Why visualise data?

By Chris McDowall 03/09/2010


Why visualise data? In the introduction to his classic text, The Visual Display of Quantitative Information, Edward Tufte answers this question in three words. “Graphics reveal data”.

To illustrate his point Tufte asks the reader to examine four datasets of eleven (x, y) datapoints, collectively known as Anscombe’s quartet. I’ve reproduced them in the figure below.

seeingdata1

The datasets that constitute Anscombe’s quartet share identical basic statistical properties and can be described by the same linear model. Take a look at them. What can you see? If you’re anything like me, the answer is probably “not very much”.

When the data is graphed their characteristics and differences become immediately apparent.

seeingdata2

Visualisation renders complex data accessible. Graphics and interactive visualisations make visible things we never expected to see. They lend tangibility to abstract concepts, reveal patterns and highlight outliers. Good visualisation tells stories, surprises us and challenges our assumptions.

This blog will examine the state of data visualisation in New Zealand and abroad. I plan to draw attention to the interesting, the timely and the beautiful. I hope you will join me.


0 Responses to “Why visualise data?”