Does association always imply causation?

• A. Sometimes
• B. Absolutely not
• C. Always
• D. It depends on sample size

Answer: (B)

Two variables moving together does not prove that one causes the other. An association simply means they vary in a related way, but this could be due to a third factor (confounding), reverse causation, or biases in measurement. Causation, on the other hand, requires that the cause precedes the effect (temporality) and that alternative explanations are ruled out. In observational data you can see strong, consistent associations, sometimes with dose-response patterns, but unmeasured confounding can still account for the link. Establishing causality typically needs experimental evidence (like randomized trials) or robust methods that mimic randomization. For example, ice cream sales and drowning rates rise together in summer, showing an association driven by a common factor (season), not that ice cream prevents drowning. Large samples can reveal associations, but they do not by themselves prove causation.