Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

What is your favorite statistical quote?

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

share|improve this question
6  
Should this question really be "famous quotes about statistics"? – naught101 Nov 3 '12 at 4:29

143 Answers 143

All models are wrong, but some are useful. (George E. P. Box)

Reference: Box & Draper (1987), Empirical model-building and response surfaces, Wiley, p. 424.

Also: G.E.P. Box (1979), "Robustness in the Strategy of Scientific Model Building" in Robustness in Statistics (Launer & Wilkinson eds.), p. 202.

share|improve this answer
2  
I use this quote a lot to explain the difficulties in mathematicians transitioning to statistics – user549 Jul 29 '10 at 18:48
18  
This sentence itself is a model (an epistemological one) – user603 Sep 10 '10 at 20:00
3  
but see a nice discussion around this quote on Gelman's blog, j.mp/9SgIBO – chl Sep 11 '10 at 10:21
12  
And this is an actual quote, as opposed to something "attributed to" Box. It appears, e.g., in Box & Draper (1987), Empirical model-building and response surfaces, Wiley, on page 424. Yes, I did go and look it up before using it in a paper. – Stephan Kolassa Oct 14 '10 at 15:53
13  
Sadly, too many people use it to excuse themselves from the flaws in their models. In my personal experience, it's usage is an alarm sign. – JohnRos Feb 2 '12 at 13:35

"An approximate answer to the right problem is worth a good deal more than an exact answer to an approximate problem." -- John Tukey

share|improve this answer
15  
I like this one, could be put as an advise when people write questions on this site ? – robin girard Jul 27 '10 at 8:48
7  
Absolutely...asking the right question is one of the most important skills. – Shane Jul 27 '10 at 14:17
5  
I remember once where a private industry company commissioned a mathematician to solve a garbage collection routing problem. Long story short, the mathematician complained that the company was only interested in finding a "close enough" solution rather than an optimal solution. I think, ultimately he was fired, and an operations researcher was brought in instead. – dassouki Jul 27 '10 at 17:59
2  
@dassouki I think the quote is more about the question .... something like science is not about finding good answer but about finding good questions ! – robin girard Jul 27 '10 at 20:21
8  
"Far better an approximate answer to the right question, which is often vague, than an exact answer to the wrong question, which can always be made precise." John W. Tukey 1962 The future of data analysis. Annals of Mathematical Statistics 33: 1-67 (see pp.13-14) No doubt he said similar things at other times, but that's a precise source, and the version I usually see quoted. – Nick Cox Apr 27 '13 at 23:02

"To call in the statistician after the experiment is done may be no more than asking him to perform a post-mortem examination: he may be able to say what the experiment died of."

-- Ronald Fisher (1938)

The quotation can be read on page 17 of the article.

R. A. Fisher. Presidential Address by Professor R. A. Fisher, Sc.D., F.R.S. Sankhyā: The Indian Journal of Statistics (1933-1960), Vol. 4, No. 1 (1938), pp. 14-17. http://www.jstor.org/stable/40383882

share|improve this answer
    
I read a slightly different version of this quote by Fisher: "Hiring a physician after the data have been collected is like hiring a physician when the patient is in the morgue. He may be able to tell you what went wrong, but he is unlikely to be able to fix it." – Peter Flom May 27 '11 at 18:10
2  
@Peter Was it really "Hiring a physician after the data ..." or should "statistician" be in there somewhere? – Dason Nov 4 '11 at 14:06
2  
@dason You're right! Someone edited my post, I think – Peter Flom Nov 4 '11 at 21:05

87% of statistics are made up on the spot

-Unknown

Dilbert.com Dilbert.com

share|improve this answer
3  
imgur.com/0dsVC.gif – J. M. Oct 23 '10 at 11:59
1  
And 45.8% of people don't believe that statistic – probabilityislogic Feb 5 '11 at 13:12
    
ROFL ROFL Scott Adams kills me – Hack-R Apr 30 '15 at 19:44
    
Ha! Every time I see a forecast that contains too many significant digits I think of this quote. "The number of cell phone owners is forecast to be 4,372,138,975 by the year 2020." Really? As if anyone could forecast better than 4.3B or 4.4B. – JoeTaxpayer Sep 4 '15 at 11:31

Statistics are like bikinis. What they reveal is suggestive, but what they conceal is vital.

-Aaron Levenstein

share|improve this answer
6  
And life's more fun without them? Guess you can only take a metaphor so far... – naught101 Aug 21 '13 at 2:27

Statisticians, like artists, have the bad habit of falling in love with their models.

-- George Box

share|improve this answer

In God we trust. All others must bring data.

(W. Edwards Deming)

share|improve this answer
48  
God must bring data too. – KalEl Aug 19 '10 at 9:33
8  
God can make up data. – Leo May 28 '11 at 4:11
5  
@Leo What data do you have to support that hypothesis? :) – probabilityislogic May 28 '11 at 13:23
1  
Ooh, is that a new version of the Omnipotence Paradox? If god made up new data, how could you prove that it wasn't there all along? – naught101 Nov 3 '12 at 4:33
4  
It's axiomatically true. – abaumann Apr 16 '13 at 7:16

Prediction is very difficult, especially about the future.

-- Niels Bohr

share|improve this answer
15  
Prediction about the past can also be surprisingly tricky! – walkytalky Aug 19 '10 at 7:58
3  
This one has been attributed to many different people larry.denenberg.com/predictions.html and it's disputed that it would be Niels Bohr en.wikiquote.org/wiki/Niels_Bohr – gerrit Jul 25 '12 at 16:16

All generalizations are false, including this one.

Mark Twain

share|improve this answer
    
This is brilliant! – probabilityislogic Jan 19 '11 at 3:51
    
It is, except are there any generalisations that are entirely true? – naught101 Mar 21 '12 at 7:15
    
@naught101 Definitions and the laws of nature (once we know them) are generalizations that I consider true. Though the former are not very interesting as in: all "true generalizations" are true. – ziggystar Feb 3 '13 at 15:20

The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data

Tukey

share|improve this answer
5  
As a biological scientist, I find myself muttering this to myself during a lot of seminars... – N Brouwer Oct 18 '12 at 18:05
    
This should be taught in management school long before the chapter on KPI calculation – rumtscho Feb 6 '14 at 20:04

If you torture the data enough, nature will always confess.

--Ronald Coase (quoted from Coase, R. H. 1982. How should economists chose? American Enterprise Institute, Washington, D. C.). I think most who hear this quote misunderstand its profound message against data dredging.

share|improve this answer

A big computer, a complex algorithm and a long time does not equal science.

-- Robert Gentleman

share|improve this answer
43  
Still it looks promising. – mbq Jul 27 '10 at 11:20
2  
Just curious, where did he say/write that? – Hack-R Apr 30 '15 at 19:49

Statistical thinking will one day be as necessary a qualification for efficient citizenship as the ability to read and write.

--H.G. Wells

share|improve this answer
    
By God, he was right! – KalEl Aug 19 '10 at 9:32
10  
I don't know, you've seen many efficient citizens lately? – Raskolnikov Dec 3 '10 at 23:26
2  
Still waiting... – naught101 Nov 3 '12 at 4:34
1  
An anonymous user asked for a source for this 'quote'; he/she also indicated that Gigerenzer noted that he searched Wells published output in vain for the original. – chl Apr 25 '13 at 7:00

There are no routine statistical questions, only questionable statistical routines.

D.R. Cox

share|improve this answer
3  
Rolf Sundberg attributed this quote to J.M. Hammersley in a 1994 article: dx.doi.org/10.1016/0169-7439(93)E0041-2 – onestop Jan 28 '11 at 20:41
1  
The following was an attempted edit by an anonymous user: "Comment: It is told after the qoute that I have attributed this (excellent) quote to Hammersley. The reason for my attribution of it to Hammersley was that I asked David Cox before I used the quote, and he answered that it was not originally his, but Hammersley's phrasing. Rolf Sundberg". – gung May 23 '13 at 19:51

Strange events permit themselves the luxury of occurring.

-- Charlie Chan

share|improve this answer
12  
I don't mind the down vote, but I maintain that this is a deep statistical point, not to be taken lightly. ;-) – ars Jul 27 '10 at 7:18
    
Especially if you are in the financial services sector. – DWin Jan 18 '11 at 22:21
2  
I would say that the key to cracking the meaning of this quote is to recognise that the word "strange" is relative to what your model of "normal" is. – probabilityislogic Feb 5 '11 at 13:11
1  
I this just a verbose way of saying "outliers happen", or is there something deeper I'm missing? – naught101 Feb 7 '14 at 2:14
    
A similar quote that I like is “With a large enough sample, any outrageous thing is likely to happen" (Persi Diaconis and Frederick Mosteller). – MattBagg Feb 13 '15 at 1:34

A nice one I came about:

I think it's much more interesting to live not knowing than to have answers which might be wrong.

By Richard Feynman (link)

share|improve this answer
    
If I was a betting man I'd say Richard Feynman was an agnostic – probabilityislogic Jan 30 '11 at 10:48
    
Does Feynman qualify as a statistician? – Glen_b Nov 3 '12 at 6:59

The best thing about being a statistician is that you get to play in everyone's backyard.

-- John Tukey

(This is MY favourite Tukey quote)

share|improve this answer
4  
Love this one -- a wonderful bonus of being a statistician. – Kingsford Jones Sep 11 '10 at 17:13

Statistics - A subject which most statisticians find difficult but in which nearly all physicians are expert.

share|improve this answer
2  
Credit: Stephen Senn, Statistical Issues in Drug Development, page4. media.wiley.com/product_data/excerpt/71/04700187/0470018771.pdf – onestop Nov 8 '10 at 9:03
1  
I think its because physics has a similar level of pedantry required for statistics, and the physicist has the huge benefit of wanting to get rid of uncertainty, the statistician just wants to describe it. – probabilityislogic Jan 30 '11 at 11:04
22  
Physicians \neq physicists – David Roberts Aug 18 '11 at 1:14
1  
This may be my new favorite – Fomite Jul 22 '12 at 21:51
1  
Although I could imagine this applying to physicists as well. – MHH Feb 7 '14 at 1:16

He uses statistics like a drunken man uses a lamp post, more for support than illumination.

-- Andrew Lang

share|improve this answer

Say you were standing with one foot in the oven and one foot in an ice bucket. According to the percentage people, you should be perfectly comfortable.

-Bobby Bragan, 1963

share|improve this answer
5  
Wrong: 200 C and 0 C average to about 100 C, which is the boiling point of water. Ovens only go down to about 150 C, and 75 C is still too hot. Now, if you have one foot in scalding water (about 55 C) and another in cold icy water... then you are probably a strange person. – alexfernandez Aug 4 '14 at 15:23
    
Double wrong. Ice can reasonably be at lower temperatures. – Firebug Apr 27 at 23:02
    
@alexfernandez My oven begins at 50°C. It is a standard oven, and all ovens in the flats that I lived in began at this temperature. – what Jun 5 at 9:40
    
@what I suppose that low-temperature cooking has brought down minimum temperatures, but I doubt that in 1963 this was the case. – alexfernandez Jun 5 at 21:35

Absence of evidence is not evidence of absence.

Carl Sagan

share|improve this answer
57  
Good quote, but it's not true! Absence of evidence is not proof of absence, but it certainly is evidence. Why do we think magnetic monopoles (or unicorns, for that matter) don't exist? Because we've looked and haven't found any. – John D. Cook Aug 17 '10 at 18:15
10  
Besides, Tzippy is misquoting Sagan, since Sagan never believed that. He in fact listed it among the fallacies in his baloney detecion kit. – Raskolnikov Dec 3 '10 at 23:31
7  
@JohnD.Cook, +1. However, your comment relies on the fact that we have looked, and that there was a reasonable chance of having found evidence if it really were there; consider, for example, the various 'missing links' that were ultimately found (and those that have not yet been). – gung Feb 6 '12 at 18:32
    
Does Sagan qualify as a statistician? – Glen_b Nov 3 '12 at 7:00
1  
@Glen_b It appears that Sagan might have said that in some sense of irony, if it all, being a critic of Martin's quote. That's something to contemplate, for me, since cosmology is so full of examples where predictions have been made to account for inexplicable sources of error that have turned out to be correct (or not quite debunked), e.g. cosmic background radiation, dark matter, and the Big Bang Theory. – AdamO Jan 13 '14 at 17:21

Everybody believes in the exponential law of errors [i.e., the Normal distribution]: the experimenters, because they think it can be proved by mathematics; and the mathematicians, because they believe it has been established by observation.

Whittaker, E. T. and Robinson, G. "Normal Frequency Distribution." Ch. 8 in The Calculus of Observations: A Treatise on Numerical Mathematics, 4th ed. New York: Dover, pp. 164-208, 1967. p. 179.

Quoted at Mathworld.com.

share|improve this answer
6  
This is a rather free translation of a saying attributed to Gabriel Lippmann by Henri Poincar\'e in his Calcul des probabilit\'es (1896/1912). Original was in French, naturellement. Lippmann won the Nobel Prize for Physics in 1908. – Nick Cox Apr 29 '13 at 22:30

All we know about the world teaches us that the effects of A and B are always different---in some decimal place---for any A and B. Thus asking "are the effects different?" is foolish.

Tukey (again but this one is my favorite)

share|improve this answer
    
It has actually led to very interesting articles... :) – Tal Galili Jul 31 '10 at 1:07
    
@Tal: Fully agree! I think the whole area on optimal separation in minimax testing is starting from this idea ... and it is still so confused for a lot of statistician. For those interested see the paper of donoho projecteuclid.org/… (and the references in the paper ! since things are much older than donoho's paper) – robin girard Jul 31 '10 at 6:19

... surely, God loves the .06 nearly as much as the .05. Can there be any doubt that God views the strength of evidence for or against the null as a fairly continuous function of the magnitude of p? (p.1277)

Rosnow, R. L., & Rosenthal, R. (1989). Statistical procedures and the justification of knowledge in psychological science. American Psychologist, 44(10), 1276-1284. pdf

share|improve this answer
    
I would love to take this one as my accepted answer ! too good to be true ! – robin girard Jul 27 '10 at 17:10
    
Anyway, I can give a citation of Lehman about that: "There is some convenience in such standardization since it permits a reduction in certain tables needed for carrying out various tests". – robin girard Aug 30 '10 at 12:37
    
Rosnow & Rosenthal's is a very useful, eye-opening quote that is almost correct. – rolando2 Mar 17 '11 at 12:15
    
@rolando2: please expand. What's not correct about it? – naught101 Aug 21 '13 at 2:34
    
@naught101 - At this point I can't think of anything :-) – rolando2 Aug 21 '13 at 16:32

"It's easy to lie with statistics; it is easier to lie without them."

-- Frederick Mosteller

share|improve this answer

I don't know about famous, but the following is one of my favourites:

Conducting data analysis is like drinking a fine wine. It is important to swirl and sniff the wine, to unpack the complex bouquet and to appreciate the experience. Gulping the wine doesn’t work.

-Daniel B. Wright (2003), see PDF of Article.

Reference: Wright, D. B. (2003). Making friends with your data: Improving how statistics are conducted and reported1. British Journal of Educational Psychology, 73(1), 123-136.

share|improve this answer
    
Like it, though it does remind me of en.wikipedia.org/wiki/Swiss_Toni – onestop Sep 7 '10 at 7:18

Do not trust any statistics you did not fake yourself.

-- Winston Churchill

share|improve this answer
5  
This quote seems to be known only in Germany and there is doubt that it is authentic, see the link below where the State Office of Statistics in Baden-Württemberg show results of their research about this quote (sorry its only available in German). The Times, e.g., said that they never heard about it. statistik.baden-wuerttemberg.de/Veroeffentl/Monatshefte/… – psj Nov 7 '10 at 10:58
4  
The alternative form is "I only believe in statistics that I doctored myself" sometimes claimed to have been put into Churchill's mouth by Goebbels during a propaganda dispute over wartime losses. – Henry Nov 23 '11 at 22:40
    
I admit that I did not investigate about the quote's origin. However, the core of the statement remains true. Statistics, especially in mass media, are never presented with the necessary information to estimate their validity or correctness. – ymihere Jan 17 '12 at 9:45
2  
Does Churchill qualify as a statistician? – Glen_b Nov 3 '12 at 7:01
1  
@Glen_b If he ever faked some data, then sure! – Darren Cook Feb 1 '13 at 13:27

My greatest concern was what to call it. I thought of calling it 'information,' but the word was overly used, so I decided to call it 'uncertainty.' When I discussed it with John von Neumann, he had a better idea. Von Neumann told me, 'You should call it entropy, for two reasons. In the first place your uncertainty function has been used in statistical mechanics under that name, so it already has a name. In the second place, and more important, no one really knows what entropy really is, so in a debate you will always have the advantage.'

Claude Elwood Shannon

share|improve this answer

The subjectivist (i.e. Bayesian) states his judgements, whereas the objectivist sweeps them under the carpet by calling assumptions knowledge, and he basks in the glorious objectivity of science.

I.J. Good

share|improve this answer
    
oh the bayesian is soooo Good... – robin girard Jul 27 '10 at 15:39
    
I love this one. It is great ! – steffen Dec 6 '10 at 9:00
    
Why are Bayesian always equated with subjectivist? -- What about E.T.Jaynes and the other 'objective Bayesians'? What about all subjectivity within the 'objectivist' frequentism? – gwr Dec 5 '15 at 16:04

Figures don't lie, but liars do figure

--Mark Twain

share|improve this answer
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

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.