Hi ,

​

Visualising your data before starting your analysis is a lot like going to the dentist.

We all know we should do it, and in the long run, will be better off if we do do it. Yet, in the short term, generally things aren't going to fall apart if we don't do it, and there is always something else on our to do list that we would rather do.

Yet, just like those photos of mouths full of rotting teeth that are designed to scare you into making your next dental appointment, here is a set of (albeit less gross) plots that will hopefully "scare" you back into the habit of visualising your data before performing any modelling.

All the following data sets have the same basic summary statistics (to 2 decimal places):

Mean(x) = 54.26, Mean(y) = 47.83, SD(x) = 16.76, SD(y) = 26.93, Pearson’s r = -0.06

And so does this one...

👇

👇

👇

👇

Yes, this is real (the plots were all taken from HERE).

You might think you know what your data looks like from the summary statistics. But until you visualise it, you will never truly know if your data is just some boring lines of dots... or a datasaurus.

Talk again soon,

Dr Genevieve Hayes.

All past emails are available in the archive.