The Secret to Causation/Correlation

You’ve probably read about or seen CR questions that confuse “causation” with “correlation,” but what does it mean in the real world? And why does it matter?

In statistics, let’s say we have data from two variables: x and y. They have a direct relationship. As one increases, the other increases. This could lead scientists to draw a conclusion that one variable causes the change in the other variable. But this isn’t necessary true! In fact, some third variable may be affecting both x and y! In order to prove causation and validate that one variable is indeed directly influencing the other and is the reason behind the detected correlation, further studies would need to be conducted, altering parameters, and recording outcomes.

The “Secret”: Just because X and Y are connected, doesn’t mean X causes Y.

Let’s take a look at a GMAT question:

Question #1

A recent study from USC determined that although only 22 percent of students wore USC Trojans paraphernalia to school on a regular basis, over 65 percent of students who passed their finals were wearing USC Trojans clothing at the time. The study concluded that students who wear USC Trojans clothing are more likely to do well on their finals than students who don’t.

Which of the following would best strengthen the study’s conclusion?

(A) Of the students who failed their finals, none of them have ever been seen in class wearing USC-branded clothing.

(B) The week of finals every year is always on the same week as a big football game.

(C) USC clothing is well-insulated and the classrooms at USC are kept highly air-conditioned.

(D) Many students who did not pass their finals regularly attend tailgate parties.

(E) Students at USC tend to show more school spirit than students at other colleges.

Here our “X” is USC clothing and “Y” is better finals scores. The weakness in the argument is that there is no substantial evidence linking students who wear the clothing to better finals scores. None of the answer choices really give us “substantial” evidence to PROVE a causal relationship, but this piece of evidence would support the study’s conclusion if it were true.

If you chose (B), this piece of evidence implies coincidence and thus weakens the conclusion.

If you chose (C), this suggests students chose to wore the clothing on finals day because of the cold classrooms.

If you chose (D), while we may assume students who attend tailgate parties perhaps wear USC-branded clothing more often, if anything this would weaken the conclusion, since we want to prove that the students who wear USC paraphernalia do BETTER on their finals.

If you chose (E), this is outside the scope of the argument.

(A) is the correct response. This is a classic “causation/correlation” argument.

Remember to consider other possible explanations in Critical Reasoning questions. Sometimes all you’ll need to do to get a question correct is recognize the possibility of an alternative. For more articles on causation and correlation, check out this three part series.

Plan on taking the GMAT soon? We have GMAT prep courses starting all the time. And, be sure to find us on Facebook and Google+, and follow us on Twitter!

Vivian Kerr is a regular contributor to the Veritas Prep blog, providing tips and tricks to help students better prepare for the GMAT and the SAT.