Age of Miss America correlates with Murders by Steam, Hot Vapours, and Hot Objects

Written by: Rachel Minkin

Primary Source: Information Literacy Resources

This week’s blog post comes to us from Abraham Wheeler, health science librarian.

I just came across a fantastic website where the author has written a program that find correlations between completely unrelated sets of data. This is then turned into funny and sometimes frightening charts. The charts demonstrate a couple of really important things to keep in mind about data. The expression “correlation does not imply causation” is a good one remember when examining claims being made by an author. Discovery of correlations can be a good way to generate a hypothesis, but is a weak proof of a hypothesis. The discovery of a correlation is the beginning of a scientific inquiry into a subject, not the endpoint.

It also raises a potential problem with the concept of “Big Data” or large data sets. It is very possible for spurious statistical relationships to appear. The more data points you examine, the more likely you are to have coincidences appear and possibly seem meaningful.

Thank you, Abe, for bringing this site to our attention. How we introduce the use of data and statistical analysis to students at the undergraduate level matters.. check out the website yourself for charts you could use in your classes.

The following two tabs change content below.
Rachel Minkin
Rachel M. Minkin is an Information Literacy Librarian, responsible for leading information literacy sessions for the first-year students taking Tier One Writing class in the Department of Writing, Rhetoric and American Cultures. Prior to beginning at Michigan State in August 2011, Rachel was a Reference and Instruction Librarian at Lansing Community College (Lansing, MI) and the Graduate Theological Union (Berkeley, CA).