Correlation does not prove causation. Two variables can rise and fall together because a third variable drives both, or because the cause runs in the opposite direction from the one you assumed.
Why it matters
Many health claims, like "more sleep raises grades" or "social media causes anxiety", come from surveys rather than experiments. The course wants you to spot when a causal claim goes further than the data allow.
How the exam tests it
A stem gives a survey result and a researcher's causal conclusion, then asks what is wrong with it. The best answer names either a specific third variable, such as after-school work hours affecting both sleep and grades, or the directionality problem.
Don’t confuse it with
Students confuse the third-variable problem with the directionality problem; a third variable is a separate factor causing both, while directionality means the supposed effect may actually be the cause, as when aggressive children provoke more spanking.