### GMAT Critical Reasoning: Infer like a Master

Did you know that you can attend the first session of any of our online or in-person GMAT courses absolutely free? We’re not kidding! Check out our upcoming courses here.

Inference questions can appear in GMAT Critical Reasoning, Reading Comprehension, and Integrated Reasoning, so it’s crucial to master what the test is—and is not—asking you to do. The good news? Your goal is the same, regardless of the question type.

Try this problem from the free questions that come with the GMATPrep® software and then we’ll talk about how Inference questions work!

Roland: The alarming fact is that 90 percent of the people in this country now report that they know someone who is unemployed.

Sharon: But a normal, moderate level of unemployment is 5 percent, with one out of 20 workers unemployed. So at any given time if a person knows approximately 50 workers, one or more will very likely be unemployed.

Sharon’s argument is structured to lead to which of the following as a conclusion?

(A) The fact that 90% of the people know someone who is unemployed is not an indication that unemployment is abnormally high.

(B) The current level of unemployment is not moderate.

(C) If at least 5% of workers are unemployed, the result of questioning a representative group of people cannot be the percentage Roland cites.

(D) It is unlikely that the people whose statements Roland cites are giving accurate reports.

(E) If an unemployment figure is given as a certain percent, the actual percentage of those without jobs is even higher.

Let’s start by talking about what you need to do for Inference questions in general. Then we’ll tackle the problem.

Inference arguments will provide you mostly with a series of premises and they’ll ask you to infer or to draw a conclusion in some way.

Now, here’s the key: the test does not want you to infer or conclude in the way that you likely would in the real world. (Did I emphasize that word strongly enough? ?) In the real world, we conclude things that are likely to be true, given certain information.

On GMAT Critical Reasoning, you’ll need to hold yourself to a higher standard: What must be true, based on the information you were given? (Not just what is likely or reasonable to believe.)

Okay, let’s dive into this problem and see how this “what must be true?” concept works.

#### Step 1: Identify the Question

How do you know that this is an Inference question in the first place?

The question stem refers to the answer choices as possible conclusions:

GMAT Critical Reasoning (CR) questions may also ask you what can be inferred. (On Reading Comp and Integrated Reasoning, you might see the words suggest or imply in the question stem.)

In all cases, the question is asking you what must be true based on the information you’ve been given so far.

#### Step 2: Deconstruct the Argument

Roland finds it alarming that 90% of people say they know someone who is unemployed.

Sharon’s response starts with the word but, so she doesn’t agree with something that Roland said. What?

According to Sharon, a normal, moderate level of unemployment is 5%. If you know 50 people who work, then 5% = 1 person.

So, if the unemployment rate is 5%, then you’d know 1 person who is unemployed. She seems to think that it’s not a big deal that 90% of people know someone who is unemployed.

Do you know 50 people who work? You may not be close friends with 50 workers, but chances are you do know 50 people who work—your co-workers, your friends and family, acquaintances at the gym, the servers at your favorite restaurant, etc.

Hmm. Sharon seems to think that it’s not a big deal that 90% of people know someone who is unemployed and maybe she has a point! If I know 50 people who work, and if just one is unemployed right now, that reflects an unemployment rate of just 5%—not a very high rate. So most people probably do know someone who is unemployed at the moment.

Note that Sharon doesn’t dispute Roland’s statistic. She just says that his figure isn’t surprising. She thinks it’s pretty normal and explains why.

Here’s what my notes looked like, taken as I read the argument:

The bracketed text at the end was not stated in the argument. This is my own conclusion, in my own words. I use the brackets to signal to myself that I’m doing my own thinking here; I’m not writing what the argument actually stated.

#### Step 3: State the Goal

On Inference questions, the goal is to find the answer that must follow from the given information in the argument.

The most common trap on this question type is an answer that goes too far—what I call a real-world conclusion. A real-world conclusion might be true, and even might be likely to be true, but it doesn’t have to be true…so it’s not the right answer on the GMAT.

#### Step 4: Work from Wrong to Right

All right, let’s dive in.

(A) The fact that 90% of the people know someone who is unemployed is not an indication that unemployment is abnormally high.

Roland is alarmed. Sharon isn’t and she has a reasonable explanation for why Roland’s statistic isn’t actually alarming. This choice fits Sharon’s argument: she doesn’t think that Roland’s statistic indicates that unemployment is unusually high.

(B) The current level of unemployment is not moderate.

This one is a double trap. First, this goes along with what Roland is saying, but the question asks about Sharon’s argument, not Roland’s. That’s trap #1.

Trap #2: We don’t technically know what the current level of unemployment is! The argument says only that 90% of people currently know someone who is unemployed.

If anything, we could possibly real-world infer from Sharon’s argument that the current level is somewhere around 5%, and she calls this level normal and moderate. So a choice that says it is not moderate does not match what Sharon is saying.

(C) If at least 5% of workers are unemployed, the result of questioning a representative group of people cannot be the percentage Roland cites.

This one is directly contradicted by the information given; if you follow Sharon’s math, she shows that a 5% unemployment stat could easily lead to 90% of people knowing that someone is unemployed.

They’re trying to trap you into thinking that Sharon’s “but” means that she disputes Roland’s statistic in the first place. She doesn’t; she just thinks that his interpretation of what the statistic means is wrong.

(D) It is unlikely that the people whose statements Roland cites are giving accurate reports.

This is another trap based on Sharon’s starting word, but. Sharon is not disputing Roland’s statistic, just how he interprets it. He finds it alarming; she finds it normal.

(E) If an unemployment figure is given as a certain percent, the actual percentage of those without jobs is even higher.

In the real world, this may very well be true. The method used to estimate the percentage may be missing certain categories of people or something like that. The argument, though, doesn’t provide enough information to conclude that this must be true. I’d call this one a real-world trap: it probably is true a lot of the time in the real world, but it doesn’t have to be true.

The correct answer is (A). Sharon’s argument is set up to conclude that Roland’s 90% figure actually isn’t unusual and so unemployment isn’t necessarily abnormally high. If anything, Sharon’s figures seem to show that the 90% figure goes along with “normal, moderate” levels of unemployment, not unusually high levels.

What did you learn on this problem? Come up with your own takeaways before you read mine below.

#### Key Takeaways for Inference Problems

(1) Know how to identify the question type. On GMAT Critical Reasoning, this usually means some mention of the word infer or some reference to a conclusion in the answers below (not in the argument above).

(2) Lay out the facts very clearly. Articulate to yourself what the facts do and don’t tell you. Don’t assume anything and don’t try to draw any real-world conclusions!

(3) Look for an answer that must follow from the information you were already given. Be wary of “real-world conclusion” traps: ones that might be true or are even pretty likely to be true but don’t have to be true. You’re not looking for a “reasonable” conclusion. You’re looking for a “must be true” conclusion. ?

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

Can’t get enough of Stacey’s GMAT mastery? Attend the first session of one of her upcoming GMAT courses absolutely free, no strings attached. Seriously.

Stacey Koprince is a Manhattan Prep instructor based in Montreal, Canada and Los Angeles, California. Stacey has been teaching the GMAT, GRE, and LSAT  for more than 15 years and is one of the most well-known instructors in the industry. Stacey loves to teach and is absolutely fascinated by standardized tests. Check out Stacey’s upcoming GMAT courses here.