What is your favorite statistical quote?

This is community wiki, so please one quote per answer.

  • 8
    Should this question really be "famous quotes about statistics"? – naught101 Nov 3 '12 at 4:29

148 Answers 148

Figures don't lie, but liars do figure

--Mark Twain

  • 2
    I don't get what is so deep in that one, is it only playing with words ? – robin girard Aug 3 '10 at 19:29
  • I like to think of it as the statisticians equivalent of "guns don't kill people, people kill people" not very deep, but important to realise from time to time – jilles de wit Aug 4 '10 at 9:17
  • So it's an inane platitude used for quibbling over semantics? "Cigarettes don't cause cancer; people cause cancer." "Landmines don't maim people; people maim people." – Lèse majesté Aug 5 '10 at 0:36
  • 4
    well... there is a fine line between inane platitude and profound wisdom. I like the quote for it's poetic quality. Any insight is of secondary importance to me. – jilles de wit Aug 5 '10 at 7:25
  • 1
    Does Twain qualify as a statistician? – Glen_b Nov 3 '12 at 7:00

This is unlikely to be a popular quote, but anyway,

If your experiment needs statistics, you ought to have done a better experiment.

Ernest Rutherford

  • I think it is really popular.... but not it statistic (we don't want to loose our job because physisics are improving their experiment :) ). Anyway I think Rutherford belongs to this class of spiritual scientist... +1 – robin girard Aug 9 '10 at 12:32
  • All physical experiments I have ever seen have a standard deviation attached to it in some form (most commonly a +/- range). So he must have been joking. – KalEl Aug 19 '10 at 10:25
  • 5
    I think this quote means that the results of the experiment should be "obvious", and statistics just lets one put a precise figure as to "just how obvious". The word needs is the key. – probabilityislogic Jan 30 '11 at 10:56
  • 2
    I thought this was the tagline for Mythbusters... – JMS May 27 '11 at 20:31
  • 1
    I really like this quote, for me it is clear that it is not a joke (@KalEl). This recalls that most often it is not the data that talks but the experiment, standard deviations are here to confirm that your experiment is talking loud. – robin girard Sep 20 '11 at 7:07

The plural of anecdote is not data.

-- Roger Brinner

(in the context of Anecdotal_evidence)

  • 6
    Surely it is, as long as the anecdotes aren't sampled with bias? – naught101 Nov 3 '12 at 6:38
  • 1
    @naught101 Please provide an example? – Jase Nov 21 '12 at 10:15
  • 3
    @Jase: an anecdote is a chunk of information that is true, but may not be representative of the truth (i.e. it's biased toward the point that the story teller is trying to make). But that doesn't say anything about multiple anecdotes. If you could show that the biases in each anecdote in a set were independent, then they would probably cancel to some extent, allowing reliable analysis. Of course, this is a stupidly inefficient way of collecting data, and because it would be so difficult, there are no examples, because no-one has ever done it. And I was mostly just being a smart arse :D – naught101 Nov 21 '12 at 10:23
  • 2
    I would think that anecdotes are fundamentally biased: They caught the attention of someone, and caused emotions that made him remember it. There is no way human attention can be statistically independent, I assume. – Volker Siegel Sep 3 '15 at 16:09

"Million to one chances crop up nine times out of ten."

-Terry Pratchett

Those who ignore Statistics are condemned to reinvent it.

-- Brad Efron

  • Do you think that by "those" he means "everyone" ? what was the context of this citation ? it seems a bit strong like this :) – robin girard Aug 30 '10 at 12:42
  • 5
    I would go one step further, within statistics. Those who ignore Bayesian statistics are condemned to reinvent it. – probabilityislogic Jan 30 '11 at 11:08

…the statistician knows…that in nature there never was a normal distribution, there never was a straight line, yet with normal and linear assumptions, known to be false, he can often derive results which match, to a useful approximation, those found in the real world.

George Box (JASA, 1976, Vol. 71, 791-799)

The death of one man is a tragedy. The death of millions is a statistic.

-- Kurt Tucholsky, in: Französischer Witz, 1925

  • 8
    According to Wikiquote it is misattributed to Joseph Stalin; the origin is Kurt Tucholsky: en.wikiquote.org/wiki/Joseph_Stalin#Misattributed – Peter Mortensen Aug 7 '10 at 0:36
  • Here is the (so far much more likely) source: Französischer Witz by Kurt Tucholsky dating back to 1925: »Der Krieg? Ich kann das nicht so schrecklich finden! Der Tod eines Menschen: das ist eine Katastrophe. Hunderttausend Tote: das ist eine Statistik!« (German original) – gwr Dec 5 '15 at 15:56

"It is easy to lie with statistics. It is hard to tell the truth without statistics." - Andrejs Dunkels

  • The best antidote to "...lies, damned lies, and statistics." – Thylacoleo Aug 18 '10 at 10:44
  • depending on what you call 'the thruth'... I don't think I give so much value to statistical truth :) – robin girard Aug 30 '10 at 12:39
  • I think it's a great one (I think it goes well with - "no models are true, some are useful") – Tal Galili Oct 20 '10 at 2:27

"The first time I was in a statistics course, I was there to teach it"

John Tukey (link)

  • 3
    He's not the only one--and that's not necessarily a good thing. – whuber May 18 '12 at 17:00
  • 8
    Haha, true. Tukey gets a pass though! – Neil McGuigan May 18 '12 at 17:57

I keep saying that the sexy job in the next 10 years will be statisticians. And I'm not kidding.

Hal Varian

  • I guess Val Harian is not a statistician if he is not kidding... what is a sexy job ? for me it is like the sitation with the sword of the century... fun but a bit trivial :) – robin girard Aug 12 '10 at 7:58
  • 2
    This needs to be corrected. It was Hal Varian that said it. nytimes.com/2009/08/06/technology/06stats.html – vqv Dec 20 '10 at 4:23
  • 4
    in what year did he write this? – Leo Schalkwyk Mar 29 '12 at 9:43
  • 3
    @Leo: He didn't write it; he said it (at the link provided by vqv). It was only a couple of years ago, so, you still have time. ;-) – cardinal Mar 29 '12 at 11:57

"To find out what happens when you change something, it is necessary to change it.”

Box, Hunter, and Hunter, Statistics for Experimenters (1978).

  • 7
    Tell that to the theoretical physics community... – naught101 Nov 3 '12 at 6:41

The greatest value of a picture is when it forces us to notice what we never expected to see.

-- John Tukey

There are three kinds of lies: lies, damned lies, and statistics.

-- probably: Charles Wentworth Dilke (1843–1911).

  • 18
    I hate this quote. It makes professions using statistics look like you could cheat. But, when someone profoundly uses statistics one knows that actually you cannot cheat. Because when provided with enough information about the statistical procedures used, one can draw a conclusion on the soundness of the procedures/results. If not enough information on the statistical (and other) procedures are provided, you should immediately question the results. – Henrik Jul 27 '10 at 16:15
  • 5
    That would be true if everyone were knowledgeable enough in statistics to drive the correct conclusions. Alas, that quote is very applicable to many of those amusing human beings called politicians... – nico Jul 29 '10 at 11:22
  • 4
    Whoever said this had no basic understanding of Statistics, or he was joking. – KalEl Aug 19 '10 at 10:21
  • 10
    I think this quote is more of a cynical but realistic view of how statistical data is mostly used in debates (i.e. selected to support a preconceived notion rather than produced to test a hypothesis) – jilles de wit Sep 7 '10 at 9:12
  • 6
    There are three kinds of lies: Lies, damned lies and fake statistics. – Joris Meys Sep 13 '10 at 22:01

We are drowning in information and starving for knowledge.

Rutherford D. Roger

60% of the time, it works every time.

-Brian Fantana

"The Central Limit Theorem is about the journey and the Strong Law of Large Numbers is about the destination." stats.SE user cardinal in a comment on this question

While the individual man is an insoluble puzzle, in the aggregate he becomes a mathematical certainty. You can, for example, never foretell what any one man will be up to, but you can say with precision what an average number will be up to. Individuals vary, but percentages remain constant. So says the statistician.

Arthur Conan Doyle

  • Clearly this person doesn't quite grasp the power of combinatorial calculations. Normal distributions appear because they are easy to realise. – probabilityislogic May 22 '15 at 0:31

The statistician cannot evade the responsibility for understanding the process he applies or recommends.

-– Sir Ronald A. Fisher

It ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so.

Mark Twain (okay, so he's not a statistician)

  • 1
    Google finds 12.3 million hits for this quotation. After running down the first eight pages of them, I haven't found a single site that actually gives a source--they all seem to be quoting each other rather than Mr. Twain. Does anyone know where he wrote this? Or maybe it's apocryphal. – whuber Nov 8 '10 at 14:05
  • 2
    Heh, it was also used in the beginning of "The Big Short": imdb.com/title/tt1596363 ( filmcomment.com/blog/deep-focus-the-big-short ) – Konstantinos Jan 17 '16 at 21:07
  • I am almost sure that this is from "Huckleberry Finn." Ain't it? – Ulisses Braga-Neto Oct 2 '16 at 1:25

This one is brand new, and Allen Wilcox is an epidemiologist, not a statistician, but whatever, I'm running with it.

Data do not speak for themselves - they need context, and they need skeptical evaluation

  • 5
    I am upvoting this because a minor variation is so apt for our site: "Your data/output/code/formula do not speak for themselves: they need context and they need sceptical evaluation." – whuber Aug 4 '15 at 14:48

Correlation doesn’t imply causation, but it does waggle its eyebrows suggestively and gesture furtively while mouthing ‘look over there’.


"Extraordinary claims demand extraordinary evidence."

Often attributed to Carl Sagan, but he was paraphrasing sceptic Marcello Truzzi. Doubtless the concept is even more ancient.

David Hume said, "A wise man, therefore, proportions his belief to the evidence".

One could argue this is not a quote about statistics. However, applied statistics is ultimately in the business of evaluating the quality of evidence for or against some proposition.

My thesis is simply this: probability does not exist. - Bruno de Finetti

If I can't picture it, I can't understand it.

-Albert Einstein

I acknowledge that Einstein wasn't a statistician. However, Michael Friendly uses this quote in arguing for a greater role for visualizations in data analysis. I share that goal, and I think the quote works nicely.

  • 2
    +1 That's a nice one. The funny thing is this, sometimes people can picture it (make or see a plot of something) and still not understand it. :) – Graeme Walsh Jul 4 '13 at 0:44

The primary product of a research inquiry is one or more measures of effect size, not p values.

Cohen, J. (1990). Things I have learned (so far). American Psychologist, 45, 1304-1312.

The Earth is round. p < .05

Jacob Cohen

  • 6
    The earth is an oblate spheroid. So I guess that depends on what degree of accuracy you care about.. – naught101 Feb 7 '12 at 2:24

When I see articles with lots of significance tests, I say that the statisticians are p-ing on the research.

Herman Friedmann (by recollection, he said this in class)

  • Who's Hermann Friedmann? I've tried searching for him, but couldn't find anything... – An old man in the sea. Nov 29 '16 at 22:48

May I add this one, because I like Jan's contributions to psychometrics and statistics...

Causal interpretation of the results of regression analysis of observational data is a risky business. The responsibility rests entirely on the shoulders of the researcher, because the shoulders of the statistical technique cannot carry such strong inferences.

Jan de Leeuw, homepage

An ecologist is a statistician who likes to be outside.

-- apparently a good friend of Murray Cooper.

I just can't help myself, this is a provocative quote from E. T. Jaynes:

Many of us have already explored the road you are following, and we know what you will find at the end of it. It doesn't matter how many new words you drag into the discussion to avoid having to utter the word 'probability' in a sense different from frequency: likelihood, confidence, significance, propensity, support, credibility, acceptability, indifference, consonance, tenability; and so on, until the resources of the good Dr Roget are exhausted. All of these are attempts to represent degrees of plausibility by real numbers, and they are covered automatically by Cox's theorems. It doesn't matter which approach you happen to like philosophically; by the time you have made your methods fully consistent, you will be forced, kicking and screaming, back to the ones given by Laplace. Until you have achieved mathematical equivalence with Laplace's methods, it will be possible, by looking in specific problems with Galileo's magnification, to exhibit the defects in your methods.

protected by Community Nov 19 '15 at 6:14

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

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