The graphic displays the results of a study in which subjects were ask

Hi there,

I really don´t know how to read this graph. I am going to take 1 group (“early in campaign”) to explain how I think it should be read. Would someone please check whether my reasoning is correct?


“Early in the campaign”, there were 4 types of people:
- those who have a level 1 of exposure to tv campaign news coverage
- those who have a level 2 of exposure to tv campaign news coverage
- those who have a level 3 of exposure to tv campaign news coverage
- those who have a level 4 of exposure to tv campaign news coverage

- 65% of those who have a level 1 of exposure “early in the campaign” considered changing their vote during the campaign. Likewise, 35% of this group didn´t consider so.

- 55% of those who have a level 2 of exposure “early in the campaign” considered changing their vote during the campaign. Likewise, 45% of this group didn´t consider so.

- 49% of those who have a level 3 of exposure “early in the campaign” considered changing their vote during the campaign. Likewise, 51% of this group didn´t consider so.

- 40% of those who have a level 4 of exposure “early in the campaign” considered changing their vote during the campaign. Likewise, 60% of this group didn´t consider so.


This reasoning should be repeated with the remaining 3 groups:
- long before election day
- before campaign
- on/near election day


- All this information is presented in relative values (percentages).
- All percentages are taken not only from within each particular group (“early in campaign”) but also within each particular subgroup (“early in campaign” & “exposure 1”).
- We have absolutely no idea what group is larger/smaller (“early in campaign” vs. “before campaign”)
- We have absolutely no idea what subgroup is larger/smaller (“early in campaign & exposure 1” vs. “early in campaign & exposure 2”)

Why this can be counterintuitive: because the x-axis doesn´t represent time.


One thing: I am perfectly aware that in order to answer this type of question, one should be “question-based” and not “prompt-based”. In other words, there are a lot of irrelevant data and we surely don´t need that much info to answer this particular question.

However, I want to make sure I know how to read this graph since I´ve seen this type in other occasions!!

Would you help me?

Thanks!!

Hi minwoswoh,

In the IR section, you'll face at least one prompt that is real "data heavy." You've correctly noted that you don't need to understand all of the data to answer the given question (and oftentimes, you only need to "find" and "pull out" the small chunk of relevant data to get to the correct answer). You should always start with a quick analysis of the table (what does each axes represent and are there any general trends/patterns that you can quickly spot), then read the text description next to the table. While it's tempting to want to try to immediately understand everything involved, you shouldn't put too much energy into trying to define every detail because the attached question(s) almost never require that level of work.

Your approach was fine, although some of the work was unnecessary (given the question that came with this prompt). Test Day requires some flexibility on your part - sometimes you'll have to do more work to answer an IR question, sometimes you'll have to do less. Instead of trying to process every nuance though, take a broad assessment of what you have to work with, then note what information the question requires you to focus on.

GMAT assassins aren't born, they're made,
Rich

Hi Rich,

Thank you so much for taking the time to answer to my question.

I completely agree with you on the process to solve this IR tricky questions. However, what actually happened to me with this question was that I solved this problem with this "question-based approach", but when I finished my practice set and then came back to check my answers, even then I was still overwhelmed with the problem!

That´s what I wanted to check with the forum that I have correctly understood the graph.

To conclude, can we say that there are 16 different groups in the graph (4 subgroups of 4) and that we cannot possibly know which is bigger/smaller?

Thanks!

Hi minwoswoh,

I agree that there's no way to know how many people were polled in each "group"; there's also nothing to imply that the groups were all the same size.

From the first sentence of the text (next to the chart), it's not perfectly clear whether there are 16 groups or the same 4 groups polled 4 times. From the language, I actually think that it's the latter option though. Here's why:

1) The phrase "...subjects were asked about their LEVELS of exposure...at various TIMES during the election." The words "levels" and "times" implies that the subjects were polled about MORE THAN ONE point in time.
2) The phrase "...and whether they considered changing their votes....at ANY TIME during the election..." The words "any time" also imply MORE THAN ONE point in time. That question makes no sense if you're only asking a group about their opinions at just one point in time.

Notice how it doesn't really matter whether you think it's 16 groups or 4 groups though - the answer to the question doesn't change. In that way, GMAT prompts are always carefully worded - the Test results wouldn't be useful if the questions were debatable or open to interpretive 'bias.'

GMAT assassins aren't born, they're made,
Rich

Hi Rich,

Thanks again. You are indeed correct. It makes more sense that there are a total of 4 groups and each was polled 4 times.

With that reasoning, we can make such statements as "before the campaign, the more exposure people receive to television campaign news coverage, the less they consider changing their vote. On the other hand, near or on election day, the more exposure people receive, the more likely they are to consider changing their vote"

Tricky and interesting graph, indeed...

Thank you so much for your help, Rich.

Hi minwoswoh,

Your deductions in your last post are exactly what you're supposed to "see" in these graphs. Those are 2 "big picture" deductions and the IR section tends to reward Test Takers who can see those patterns.

Notice how the skills you used to make those deductions are rooted more in "CR skills" than in "math skills" - that's an important aspect of IR that a lot of people miss. The IR section is more about connecting ideas and spotting logical patterns than about high-level math skills.

GMAT assassins aren't born, they're made,
Rich

Thank you so much for all your help, Rich!

Why is it not subject with 'low' exposure to tv news and 'early in campaign'? I reached the same answer as the official, but now I am second guessing myself and can not seem to come to an agreement with myself. I see the horizontal lines and think- its the same level of people considering changing votes.
Does the negative slope have an impact here? is that why this is not the answer?

Hi deeuk,

While that 'combination' (low level of exposure/early in the campaign) certainly shows a high chance of 'vote changing', it looks a little "lower" in the table than the correct answer (high level of exposure/on-near election day). In real basic terms, the "highest point" is the correct answer.

GMAT assassins aren't born, they're made,
Rich