# "Famous" set of correlation plots used for teaching

I am familiar with Anscombe's quartet

I am now looking for a different set of plots. From what I remember, these all had correlations of zero, and there was a question here on CV, and I think it was about dependence in the absence of correlation and it was used as part of an answer.

I have tried a few searches such as this one but so far I've not found it.

• Was it this, maybe? Commented Jan 17 at 13:23
• stats.stackexchange.com/questions/633316/… Commented Jan 17 at 13:43
• Wikipedia has one nice plot that I am sure is copied somewhere on this site! Commented Jan 17 at 13:55
• @COOLSerdash It isn't that one, though that one may be sufficient for my needs. Thank you ! Commented Jan 18 at 20:36

Answering my own question after waiting to see if @kjetil b halvorsen convert's their comment. If they subsequently do then I will delete mine.

This is the one I was looking for:

Several sets of (x, y) points, with the Pearson correlation coefficient of x and y for each set. The correlation reflects the noisiness and direction of a linear relationship (top row), but not the slope of that relationship (middle), nor many aspects of nonlinear relationships (bottom). N.B.: the figure in the center has a slope of 0 but in that case, the correlation coefficient is undefined because the variance of Y is zero.

For completeness I will also include some others which also demonstrate the same ideas. First, the "Datasaurus" that @COOLSerdash suggested:

This package wraps the awesome Datasaurus Dozen dataset, which contains 13 sets of x-y data. Each sub-dataset has five statistics that are (almost) the same in each case. (These are the mean of x, mean of y, standard deviation of x, standard deviation of y, and Pearson correlation between x and y). However, scatter plots reveal that each sub-dataset looks very different. The dataset is intended to be used to teach students that it is important to plot their own datasets, rather than relying only on statistics.

The Datasaurus was created by Alberto Cairo. Datasaurus shows us why visualisation is important, not just summary statistics.

And finally, Anscombe's quartet:

All four sets are identical when examined using simple summary statistics, but vary considerably when graphed

• The R code I posted at stats.stackexchange.com/a/152034/919 will create any of these examples as well as any other example you care to conceive of.
– whuber
Commented Jan 24 at 17:35
• @whuber that is awesome ! Commented Feb 6 at 20:56