The definition of Anecdotal Evidence, via Wikipedia:
(2) Evidence, which may itself be true and verifiable, used to deduce a conclusion which does not follow from it, usually by generalizing from an insufficient amount of evidence. For example “my grandfather smoked like a chimney and died healthy in a car crash at the age of 99” does not disprove the proposition that “smoking markedly increases the probability of cancer and heart disease at a relatively early age”. In this case, the evidence may itself be true, but does not warrant the conclusion.
Contrast that with Scientific/Empirical Evidence:
Scientific evidence is evidence which serves to either support or counter a scientific theory or hypothesis. Such evidence is expected to be empirical and properly documented in accordance with scientific method such as is applicable to the particular field of inquiry.
Anecdotal evidence is using your personal experiences and stories to illustrate your point. Empirical evidence is measured, unbiased, and replicable. Yes, numbers can lie, especially if you mess up the calculations, but making conclusions based off one person’s anecdotal evidence is just dangerous.
Anecdotes are fine, just have the numbers to back them up.
[Img credit: Sébastien Magro via Flickr]