The Case of the Red Shoes, or, Causation vs. Correlation on the GMAT

by on October 23rd, 2012

red-shoesGuest post from: Jonathan Bethue

It’s Monday morning. You’re checking your email when your boss walks up to you and says, “You’re fired.”

You look up in stunned silence. “What? Why?”

Well it isn’t easy to explain, unfortunately, but it seems that you are depressing our sales.”

What? How!?” you demand.

Your boss replies, “I’m not sure. All I know is that you started wearing those red shoes six months ago, and since then, sales have gone down. Therefore those loud shoes must be depressing our sales.”

While this scenario might seem a bit ridiculous, the logic involved in the boss’ decision is surprisingly common. What seems absurd (and possibly litigable!) in the example above is actually a common method of argument. What we’re dealing with here is confusion between causation vs. correlation—an all-too-frequent visitor in the land of GMAT critical reasoning.

In the scenario above, your boss concluded that one event caused another—that is, that your red shoes caused the depression in sales—based only on the evidence that the two events occurred one after the other. On the GMAT, you’ll also see arguments conclude that one event caused another based on the fact that two events happened at the same time. In effect, your boss is relying on an unwarranted assumption that two events that occur together (that is, two correlated events) must also have a certain cause-and-effect relationship.

The idea that correlation implies causation rears its head often when we engage in debate. Ever hear someone say, “Man, back in the 90’s we had a great economy, budget surpluses, zero wars, and occasionally music videos on MTV. Clinton sure was a great president!” Or, how about this common assertion from the right: “There were zero attacks on American soil after Bush invaded Iraq. He kept us safe!” Or, for a less political example, think about the fans of the Dallas Cowboys, who tried to ban quarterback Tony Romo’s former girlfriend Jessica Simpson from attending football games, as Tony supposedly performed dismally whenever she was there.

Easy as the flaw is to see when dealing with abstractions, correlation is often compelling in our daily experience. Try telling the cancer survivor who prayed every day, or the basketball player who won the championship while wearing his lucky socks, that they’re operating under an “unwarranted assumption.”

But on the GMAT, being able to identify this flawed logic is crucial—and will do wonders for your GMAT score! Critical Reasoning questions are BFF’s with the causation / correlation fallacy; you’ll come across causation / correlation on both strengthening and weakening questions. When the GMAT asks you to deny an assumption or undermine a conclusion, often the correct answer will exploit the causation / correlation fallacy. When they ask you to strengthen an argument, you may have to strengthen the correlation.

Let’s say a question tells you that a new plant food guarantees faster growing and healthier plants. You use the plant food, the plants grow healthier. Therefore, the plant food caused the improved growth.

The correlation should be clear.

If the question asks you for a statement that would weaken the conclusion, think: How can we make this correlation into a fallacy? Since you’re looking to undermine the conclusion, try to find the answer choice that suggests an alternate cause. Perhaps choice C says something like “at the same time the new plant food was used, the plants were given more water and sunlight.” This lessens the correlation between the new plant food and the healthier growth by introducing the possibility that the increase in water and sunlight caused the plants to grow healthier.

If this were a strengthening question, we would likely need something that strengthens the correlation. An answer choice that strengthens the correlation might read something like, “In an experiment with a control group that had identical plants and environmental conditions, it was found that the plants that received the plant food grew faster and larger.” This would make it much less likely that the correlation is pure coincidence—and make it much more likely that the new plant food caused the healthier growth.

Keep an eye out for this sneaky little fallacy on reading comprehension questions too. With a bit of practice, you’ll find it stands out more than a pair of bright red sneakers.


  • Great Insight.

  • this article makes so much sense while clarifying a very subtle (yet important) point succinctly!

  • Great article. Could you pls share a sample OG question with this concept and apply strengthen/weaken to it?


  • Thanks for the note paddle_sweep,

    If you take a look at OG11 q116 / OG12 q 118, you'll find a weakener question that relies on your ability to recognize that correlation is not causation. The evidence provided states that the firms with PRP contracts showed higher productivity last year. You are asked to weaken the argument that PRP leads to an increase in productivity. The correct answer, choice D, presents another factor that coincides with the increase in productivity, and this could provide an alternate cause for this increase in productivity; the PRP contracts were introduced at the same time as another change, modernization, so the link between PRP--->productivity is effectively weakened.

  • It would be simpler if you could just explain post hoc ergo propter hoc and cum hoc ergo propter hoc and help people understand that temporal succession does not translate into causation and happenstance does not translate into causation!
    Informal fallacies abound in the GMAT and explaining them by giving them their proper names and making people understand how to deal with them makes life a lot easier for both the teacher and the learner.

  • This is a great article. It definitely reassures GMAT preppers that these types of questions are not as intimidating as they may seem at first, and that eventually they'll "stand out more than a pair of bright red sneakers" with more and more practice. I say this because I am currently finishing up my first month of GMAT prep, and CR questions are currently a bit intimidating to me.

    Thanks, Josh.

  • great article. thanks for posting this

Ask a Question or Leave a Reply

The author Knewton 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.