6
$\begingroup$

Who said:

"Let the Data Speak for Themselves"

-- Ronald Fisher or John Tukey?

$\endgroup$
5
  • 6
    $\begingroup$ It is highly unlikely Fisher would have said this. Tukey said many things that might sound like it, but I can find no direct quotation. A lot of people have said precisely this phrase (without any attribution at all), as a Google search will attest. At least one person claims this quotation appears in Tukey, citing his EDA book (statisticsviews.com/details/feature/5393251/…). I doubt he's right, for here's one EDA quote: "NO BODY OF DATA TELLS US ALL we need to know ABOUT ITS OWN ANALYSIS" (emphasis in the original, p. 115). $\endgroup$
    – whuber
    Commented Mar 29, 2014 at 21:11
  • 2
    $\begingroup$ A modern claim from a Wired Magazine article in 2008 is that "With enough data, the numbers speak for themselves" (wired.com/science/discoveries/magazine/16-07/pb_theory). For counterpoint see Tim Harford's article (yesterday) at ft.com/cms/s/2/…. Harford concludes, "that [quotation] seems hopelessly naive in data sets where spurious patterns vastly outnumber genuine discoveries." $\endgroup$
    – whuber
    Commented Mar 29, 2014 at 21:50
  • 2
    $\begingroup$ The historical carefulness of the statisticsview.com article can be judged from its reference to the Rothamsted Research Centre. In Fisher's time it was the Rothamsted Experimental Station. I agree with @whuber that neither quotation sounds like Fisher or Tukey. $\endgroup$
    – Nick Cox
    Commented Mar 30, 2014 at 1:43
  • $\begingroup$ Thanks. In fact, I became confused, because I got this in the net archive.org/details/ost-engineering-book-1 : "What one can learn from a data set depends critically on what question one asks. R. A. Fisher once said "let the data speak for themselves". It appears that the data are more than capable of this, but they do not speak spontaneously; they need someone who is willing to ask the right questions, suggested by cogent prior information." (Lecture Notes in Statistics, Edited by J. Berger, S, Fienberg, J. Gani, K. Krickeberg, and 6, Singer - 48 - pg.75) $\endgroup$
    – user42729
    Commented Mar 30, 2014 at 21:33
  • 1
    $\begingroup$ It's most likely that anyone who says something like this is, or should be, saying it in context and reacting to the opposite view which is being overplayed at the time in some debate. In particular, Fisher and Tukey both counselled against over-interpreting particular datasets, Fisher urging the importance of a significant result being reproducible and both Fisher and Tukey being fully aware of the need for looking at other datasets and using whatever other guidance is available, including understanding of mechanisms. $\endgroup$
    – Nick Cox
    Commented Feb 28, 2022 at 13:07

3 Answers 3

3
$\begingroup$

The earliest I can find in Google Books is from The Lookout, Seamen's Church Institute of New York and New Jersey, 1915 so unlikely to be John Tukey, who was born that year.

A very similar quotation appears to come from records of the Protestant Episcopal Church in the United States of America in 1917 so it is possible both are reprinting something earlier.

$\endgroup$
1
  • 3
    $\begingroup$ I could see Tukey writing this the year he was born. $\endgroup$ Commented Mar 29, 2014 at 21:49
1
$\begingroup$

For what it's worth, Jaynes (Logic of Science, 2003) attributes it to Fisher:

R. A. Fisher’s maxim: ‘Let the data speak for themselves!’ which has so dominated statistics in this century. The data cannot speak for themselves; and they never have, in any real problem of inference.

$\endgroup$
0
$\begingroup$

Google finds this paper on pubmed, with this exact quote in title which attributes it to Aristotle and Newton:

This contrasts sharply with the second approach suggested by Aristotle and revived by Newton in the 18th century that places data at its center: "Let the data speak for themselves."

$\endgroup$
1
  • 2
    $\begingroup$ That's pretty weird as a summary of Newton's contributions in particular. The leading examples being perhaps Albert EInstein and Mark Twain, great people are often attributed quotations no one can find exactly. Of course, we typically don't have recordings of what they said in conversation, but we can search much of what they wrote. In the case of Einstein there is an entire section on what he probably didn't say in Alice Calaprice's definitive collection. $\endgroup$
    – Nick Cox
    Commented Feb 28, 2022 at 13:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.