There are lies, damned lies and statistics

Why do we say "there are lies, damn lies, and statistics"?

The listing is ranked based on unreliability. Simple lies are those lies without any proof. They easily generate doubts, because they do not come with a reliable source to support. Damn lies can be more misleading. However, lies using statistics are statements open to interpretation, which can obscure the truth silently.

How can statistics lie?

Numbers themselves do not lie, rather how people interpret them can make a big difference. "It is possible for anyone to manipulate a data set so the extracted statistics suit their argument." The language used to state their conclusion can be confusing. The graphs may also have misleading effects when using unnecessary angles, areas, heights to compare between data. This is exactly why professional statisticians prefer to use the simplest graphs to visualize their results. (However, for common people, this gives an impression that statistics is dull.)Moreover, sometimes when people use statistic tools to analyze data, they do not lie intentionally, but their poor-made graphs and inexact descriptions express their conclusions otherwise.

How can we avoid falling into statistical tricks?

* Check the source of data and try to find cross justifications.

* Read the captions, interpretations and conclusions carefully.

* Keep a critical mind.

In the end, allow me to show you my favorite statistical jokes!

Reference

Sean Ding, Margaret Weiss from Quora