Looking for suggestions to improve quant speed/score

Hey guys,

Great discussion so far! If I could add a few suggestions...

1) Data Sufficiency

Another great way to get your mind in tune with DS questions is to train yourself to "think like the testmaker" by turning each DS question into a series of a few questions - ask yourself how they could change one word, symbol, etc. to elicit an entirely different answer. For example, there's an OG question that reads:

For nonzero integers m and n, is m/n > (mn)^4?

1) m/n > 0
2) m < 0

For this one, since we know that m and n are integers, there's no chance that m/n could ever be greater than (mn)^4. Thinking logically, on the left we take the same m and divide it, whereas on the right we multiply it by m^3 n^4. Even if either is negative, the right hand side will end up positive because of the even exponent. At worst we can construct a tie if all values are 1 or -1. The answer is D.


But to become a true master of DS, ask yourself how they can use the same setup to elicit a different answer. One way - if they just changed the direction of the inequality in the question to:

Is m/n < (mn)^4?

Usually the answer will be "yes" - take 4 and 2, 4/2 is much less than (4*2)^4. BUT, since we know that we can construct a "tie" with all 1s, then we can still get that answer "NO". Here, we learn that with inequalities you have to be careful because the yes/no question format can be tricky with the potential for a "tie".


2) Timing

One of my favorite timing drills is one I call "Quick First Step", in which you take 10 or 20 quant problems and start on each in 30 seconds, then move on to the next. When you're done with 30 seconds/question, go back and finish the question. Here, your goal is first to train yourself to begin questions quickly by drawing relationships and identifying your known quantities - so often people spend 30 seconds reading, 15 seconds worrying, and then it's almost a full minute before they start "doing". You want to actively read and begin work on a problem, both for pacing and confidence reasons. Second, you'll also learn which mistakes you tend to make when setting up a problem quickly. I'd argue that most quant mistakes come in the first 30 seconds and last 30 seconds you spend on a problem - you're either setting it up incorrectly or beginning with false assumptions, or you're missing steps or leaving a problem short when you finish. This drill allows you to focus on that first-30-seconds portion that can be so crucial.

Thanks for your response. In other words, the more practice questions I solve the faster is a response time. Make sense. I often wonder how on earth someone has time to rephrase a question and then evaluate it in DS

Yeah, DS questions can be harsh sometimes However, the more answer explanations you read, the more equipped you are. There are special chapters in MGMAT guides also, as I remember, you can check how to rephrase there. Keeping attention to small details is very important when solving DS questions. Always notice words like "positive", "integer" and so on, this may be crucial.
Forgot to say, I think your goal is quite achievable since you have decent verbal and quant is much easier to master. Good luck!

Don't just do OG problems--reread the MGMAT books and find out how to solve all the most common types of problems. You need to be able to recognize all the ways the GMAT uses to make simple questions look more obscure...for example, how complicated-looking questions can be simplified to (a^2 - b^2) which obviously equals (a+b)(a-b). And you MUST memorize all the equations and formulas, exponent rules, characteristics of geometric shapes, etc.

Yeah, I noticed that about DS questions. By the way, I find it time consuming to draw tables for work/rate and speed/distance problems. By the time I fill it in, like 60-80 seconds are gone.

Great post Brian! I think the 30 second "Quick First Step" drill is going to highlight one's weaknesses and gaps in the knowledge of basic math concepts. I'll definitively try that.