). During the practice tests above (especially the 770 one), I had to pause multiple times during the test due to anxiety, and I didn't do the essay questions either. Therefore, the scores are absolutely inflated, and I feel awful every time I read a post by you that warns people not to do this.

    Now, I do think I have a shot at my target score (720+) - I usually find that questions I pause on I end up getting wrong anyway. The real benefit of pausing is to calm my nerves so I can keep going with the test.

    I am going to be on vacation until the 14th, and I really hope you can help me put together a "last push" plan to get my target score (720+). I am going to get about 10 hours each day to study from now till the 14th. Despite my less-than-ideal practice test setup, I do know all of the 700+ materials well, and I just need to execute it right.

    Here is a high-level overview of what I'm looking to do from now till 8/14. Can you provide some feedback?

    Week 1 - 8/1 - 8/6

    1.) Analyze all 4 practice tests to understand strength and weakness
    2.) Use Green and Purple Supplementary book to target weaknesses
    3.) Take remaining MGMAT Test #5 and #6, timed, with essays, no excuses

    Week 2 - 8/7 -8/13

    1.) Continue Green and Purple Supplementary books for target treatment
    2.) Take GMATPrep Test #1 and #2
    3.) Finish last batch of target treatments
    4.) Chill on 8/13 (I know this is important)

    A note about the practice tests - I know you're not supposed to be taking practice tests in such frequency, but I find my biggest problem is stamina, timing, and anxiety management. I really need to get used to taking 3.5 hour tests and doing more tests is the only way that I can improve on this area.

    Optional Activities - Do you think any of the following should be bumped into my 2 week plan?
    1.) Weekly Challenge Archive
    2.) MGMAT Question Banks
    3.) GMATPrep Flashcards
    4.) Already ruled out - PowerScore CR Bible
    5.) Other activities that I am leaving out?

    I apologize for the long post and I really think I have a good shot at getting a score if I spend my time wisely, be disciplined in my practice tests, and manage my anxiety appropriately.

    Thank you Stacey for your help. GMAT is one of the most stressful ordeal I've ever had to go through, and you and the MGMAT Staff have gotten me a long way. Please just help me push through this last two weeks! (and please don't make me feel worse about my practice tests than I already am ;) )

    Regards,

    James

    • We do need to have these kinds of conversations on the forums, yes. Copy your post over there, then send a link to the post to me via PM (and also to any other instructors whose input you'd like to have).

      Couple of things: don't try to do too much in 12 days and burn yourself out. Figure out what your strengths and weaknesses are from your recent tests, yes, but then prioritize. Which areas are your biggest weaknesses AND are things that show up frequently on the test? (If you're not show what shows up more or less frequently, ask the experts!)

      No to the challenge archive. Most of those are too hard and require solution methods that are more cumbersome than what you'd be expected to do on the real test. Not worth your limited time at this point.

      Re: stamina, you can also set up your study sessions to help you practice stamina. Figure out what you're going to do for an entire 2-hour period. Then sit down and do it, taking only one 8-min break in the middle. Don't answer your phone. Don't eat or drink anything, except on the break, etc.

  • Hey Stacey,

    As requested I sent two PMs to your user name with a couple follow-ups. However, the message is not shown in my Sent Box so I'm not sure if it actually went through. Can you let me know when you have received them :) If not, I'll resend again.

    Thank you so much!

    James

  • FYI - I think PMs stay in your outbox until the person you sent it to actually opens the message up!

  • Oh okay - I see them in my Outbox :) I assume you've gotten them? Let me know if I need to resend (sorry first time using the forums not sure if I'm doing everything correctly)

    Good news... 770 on my last practice test, but inflated again due to a few pausing during math. Verbal was smooth sailing, however.

    Can definitely use your advice! Looking forward to your PM responses.

    James

  • Hi Stacey,

    Thanks for the great article. Question for you. When making careless mistakes, you pose the question; "What habits could I make or break to minimize the chances of repeating that careless mistake in future?"

    Could you recommend some habits/techniques to overcome this weakness?

    Thanks,
    James O

    • I can make recommendations for specific weaknesses, yes... you just have to tell me what those weaknesses are. What you do to remedy a weakness is specific to that actual weakness. :)

      For instance, if you notice that you make more mistakes when you do things in your head (which is true for all of us!), the remedy is to write ALL of your work down. If you notice that you frequently mis-translate equations when translating from English into math, then one remedy is to check your translations with real numbers. (You don't check your work on every single thing all the time, but you do get into the habit of checking your work every time on things for which you have a history of making errors.) And so on.

  • Hi Stacey,

    I took 5 Manhattan Gmat tests and got an average of 710.I always got 16-17 incorrect in the verbal section. The same 16-17 incorrect got me a score of 26-27 in verbal.I know that the order of getting the questions wrong does matter.
    Moreover, I took gmat a few days back and got 29 in verbal.I know that I am lacking in the verbal section as I need to analyse each and every mistake from my tests as well as the OG.

    I am starting my preparation again, but I need some information from you.I went through 'http://www.manhattangmat.com/forums/post48911.html#p48911' post and got the question wrong.I thought of 'A' as the answer.Now, can I find these kind of concepts in Manhattan's SC guide, 4rth edition?

    Regards,
    Manas

    • The question that you cited deals primarily with idioms, and yes, this is discussed chapter 9 of the 4th edition SC strategy guide.

      Note: there are thousands and thousands of idioms in any language, so they cannot all be listed in one concise book. We list ONLY the ones that we actually see on official questions. So, everything that is on the list was actually tested at least once on an official question.

      That does not mean, however, that everything that has ever been tested is on that list - there could be other things that we just haven't seen - when they release new questions, etc.

      As it turns out, "intent" is listed in the idioms list in the 9th chapter. So go take a look there. :)

  • Thanx for the reply Stacey! I was wondering whether we have to rote the idioms whenever we see them or do we have to learn them by some rules.As stated in that post 'the NOUN "intention" must be followed by "of VERBing". that's just the way it goes', do we have to learn rules regarding idioms?

    Or when we start going through the 9th chapter , we will keep applying the rules consciously?What is the trick?

    • Idioms are memorization issues. There aren't overall rules - it's just "this word goes with this preposition" or "this word takes an infinitive" or whatever. And you have to memorize what each one uses -there isn't a pattern.

  • Hmmm!! I will keep that in mind. Stacey, suppose I get a question in which I am confused between two options.

    One consists of 'being'.
    Second consists of 'ability of'

    Rest of the structure seems fine in both.What should I choose?

    Sorry for creating a hypothetical situation, but I want to know what to choose between an incorrect idiom and an awkward structure.

    • I'd have to see the sentence. "ability of" by itself is not always incorrect. "ability of " is always incorrect.

      The ability of the boy to run for 3 hours straight astounded his parents.

      That's awkward, but you can say it. As I said before, you can't say "ability of " - if you want to use a verb, you have to use "ability to." But I don't have a verb following "ability of" in my above sentence.

      If you do have a 100% always incorrect idiom in one and "being" (or any other awkward structure) in another, incorrectness always trumps awkwardness. But it does actually have to be something that is definitely incorrect. :)

    • Although I should point out that we do still have "to run" after ability - but we have "of the boy" in between. :)

  • Hi Stacey,

    I had two doubts in your explanation above.
    Firstly,What i understood from the above discussion of "Ability of" is that:

    "Ability of" + anything but a verb would be Correct
    "Ability of" + verb would be always incorrect

    Is my understanding correct?

    Also consider the below example:

    His ability of running

    Is this correct usage of "Ability of"??

    Secondly,if we are stuck between an incorrect idiom and
    an awakward sentence do we choose the incorrect idiom??

    Thanks,
    AS

    • If you have something that is actually incorrect, then do NOT choose that over something that is merely awkward. Incorrect is worse than awkward.

      Ability of + noun = okay. "of" is a preposition, so it would create a prepositional phrase, for which we need a noun

      A verb requires "ability to"

      "Ability of running" technically is using a noun, since an -ing word can be a noun, but this is actually the noun form of a verb and that is considered awkward in this idiom. It's cleaner to use the word in verb form, so I would expect the correct answer to use the verb form.

  • Brilliant! I feel armed with new knowledge. I can taste a higher score already.

  • Thanks Stacey! Great Article! wish could have seen this earlier..now i know why i scored 620 even after completing OG :-)

  • Thankyou,This is really useful:)

  • Hi Stacey,
    I have followed all of your articles in this series and they have been very helpful :) However, i have one doubt. You mentioned that test takers should be able to relate problem faced on test to some problem that they did in the past. Does that mean we should learn / remember the problems we do and if that is the case i would like to know how to achieve that. At present i can only deduce what type of ques (say assumption on CR) but i am not able to relate to some previous ques. and it might be a reason that my score is stuck at 680-690. 

    • You don't necessarily need to remember the exact problem (and you definitely shouldn't try to memorize a bunch of problems), but you should be able to say, sometimes, "Oh, I remember seeing something like this before. Last time I saw something like this, XYZ was the best solution method so I'm going to use that again, or there was a big trap with ABC, so I have to be careful about that."

      You won't do this on all problems, just some - but the more for which you can do this, generally speaking, the better you're likely to do.

      Also, this is easier to do on quant than on verbal. On verbal, you'll see that you have, say, a Find the Assumption question, and you might realize that this new question using the same kind of assumption you've seen on other questions in the past - e.g., it assumes that one thing causes another thing even though it doesn't present evidence for that causation.

      Or you might see a particular trap in an SC question and realize, oh, I saw them use this same sentence structure (though not the same words, obviously) on another question, and it was a trap because it sounds fine but actually this part is ambiguous (or whatever the problem is).

      It's possible that this is one reason your score is stuck. Timing problems often also cause our scores to get "stuck" at a certain level, as well as holes in our foundational knowledge of the material being tested.

    • Thanks Stacey, i have started building more PROCESS oriented approach so as to remove any holes in foundation. For e.g. In strengthen CR ques, now i have reference points on what can be possible answers. Now i know that correct answer to strengthen will EITHER strengthen the assumption OR provide some new evidence. Even earlier i was able to solve strengthen ques. but i guess over time this process will give me more confidence.

      1) Do you think my process oriented approach for removing foundation holes is correct 
      2) What should i do to recall the structure of earlier problems because as you mentioned this can also be one of the reasons for my stagnant score

    • Yes, a process-oriented approach is good for CR - you should know what you're supposed to be doing for each Q type, what the likely traps are (which will help you spot wrong answers), what kind of reasoning is necessary to get to the right answer (based on this question type), etc.

      On Strengthen, for example, the correct answer will introduce some new information. That information does not HAVE to be true, but if it is true, it will make the conclusion at least a little more likely to be valid or true.

      When studying, try to articulate why the argument works the way it does and why the correct answer strengthens / weakens / whatever. That helps you to notice the important parts of the structure from a logical perspective (I'm talking formal logic here).

      eg: (I just saw a flawed newspaper article on this yesterday!) The author concludes that feelings of loneliness lead to physical isolation from others and health problems. But the evidence only says that the study found that people who reported that they felt lonely were also more likely to be more physically isolated and to suffer from more health problems. That's correlation, not causation - we don't know which causes which.

      So now if I see conclusion language that talks about causation of some sort (one thing causes another), then I'm on the look-out for the gap between the evidence (which shows only correlation) and the conclusion (which concludes causation). If they ask me to strengthen, I need to find something that helps to show causation. If they ask me to weaken, I need to find something that makes that causation less likely to be true. If they ask me to find an assumption, I need to find something that shows that they're assuming this causation to be true (and the correct answer is often an "inversion" - they're assuming some other pattern of causation is NOT true if they're also assuming that this pattern of causation is the correct one). And so on.

  • Dear Stacey,
    Just wanted to say thanks. I re-did OG 12 with "process" oriented approach that we discussed few days back and i was more confident in each one of the question. I could easily see that basic structure and pierce through the extraneous info. I was like "oh man this causation question, these must be the correct and wrong answers" then i looked at options and found the same. It is a great feeling...

    Now i am onto SC ( i have SC 4th edition), can you please tell how should i proceed so that i nail SC with same confidence ( i've read http://www.beatthegmat.com/mba/2010/06/03/how-to-read-a-sentence-correction-problem)

  • one reason for me asking above point is that CR has fixed pattern and once you realize them, you are almost done. But does such patterns exist on SC and how could i work upon them to make sure that i can recognize them in at least 80% of ques. ( as mentioned in your article)

  • Hi Stacey, 
    ii would like to thank you again for the invaluable discussion that we had sometime back. since then i have followed your advice - recognized patterns on CR for about a week and then did mgmat sc 4th ed. thoroughly. When i was preparing those sections individually, i got almost every question correct on each section with proper reasoning ,and i prepared notes on for each section. to focus completely on the section i was doing i did not do question from other section that i already completed. i finished SC today only and started revising notes on CR, RC & SC. my new problem is that i don't have same amount of confidence and i am feeling that i might not able to achieve same hit rate on any of the section. 

    please advise me how to cope with this situation... for last 5 hours i am feeling dejected as if all my hardwork went in pain.

    • *vain (not pain)

    • That's okay, this is how it works. Learn it really well first, then start drilling it in more realistic circumstances.

      Make some short - very short - notes for yourself on each question type / pattern. Short enough to put each on one flash card. Then start drilling the patterns - but not of one Q type alone. Mix them up.

      In particular, make sure you are drilling "When I see ________" and "I will think / do _______." In other words, what are the specific clues that should immediately make you think "oh, this is an X question" or "oh, this is a trap" or whatever.

      Also make sure that you can tell me, for all CR and RC Q types, "this is what you do for an X question, this is what kind of reasoning to use for the right answer, these are the kinds of traps they like to use." Weaken? Inference? Evaluate? For SC, make sure you can tell me what reasoning is useful for different kinds of errors / grammar areas / meaning issues and again what kinds of traps they like to use.

      Also, remember that you're not going to get everything right on the real test either. It's a 3.5 hour long test - they expect you to make mistakes. You just want to get enough right. :)

  • One of the best articles on GMAT I have ever read!

  • Great article

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.