GMAT Strategy

Anyone who has experience with a disappointing first attempt can relate to this. I scored 95th percentile in verbal... with a dismal 50th percentile in quant. This is the lowest quant score I've ever received! I want to say that I 'froze' or 'choked' but to be perfectly honest, I don't fully understand why it turned out so awful. I'm not sure if I need a different tutor or a psychologist at this point!

My practice test scores have been: 530 (initial GmatPrep 1; no studying)..640/660/680/710/710. All Manhattan Gmat. Quant has always been my weakpoint. I was somewhat hoping that I would "get lucky" on the actual test and be somewhere in the 7s. My initial goal was 750, but I would've been happy with 720 or so.

Following the very useful tips given by an expert in the only other topic I've ever posted, I really tried to improve upon my strategy. I wrote out a "method" for each question type, I went through ALL OF OG, and truly spent a ridiculous amount of time reviewing my errors. I re-did all my tutor lesson plans (his method was by subject i.e., Divisibility, Rates and Work, etc). The problem is... I am amazing in problem sets. I can solve almost any problem I see when I'm going through OG, or going through my tutor lessons.

Bottom line is... no matter how great I feel about the overall content and about specific question types... when it comes to taking the actual test, clearly something is NOT clicking. I should note this has been a very long term and painstaking process (I'm not cramming). I had worked through ALL of the Manhattan strategy guides and focused solely on content for the first 4 months of my efforts. Then I worked with my tutor 3hrs a week for 2 months - this was structured as "by topic". Since then, I've worked through OG, all Manhattan CATs (reviewing errors) and I have probably over 1,000 index cards that I work through daily.

All this... for a 50th percentile? I'm not sure whether I should be embarassed or angry. Neither one will be constructive. I must retake. PLEASE HELP!

Hey Lindsey,

First of all, congratulations on a 670! If there's one thing that communities like BTG do a little irresponsibly, I think, it's putting these undeserved stigmas on scores that are actually really impressive accomplishments. What's that, a little above the 80th percentile? So better than 4/5 of examinees, nearly all of whom have graduated from college and feel qualified to succeed in higher education? I know we all dream of scores that start with 7, but in doing so I hope we don't forget that among the pool of self-selecting GMAT test-takers there's no shame in being more than a standard deviation above average!!

Now, that said, I certainly agree that you can improve and I admire your commitment to doing so. And with this as a starting point (at least an "official test starting point") you're in pretty good shape.

I have a couple thoughts that I think might help given what you wrote:

1) It sounds like most of what you've studied has been at the "content" level. But I recently heard a pretty enlightening quote from one of the higher-ups at GMAC, those who administer the test, that "we're not testing your math skills or your verbal skills; we're testing your reasoning ability and higher-order thinking. If you don't know algebra or geometry well enough you'll get questions wrong, but that's not what the questions are about. We use those topics to set up questions that assess your ability to think logically and critically - that's what we're really testing."

Now, what I'd take from that is that it's not enough to know "all the problems in the OG". The authors of the test know exactly which problems are in the OG, and their mission isn't to reward you for memorizing each individual problem setup and having a specific method for it. Their mission is to see if you can really think; to see if you can see a unique-looking problem and use the skills and frameworks that you've studied to make sense of it.

So I'd challenge you to "think like the testmaker". As you see problems, ask yourself how the GMAT authors could make them harder or just "different". When you're reviewing concepts ask how they might be able to turn them into unique questions. Challenge yourself to see concepts and question types from multiple angles and not just the A-to-Z way that that problem happened to be written.

So, for example:

-At this point you know "number properties": Even/Odd, Positive/Negative, Units Digit. But while a 67th percentile question may ask you to employ Odd/Even number properties, a much-harder question might ask you to use the real guiding principle behind those properties, which is that numbers often behave in patterns. It may not test a property that you could memorize or have seen before, but you can employ the idea that "when in doubt or when dealing with large numbers that I don't want to calculate, try to find patterns". Here are a few examples:

https://www.beatthegmat.com/remainder-q- ... 61475.html
https://www.beatthegmat.com/mba/2010/09/ ... ries-day-5 (referring to the problem/solution on that day...not the "new problem" that day)

Now, I choose both of those examples for a second reason - they're two wholly-different looking problems that both employ the same exact (and kind of unique) number property. And that's what I mean by "ask yourself how they could create a new problem or make this one harder". If you can anticipate ways in which they can twist one problem to make you use the concept in a different way, your knowledge base is much more flexible. So "they asked me for the tens digit so I..." could become "they gave me the tens digit and I...". If you've only learned from A to Z they can trap you by making you go from Z to A. So try to understand concepts both ways, or really from a 360-degree view. My favorite OG question these days is the one that goes:

For positive integers m and n, when m is divided by n the remainder is 14. And the result of m/n = 65.4. What is the value of n?

People abhor this problem. But it's classic GMAT - it calls on you to recognize that you can do not
-evenly-divisible division three ways: (so let's use the example 11/4)

Integer quotient and remainder (2, remainder 3)
Integer quotient plus mixed number (2 and 3/4)
Decimal quotient (2.75)

In each case, the 2 remains the same. So in the given problem, we don't really need to deal with 65. We know that if m/n = 65.4, then if they tell us the remainder is 14 the answer in that "quotient-and-remainder" setup is 65, remainder 14. And how do we relate the remainder to the decimal of .4? It's that middle way to express it - the mixed number. We take the remainder and divide back by n, and that gives us the decimal places. So 14/n (the remainder divided back over n) gives us 0.4: 14/n = 0.4
14 = 0.4n
140 = 4n
35 = n

And what the GMAT learns from testing this problem is whether you can see that relationship between remainder and decimal point. They're making you reverse-engineer the concept of division. If I told you to divide two numbers and I asked for any of those three formats for the solution, you'd do it without thinking. But what makes this a critical-thinking exercise is that they didn't ask that; they asked you to take the results and use them to "create" the process. They asked you to go from Z back to A. And that's more of a conceptual/strategic skill than it is a "content" skill.

I'm probably losing the whole enumerated list thing by ranting on point #1, but the point really is this - think like the testmaker. Don't learn one method to solve one problem. Think about ways they're asking you these questions and how they can tweak them to ask them in different ways. Learn a 360-degree view of concepts and question types, because harder questions have to be written in a way that rewards those who can really think.


2) I promise I'll make this point a quicker read! But the other thing that I took from what you wrote is that you fear the test a little bit. The 1,000 flashcards, the capitalization on "ALL" the OG, the focus on thoroughly covering content...all classic signs of someone who went into the test "hoping to remember" and probably worried about pacing, trying to gauge question difficulty for an idea of 'how I'm doing', etc.

Now, that's pretty normal, but recognize that test stress and fear-of-the-GMAT are typically pretty counterproductive. So I'd recommend:

-Have a "contingency plan" for pacing. Know that if you fall a couple minutes behind your target pace, rushing through multiple questions to get back on track probably means that you'll miss a few questions that you should get right. Know also that you're going to miss questions - everyone does. So you may want to plan in 3-4 "I pass" questions on which you STRATEGICALLY guess quickly to save yourself time and confidence. I think you can manage the test better (but I don't know...was pacing or just that "I have to rush" feeling a problem?) and allow yourself to relax and do the things you do well, well.

-See the GMAT "traps" (negative numbers and nonintegers as those exceptions on DS problems; questions with multiple variables that bait you into picking the value for x when they really want the value for y, etc.) as "weapons". You GET TO use those as ways to beat the test; you don't "have to remember" those or the test beats you. You know the test, so try to see that knowledge as an asset you have in your favor instead of something that creates apprehension because you worry that the test will use it against you.



Anyway, just one man's opinion on how you can improve. I hope that helps...

(whoops...double-post...sorry)

Brian I can't thank you enough for the wonderful post.

The insights are invaluable.

Akhil

GMAT Quant(even Maths in general) is not a memorizing rules by flashcard thing. This approach is fine for CR/SC but not the best for Quant. Most GMAT Quant problems (at higher levels) don't fall neatly in one category. You should try to develop an overall mastery of quantitative aptitude rather than fully relying on piece meal approach (Piecemeal approach is fine only till Q40).

The best way to do so it to solve tough problems untimed. Would suggest spending even 5-10 mins on tough problems, to fully grasp the concepts and to give your maximum effort. 200 tough questions fully understood are better than 2000 easy problems. This would force you to bend your mathematical muscle more than you would normally.

You can get a collection of 200-300 tough problems, try each on your own and ask your tutor to suggest better ways. Even posting on BTG is not a bad idea. There a quite a few quants here

Thank you so much for the replies!

Brian - your input is invaluable! I had tried to do this previously.. but clearly not all that well. I just started a new (demanding) job yesterday. Hence, now my study time must be extremely efficient... as opposed to my previous time intensive method. I actually don't regret focusing so much on content; I have to admit, my base level skills were sorely lacking. However, now it's all about strategy. I hope to be ready within a month or two, max. I need to be over this GMAT process.. life is getting too stressful and I no longer have that precious time to spend. You've already given me so much go to on. I only ask one additional question: how to take that abstract process and make a more direct plan. Even just a first step would help. I commute about 1:45 each way, all through public transit, so I plan to keep this commute time as my study time. Any input about how to implement a more strategic final plan would be so helpful.

Thank you all for such a great sight!

Your verbal is in the 95th percentile. That's good because verbal is the toughest part (so much gray area, having to choose between a right answer and a "more" right answer).

Math is black or white. You're either right or you're wrong, and can be easily remedied.

There's no reason why you can't bump up your math subscore by close to 10 points with a little bit of work

Hey Lindsey,

Glad that helped! And I'm sorry to hear about that commute...that's a ton of time to spend traveling (but congratulations on the new job)!

As far as first steps on this, obviously it's tough to give one-size-fits-all advice, but I think you can get a lot out of doing something like this:

1) Take a set of problems you've already done and revisit them, thinking about:

-Where did the wrong answers come from? Which mistakes did you make or could you have made that would have gotten you to a popular trap.

-How could you rewrite the question to add a step or subtract a piece of information and create a harder question?

-If it's a problem solving question, how might they phrase it in a Data Sufficiency context? Or if it's a Data Sufficiency question, how would you create a good trap answer in a Problem Solving context?

***Now, note that this won't work particularly well for EVERY question, but you'll find a few out of every set of ten where you can do this and gain some real insight into the test. On others, just thinking about the question less from a "I must do this" perspective and more of a "let's think about this" mindset should help.


2) Write 5 Data Sufficiency questions that you think would trap a friend (or some folks here on BTG).

There's something about having to write a question and think about the traps inherent in the DS format that gives you a real insight into the author's playbook. You're trying to make people think they have more information than they do, or you're trying to hide information that they have but won't think they will. Either way, you start to see some of the ways in which the test will cleverly suggest (but not give) or brush past (but still give) information in the statements, and that's great insight to have.

3) I swear this isn't self-promotion but check out some of the Veritas blog articles on here and at www.veritasprep.com/blog because we try to model this thought process frequently in those articles. And that will give you some concrete examples to work off of...

Keep us posted!