Your best weapon on flaw questions in the LR section is to have a reliable familiarity with the most heavily tested flaws. Today, I’m going to look at one of those flaws -correlation/causation - but I also want you to take a moment to appreciate what it can teach us about flaw questions in general, so let’s take a step back first and look at the forest before having a closer look at the particular tree.
One idea that can prove really helpful in flaw questions in general is to watch for a mismatch between the premise(s) and the conclusion. There’s usually going to be one; there almost has to be, because the premises have to be true, but because it’s a flawed argument, there’s a good chance the conclusion is false. Identifying the mismatch can point you toward any number of flaws. For instance:
Part-to-whole (or whole-to-part). The premise is about the part (or the whole), while the conclusion is about the part.
Sufficient/Necessary Confusion. Most often, the premise will include a conditional statement, and the conclusion will include its inverse or converse (incorrect negation or reversal, respectively)
Unrepresentative sample. The group the conclusion is about won’t be the same as the group the evidence is drawn from (e.g. predicting the result of a national election based on a survey of residents in a single state).
“Tunnel vision.” The premises might tell you that something is possibly, or probably true, but the conclusion will tell you that it definitely is.
And of course…correlation/causation. The premises will indicate a correlation, but the conclusion will have cause/effect language that doesn’t show up in the premises.
As an example, check out Preptest 52, Section 1, Question 2. The conclusion is the last sentence (“So” is a conclusion indicator). It’s explicitly states that there’s a causal relationship there. Napping is the cause, and insomnia is the effect. But in the passage’s other sentences, there’s no cause/effect language. The people who nap are more likely to get insomnia than those who don’t. No word(s) indicating causation. So the conclusion shifts from correlation to causation.
This question can still trip you up, because even if you catch the correlation/causation problem, (B) might be a tempting answer, since it includes the word “cause.” But (B) misrepresents the argument by attributing the same cause to “all” cases of insomnia. The passage doesn’t take that “all or nothing” stance; it says that napping is “very likely” to be the cause.
One last correlation/causation tip: When you see the correlation/causation flaw, the conclusion will be X caused Y. The first thing you should consider is whether they got it backward. Is it possible that Y caused X? In this case, it would make perfect sense. Sure, if you nap a lot, maybe that will make it harder to fall asleep, but on the other hand, it works well the other way: If don’t sleep well at night, that may well cause you to nap during the day. (D) is your answer here.
The other frequent possibility is that both X and Y are effects of Z, a usually-unmentioned third cause. But the simple reversal of possible cause & possible effect is the most likely, at least on the LSAT. Bear in mind, however, that the reversal has to make sense. For instance, there’s a correlation between height and basketball ability, from which one might conclude that being taller makes one better at basketball. You can consider the opposite, but you should quickly discard it - Practice basketball all you want, but no matter how good you get, it’s not going to make you any taller.