Data and Evidence

The plural of anecdote is data

"The plural of anecdote is data."
- Raymond Wolfinger, c. 1969–70

The line is the political scientist Raymond Wolfinger's, said in a Berkeley seminar around 1969 or 1970 and recorded by his student Nelson Polsby; Fred Shapiro traced it for the Yale Book of Quotations. The form usually heard today, "the plural of anecdote is not data", is a later inversion that means the opposite, and it is often misattributed to Wolfinger.

Both versions have a point. Wolfinger's is that data does not have a virgin birth: it comes to us from somewhere, and someone set up a procedure to collect and record it. Sometimes that person is a scientist, but she could also be a journalist; a data set is anecdotes gathered systematically. The inversion's point is that anecdotes gathered unsystematically are a biased sample, because the ones that get told are the ones that were memorable or confirmed what the teller expected (Survivorship Bias, Cherry-picking, Confirmation Bias). "It worked for me" is an anecdote; a trial is the plural.

Absence of evidence

"Absence of evidence is not evidence of absence."
- Carl Sagan

In Sagan's words, the expression is a critique of the "impatience with ambiguity" exhibited by appeals to ignorance. Despite what the expression may seem to imply, a lack of evidence can be informative: where evidence would be expected if a claim were true, its absence counts against the claim (see Argument from Silence). A careful search of a room that finds no elephant is good evidence that there is no elephant in the room.

Extraordinary claims require extraordinary evidence

"Extraordinary claims require extraordinary evidence."
- Carl Sagan, Cosmos (1980)

The weight of evidence needed for a claim should be proportioned to how far the claim departs from what is already well established. See the Sagan Standard for its history and the Burden of Proof for whose job it is to supply the evidence.

Also see

Wikipedia: https://en.wikipedia.org/wiki/Anecdotal_evidence

Sources

Fred R. Shapiro (ed.), The Yale Book of Quotations (2006), entry for Raymond Wolfinger.

Linked from

Appeal to Ignorance

Not every argument from absence is fallacious. Where evidence would be expected if the claim were true, and a proper search has found none, the absence is evidence against it; a search of the room that finds no elephant is good evidence th…

Argument from Silence

The sound form is: absence of evidence is evidence of absence where evidence would be expected (Data and Evidence). A careful search that finds nothing is informative; an absence nobody would have noticed is not. Historians use arguments f…

Hasty Generalisation

Unrepresentative samples are usually supplied by a bias rather than chosen: Survivorship Bias shows you only the cases that lasted, the Spotlight Fallacy only the cases that were reported, and Confirmation Bias only the cases that fitted.…

Data and Evidence

Data and Evidence

Data and Evidence

Data and Evidence

Safety claims require evidence. A Massachusetts study comparing localities with and without gender-identity protections found no significant association between those protections and reported criminal incidents in toilets, changing rooms a…