Know the GMAT Code: Work Fast on IR Tables

by on February 8th, 2017

Gaussian (bell) graphIn today’s latest installment of our Know the Code series, we’re going to talk about the most efficient way to tackle Table problems in the Integrated Reasoning (IR) section of the GMAT.

First, try out this Integrated Reasoning (IR) Table problem from the GMATPrep® free practice exams.

A timing note: If you’re planning to guess on 3 questions in the IR section, then you can give yourself 3 minutes and 20 seconds to do this problem.

And a logistics note: On the real test, you’ll be able to sort by the different columns in the table. That’s not possible in a blog article, so just do your best as is, but note that a question like this one can be done in much less time than 3 minutes and 20 seconds if you’re taking advantage of the ability to sort the data.

“The table summarizes information in several categories about the 9 stores in a small grocery chain. The table also includes chain-wide averages where appropriate.”

“For each of the following statements, select True if the statement can be verified to be true based on the information provided. Otherwise, select False.”

How did you do?

It was a real pain to do that without being able to sort the columns, wasn’t it? Use that knowledge to your advantage. On the real test, don’t even think about doing a Table question without Reflecting as to how sorting the table will make it easier for you to answer the various questions.

1-second Glance. It’s a table. There are some numbers and some yes/no designations. (Don’t learn all about the data yet.) You might notice that the table includes a final row with averages for the columns that contain numbers. Presumably there will be at least one statistics question in the mix.

Read and Jot. The table has data about some grocery stores. The question asks what can be proved True based on the info provided. (Still don’t learn all about what kind of data is included—not yet!)

Here’s an important step. Table questions have three parts, and you have to answer all three correctly in order to earn any points on this problem. So as you read the three statements, reflect on whether you think any might be too hard or might take too long.

There are three possible scenarios:

  1. You think you can do all three in a reasonable amount of time: Go for it.
  2. Two look straightforward, but you’re not sure about the third: Make a judgment call. If you think the two are easy and will be pretty fast, then it might be worth doing them and guessing on the third one, as you’ll still have a 50/50 shot of getting it right. But if you’re not really confident on the two easier ones, then go to scenario 3.
  3. Two or more of these look annoying / hard / time consuming: Guess immediately on all three and move on.

For example, on this problem, let’s say that you have no idea what “negative correlation” means, so you know that you’ll have to guess on the second statement. If you feel very comfortable with the first and third statements, you might move forward anyway and just know that you’ll have a 50/50 shot at this problem.

If, on the other hand, you’re at all uncomfortable with either of the other statements, then guess and move on right now.

Let’s assume that we are going to do this one; otherwise, I don’t have anything else to write. :)

On Table problems, don’t jot down any info from the table, but do jot down what the question stem wants you to do. In this case, I would just write down “True / False” or “T / F.” Taking 5 seconds to do that helps me to focus on my task.

Next, read the first statement.

“In each store whose average customer age falls between 34 and 36, the number of self-check express lanes is above average.”

Reflect. Now, go look over the table. What part of the data do you need? And is there another order that would be better than the current sort?

The first column shows stores—each store gets one row. The final column shows average customer age—and only some of the stores fall between ages 34 and 36. In this case, it would be really beneficial to sort by that final column so that you can easily review just the age range in question. Here you go:

It’s a million times easier! (Okay, I might be exaggerating a little. But it’s a lot easier. :) )

All we care about are stores A and F, because they’re the only two with average ages between 34 and 36. You might be thinking: Well, if there are only two, I could have just scanned for that.

But you don’t know in advance that there are only two! If you just scan without sorting, you might miss one without realizing it.

Okay, now let’s answer the question. The statement says that the number of self-check express lanes is above average. Is it? The average is 6.89. Store A has 8 and so does store F, so yes, they are both above the average.

The first statement is True.

Here’s the second statement.

“There is a negative correlation between the number of self-check unlimited lanes and the average customer age.”

Correlation is a statistical term and it basically means: two sets of numbers share some kind of trend or pattern. In a positive correlation, when one set of numbers increases, the other set does, too. For example, in childhood, age is positively correlated with height. In general, as children get older, they also get taller. (A positive correlation could also refer to two sets of numbers simultaneously decreasing.)

A negative correlation, on the other hand, means that these two sets of numbers don’t go in the same direction: As one increases, the other decreases.

So this statement asserts that the two data sets in question don’t follow the same trend; in other words, when one goes up, the other goes down.

You can sort by either of the columns, but think about it for a moment. Is it easier to use one vs. the other?

Technically, it doesn’t really matter. But we already sorted by age for the first statement … so it’s easier just to use the table as-is! Here it is again so you don’t have to scroll up:

The age column is going up. Is the unlimited column mostly going down? (Ignore the “averages” row.)

Yes! In general (with a couple of exceptions), the numbers in the unlimited column go from higher to lower as the age column goes from lower to higher. This is, indeed, an example of negative correlation.

The second statement is True.

Here’s the third statement:

“Stores in this table that have fewer self-check express lanes than the chain-wide average are less likely to have restaurants than stores that have more self-check express lanes than the chain-wide average.”

Ugh—this one’s like a sentence correction problem. Follow the modifiers!

“Stores in this table that have fewer self-check express lanes than the chain-wide average … ”

Express column, fewer than the average. Okay. And we need to compare them to…

“ … stores that have more self-check express lanes than the chain-wide average”

Oh, that’s the same group. Great, so I want to sort by the express column. And then what am I checking? Oh, which sub-group is more likely to have restaurants. Got it. Here’s the sort:

I love that the average row sorts in with the rest of the data. Makes my life easier!

The three stores above the averages row have a below-average number of express lanes. Of these, all three have restaurants—that is 100% of the below-average stores do have restaurants.

I don’t even need to check the above-average stores. The answer already has to be False. The third statement says that the below-average stores are less likely to have restaurants, but 100% of them have restaurants! The above-average ones can’t have restaurants at a rate above 100%—that’s impossible. (And, in fact, only one of these has a restaurant, so they are not more likely to have restaurants than the below-average group.)

The third statement is False.

As on the rest of the GMAT, expect to guess outright on a small number of questions. As of this writing, the IR section is not as important as are the Q and V sections, so you can reasonably guess on 3 or 4 of the 12 questions in the IR section. (In future, that might drop to 1 to 3.)

If you guess on 3 questions, you’ll be left answering 9 of 12 questions in 30 minutes, for an average of about 3 minutes and 20 seconds per question. Some questions, though, can be done faster, leaving you more time for the more complicated ones.

Table questions are usually an opportunity to save some time (though not always). Aim to answer these in closer to 2 minutes. It will be key to figure out how to sort the table in the way that will allow you to answer accurately and efficiently. If you don’t regularly work with tables and think about sorting data, start practicing!

Key Takeaways for Knowing the Code:

(1) Tables will always have three statements, and you will always have to answer all three correctly in order to get points on that question. So don’t just start doing the first one first. Read all three, then decide whether you want to do this problem at all.

(2) Keep Reflecting! All three of these statements can be made easier by figuring out how best to sort the data. Investing a little time to think about that will save you more time on the back end. Get through something like this in, say, 2 minutes, and that’s at least a minute that you can now spend somewhere else in the section.

(3) Turn any knowledge you gain into Know the Code flash cards:

* GMATPrep® questions courtesy of the Graduate Management Admissions Council. Usage of this question does not imply endorsement by GMAC.

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.