Showing posts with label Correlations. Show all posts
Showing posts with label Correlations. Show all posts

Tuesday, May 13, 2014

Coincidence or Spurious Correlation: Divorce Rate In Maine and the Per Capita Consumption of Butter in the US:

Here for spreading far and wide is another graphical reminder of the important distinction between correlation and causation.

Created by Tyler Vigen, Spurious Correlations is a website devoted to documenting examples of what are best described as spurious relationships – that is, relationships in which two events have no causal link, but still appear to due to either a coincidence or a third, confounding variable. The same way the rise in autism rates may look like it's caused by the rise in organic food sales, Vigen's charts show causal relationships between things like divorce rates in Maine and margarine consumption, the number of people who died by becoming tangled in their bedsheets and total revenue generated by skiing facilities, and US crude oil imports from Norway and the number of drivers killed per annum in railway train collisions. A good website to have on hand when discussing cognitive biases with antivaxxers.

More from IO9.

Wednesday, April 18, 2012

Correlation or Causation? Statistics are easy...

Need to prove something you already believe? Statistics are easy: All you need are two graphs and a leading question

Correlation may not imply causation, but it sure can help us insinuate it.

Graphics here.

Saturday, July 16, 2011

http://www.correlated.org/

"69 percent of people who have business cards have gotten a speeding ticket, compared with 53 percent of people in general.
Based on a survey of 298 people who have business cards and 743 people in general."

This is one the truly bizarre factoids available at Correlated.org. Forget the fact that these are self-selected people, and therefore it's not a scientific survey; it's just weird. But from the webpage, the Freakonomics people seem to like it.