What is the famous data set that looks totally different but has similar summary stats?

Question

There is a famous example of a collection of datasets with similar summary statistics like mean, standard deviation etc., whose visual appearances are totally different. It is named after the famous statistician who produced it.

Can someone point me to this? I tried googling this but could not get relevant results due to generic search terms.

Perhaps helpful - Are distributions with the same moments identical? http://stats.stackexchange.com/questions/84219/whether-distributions-with-the-same-moments-are-identical — D L Dahly, Mar 13 '15 at 14:30
And here: http://stats.stackexchange.com/questions/25010/identity-of-moment-generating-functions — D L Dahly, Mar 13 '15 at 14:32
Anscombe's quartet? [wiki](http://en.wikipedia.org/wiki/Anscombe%27s_quartet) — Penguin_Knight, Mar 13 '15 at 14:34

score 12 · Accepted Answer · answered Mar 13 '15 at 14:34

12

You must be thinking of Anscombe's quartet.

answered Mar 13 '15 at 14:34

NickJ

234
2
4

YES!! Exactly right. – raco Mar 13 '15 at 14:56

score 6 · Answer 2 · answered Nov 08 '15 at 22:29

Anscombe quartet is the name (as said before), and its standard plots are below.

It was constructed in from Graphs in Statistical Analysis, The American Statistician, 1973. Since then, there have been attemps to reproduce or generalize it on a broader extend, for instance:

Generating Data with Identical Statistics but Dissimilar Graphics: A Follow up to the Anscombe Dataset, The American Statistician, 2007,
.Cloning data: generating datasets with exactly the same multiple linear regression fit, Australian and New Zealand Journal of Statistics, 2009.

What is the famous data set that looks totally different but has similar summary stats?

2 Answers2