Integrated Reasoning: Table Analysis

Stacey Koprince —  January 7, 2013 — 13 Comments

GMAT IR tableI’ve been wanting to do this problem with you for a while, but I’ve been delaying because well, you’ll see when you get to the table. It takes a lot of work to recreate that in a blog post. :) But that ridiculously large table is also the reason why I want to talk about this one “ so let’s test it out!

Try the problem

This problem can be found in GMATPrep© 2.0. Normally, on table problems, you’re able to re-sort the data according to various column headers. I can’t recreate that functionality in a blog post, but I’ll give you a hint: sorting actually isn’t necessary for this particular problem.

Set your timer for 2.5 minutes. (You can take up to 3 on this one, but if you do decide to use extra time, use it wisely! Otherwise, it’s better to cut yourself off faster.)

The Consumer Price Index (CPI) measures the average prices of goods and services purchased by consumers. In the United States, the CPI-U calculates the CPI for all urban customers.

The CPI-U is calculated based on prices of food, clothing, shelter, fuels, transportation fares, charges for doctors’ and dentists’ services, drugs, and other goods and services that people buy for day-to-day living. All taxes directly associated with the purchase and use of items (such as, in the United States, sales taxes) are included in the index. An increase in CPI-U by a certain fractional amount means an increase by that fractional amount in overall prices within the relevant category.

For analyzing general price trends in the economy, seasonally adjusted prices are usually preferred over unadjusted prices because adjusting eliminates the effect of changes that normally occur at the same time and in about the same magnitude every year “ such as price movements resulting from climatic conditions, production cycles, model changeovers, and holidays.

 

Percent Changes in CPI for All Urban Consumers (CPI-U), US City Average

Category Seasonally adjusted changes from preceding month

Unadjusted 12 months ended Sep 2010

Mar 2010

Apr 2010

May 2010

Jun 2010

Jul 2010

Aug 2010

Sep 2010

All items

0.1

 -0.1

 -0.2

“0.1

0.3

0.3

0.1

1.1

Food (all)

0.2

0.2

0

0

“0.1

0.2

0.3

1.4

Food (at home)

0.5

0.2

0

“0.1

“0.1

0

0.3

1.4

Food (away from home)

0

0.1

0.1

0.1

0

0.3

0.3

1.4

Energy (all)

0

“1.4

“2.9

“2.9

2.6

2.3

0.7

3.8

Gasoline (all types)

“0.8

“2.4

“5.2

“4.5

4.6

3.9

1.6

5.1

Fuel oil

0.7

2.3

“1.4

“3.2

“1.6

0.9

0.8

11.8

Energy services

1.4

“0.5

“0.5

“1.6

0.8

0.4

“0.8

1.5

Electricity

2.1

0.7

“0.4

“2.2

0.5

0.2

“0.3

1.1

All items less food and energy

0

0

0.1

0.2

0.1

0

0

0.8

New vehicles

0.1

0

0.1

0.1

0.1

0.3

0.1

2.1

Used cars and trucks

0.5

0.2

0.6

0.9

0.8

0.7

“0.7

12.9

Apparel

“0.4

“0.7

0.2

0.8

0.6

“0.1

“0.6

“1.2

Services less energy services (all)

0.1

0.2

0.1

0.1

0.1

0

0.1

0.8

Shelter

“0.1

0

0.1

0.1

0.1

0

0

“0.4

Transportation services

0.4

0.4

0.4

0

0

0.1

0.3

3.0

Medical care services

0.3

0.3

0

0.4

0

0.2

0.8

3.7

 

For each of the following, select Yes if the statement is inferable from the given information. Otherwise select No

Yes No
The changes in seasonally adjusted prices for used cars and trucks between March 2010 and September 2010 were in most cases less in magnitude than the changes in seasonally adjusted prices of new vehicles for the same period.
The seasonally adjusted CPI-U for all items was higher in March 2010 than in the previous month.
The seasonally unadjusted change in the price of new vehicles in August 2010 over the previous month was about the same as the seasonally unadjusted change in the price of food away from home over the same period.

 

The Solution

This table is ridiculous isn’t it? And three paragraphs to start, before I even get to the question? There’s no way I’m possibly going to look through all of this data in 2.5 minutes. (Hint. That’s one clue to solving these kinds of questions!)

In general, on tables, read the information that tells you what the table is about and look at all of the big labels (such as the column headers), but do NOT try to read / remember / understand all of the actual data points. You’ll use whatever data you need when you get a specific question about that data “ but not before!

In this case, the first paragraph tells us that this CPI-U thing calculates average prices on a bunch of things for a certain population (urban consumers in the US). The second paragraph starts off by listing a bunch of categories “ ignore them. If you need to know the specific categories, you can come back here later. In fact, ignore most of the details here until we figure out what the questions are about.

The third paragraph explains that we generally prefer to use seasonally adjusted prices. A quick glance at the table shows us that it includes both adjusted and unadjusted numbers. It also explains why “ ignore this. Many, if not most, people will have used a minute or more at this point, so don’t spend that time now. If you need to know why, you can come back here later to understand.

Finally, glance at the table labels. The columns list a bunch of individual months from 2010 and those are labeled seasonally adjusted changes from preceding month. Notice two important things here: These numbers are seasonally adjusted, not unadjusted, and each monthly figure is based on the change from the preceding month. The final column is for a whole 12 month period and these numbers are unadjusted. Finally, there are a million individual categories listed in the rows. L

I’m not 100% sure that I understand the difference between adjusted and unadjusted “ but I don’t care about that right now. I’ll figure that out if I need to based on a specific question. Just keep going!

Now it’s time to examine the statements; here’s the first one:

The changes in seasonally adjusted prices for used cars and trucks between March 2010 and September 2010 were in most cases less in magnitude than the changes in seasonally adjusted prices of new vehicles for the same period.

Let’s see. They mention two categories: used cars and trucks and new vehicles. Great, I only need to look at those two rows. They also specifically mention adjusted for these, so I only care about the monthly columns. Scan until you find the right rows; luckily, they’re already right next to each other in the table.

The question asks about the magnitude of the changes in the two categories, so we need to compare them, month by month, to see which changed more and which changed less. For March 2010, the magnitude of change for used cars and trucks was 0.5, but only 0.1 for new vehicles. For that month, then, the magnitude of change was larger for the used cars and trucks. Compare the other months. In each case, the magnitude of change (the distance from zero) is larger for used cars and trucks. (This was true even in September 2010 “ the change was negative, but it was -0.7, which is 0.7 units away from 0. The magnitude of change for new cars was only 0.1.)

The statement says that the change for used cars and trucks was in most cases less in magnitude, but that’s not true. During all of the months in question, however, the change in magnitude (from zero) for used cars and trucks was greater.

Select No for statement 1.

Here’s the second statement:

The seasonally adjusted CPI-U for all items was higher in March 2010 than in the previous month.

This time, the statement is directing me to the all items category in the month-by-month (adjusted) timeframe. That’s right at the top of the table. We’re specifically asked about March 2010 compared to the previous month, but the table doesn’t include the previous month, so how can it tell us anything at all about February?

Read the labels! Right above the months, the table says Seasonally adjusted changes from preceding month. (Emphasis added.) In other words, the figure for March 2010 is based upon the change from February 2010. That change is positive, so it is true that the number for March is higher than the number for February.

Select Yes for statement 2.

And finally our third statement:

The seasonally unadjusted change in the price of new vehicles in August 2010 over the previous month was about the same as the seasonally unadjusted change in the price of food away from home over the same period.

Read carefully! What data do they want now? Seasonally UNadjusted this time, not adjusted! The only data we have for the unadjusted category is based on the 12 months ended Sep 2010 “ we don’t have month-by-month data! Therefore, I can’t tell anything at all about what happened in August 2010 vs. July 2010 “ not for the UNadjusted data.

Note something very tricky. If you gloss over the Un in the word unadjusted, you might just go check the Aug and Jul columns for new vehicles and food away from home. For new vehicles, the change was 0.3. For food away from home, the change was 0.3. Those two numbers are the same “ and so you would pick Yes, and you would fall into the trap. We don’t have any unadjusted data for the month of August alone.

Select No for statement 3.

The answers to the three statements are No, Yes, No.

Finally, notice something. How many rows and cells did you need in order to answer all three statements here? The first statement required us to use most of the cells in two adjacent rows. The second statement required us to use one cell in one row. The third statement required us to use nothing “ but, even if we fell into the trap, we’d still only have used two cells in two different rows (one of which was a row we’d already used previously).

In other words, we never have to look at the vast majority of the data in this table. They could have cut out 3/4 of the rows without changing the question at all “ except, perhaps, reducing the anxiety of a lot of test-takers when they saw a table with nearly 20 rows.

 

Key Takeaways for Table Analysis questions:

(1) You will NOT need to use all of the data, so when you see a crazy number of rows or columns, don’t let yourself get anxious; instead, remind yourself that you won’t need to look at most of this. Ditto for those three paragraphs of text at the top. Don’t bother to read the table thoroughly or take in all of the data “ just know what kind of information is available, then return to whatever you need based on the question statements.

(2) For many table questions, the ability to re-sort the data is critically important to our ability to answer these questions efficiently. Sometimes, though, we don’t need to sort the data at all “ as in this case. Learn how to decide when and how to re-sort the data.

(3) Read carefully. When you have a table, by definition, the questions are going to hinge around whether you’re examining the right cell or cells from the table. Make sure you read the statements carefully with this question in mind: What group / category are they asking me about now?

 

* All quotes copyright and courtesy of the Graduate Management Admissions Council. Usage of this material does not imply endorsement by GMAC.

 

 

Stacey Koprince

Posts

Stacey Koprince is an Instructor and Trainer as well as the Director of Online Community for Manhattan Prep. She also co-manages the company's GMAT curriculum and product line. She has been teaching various standardized tests for more than fifteen years and her entire teaching philosophy can be summed up in five words: teaching students how to think.

13 responses to Integrated Reasoning: Table Analysis

  1. Hi Stacey…First of all i must say great job .
    For the option “The changes in seasonally adjusted prices for used cars and trucks between March 2010 and September 2010 were in most cases less in magnitude than the changes in seasonally adjusted prices of new vehicles for the same period” I choose “YES”,because the parent question stem asks”“For each of the following, select Yes if the statement is inferable from the given information. Otherwise select No.”The parent question stem does not asks for “True and False”.Clearly it is inferable from table that the “The changes in seasonally adjusted prices for used cars and trucks between March 2010 and September 2010 were in most cases less in magnitude than the changes in seasonally adjusted prices of new vehicles for the same period” is not true but still inferable.

    • Hi, Nilay (and Stephanie and Asad, below!)

      When the test asks us to infer something (whether on CR, RC, or IR), it is asking us to find something that MUST be true given the information provided. That’s why I used the “true/false” shortcut in the problem. If something IS inferable, then it is something that must be True. If something is NOT inferable, then it is technically something that doesn’t have to be true (as opposed to false), but it’s annoying to write “doesn’t have to be true” every time. It’s easier just to write F for false. :) (But feel free to do something else if you prefer, such as “NNT” for not-necessarily-true.)

      So, no, you’re not actually finding that something is False. You’re deciding whether it MUST be true. If so, then answer Yes. If not, then answer No.

      In the case of this problem, statement 1 says that one thing (call it X) is in most cases less than another (call it Y). In most (all, actually) cases, the opposite is true: Y is less than X. Therefore, we CANNOT infer that X is less than Y.

      Note that they have not asked you WHETHER X is less than Y. That you can figure out, yes. They have stated that X IS less than Y and asked whether the data supports this statement. It does not.

  2. I had the exact same approach as Nilay mentioned in her response. Are we answering the more general parent question or approaching the individual question to a certain T/F answers?

  3. Hi Stacey…First of all Thank you for such a wonderful post..
    For Question 3: It is UNadjusted data which I reckon mean the data is not adjusted and do not consider effect of changes(climatic conditions, production cycles, model changeovers, and holidays) that normally occur at the same time and so it should remain same through out the year. Hence every month the UNadjusted data value has to be same so change in unadjusted price month over month is Zero (0).
    Unadjusted change in the price of new vehicles in August 2010 over the previous month = 0 which is same as seasonally unadjusted change in the price of food away from home in August 2010 over the previous month ( =0).
    Please advise where I am going wrong.

    • You read too much into it, unfortunately. “Unadjusted” doesn’t mean that there aren’t any changes for any reason. It just means that the changes weren’t adjusted specifically for *seasonal* issues. For instance, when a new product first launches, there are all kinds of month-over-month and year-over-year changes (simply because it’s new!!) that don’t necessarily have anything to do with the season.

      How do you know this for sure? The table includes an entire column for UNadjusted percent change. If the change every month is 0, then the change over that 12 month period couldn’t possibly be anything other than 0. But the table shows all kinds of changes. :)

  4. What I interpret from the heading of the questions asked is that we just need to answer Yes, if we can deduce the answer from the provided data and “No”, if we cannot? Please let me know if I am missing something?

  5. Hi all,
    I feel that one should go with the parent question stem.Hopefully Stacey will clear this confusion soon.
    Well i am posting one my doubts from RC here because of see some active members here.

    Which one of the following most accurately describes the organization of the last paragraph?
    (A) One explanation is criticized and different explanation is proposed.
    (B) An argument is advanced and then refuted by means of an opposing argument.

    Please help me to “break down” what option (B) says..

  6. Hi Stacey,

    I have got couple of questions. Could you please clarify these ?

    1. Question stem says whether we can Infer the required information. So, as Nilay and Stephanie have posted above, I chose – YES for statement 1. Are we correct in our approach here ?

    2. “The seasonally adjusted CPI-U for all items was higher in March 2010 than in the previous month.”

    for a second, I got stumped that we can not answer this unless we know the CPI-U for February;However, on reading again I think that detail is not required.

    What if we rephrase statement 2 as – “The seasonally adjusted CPI-U for all items was higher in March 2010 than that of previous month.”

    I believe, we will then need the data for February as well. Am I correct in my assumption here?

    • For Q1, see above.

      For Q2, no, your assumption isn’t correct. :( The table is not actually showing the CPI-U values. It is showing the percent change of the value from the previous month to the current month.

      Percent change, by definition, is calculated based off of two values. In this case, those two values are “this month” and “previous month.” If the percent change is positive, then “this month” had the higher value (but we still don’t know what that actual value is). If the percent change is negative, then “previous month” had the higher value.

Leave a Reply