False Cause

RSS | Click here to subscribe to this section's RSS feed

Outside the laboratory, establishing that one factor caused another is more difficult than it may seem. The likelihood that x caused y increases if:

x precedes y,
y only exists in the presence of x,
changes in x change y,
and there are no other plausible explanations for the existence and behavior of y.

However, x can precede y without causing y; and correlation does not establish causation. (The rooster that crows at dawn does not cause the sun to rise.) In experiments, scholars establish causal relations by controlling when, where and with whom x is introduced and by then comparing the behavior of groups exposed to x to comparable groups that were not.