correlation, causation, and so on

Has anybody played around much with Google’s Correlate tool? Quite amazing, in a frightening sort of way. I found it surfing from this similarly amusing, but less thorough, post. I can come up with no adequate theory to explain the nearly .73 correlation between my scribbled line and searches for “home videos clips” on Google.

However, using a steadily decreasing line, the correlation is mostly with searches for now-obsolete web technologies (how many people now search for “web page” on google?), which indicates, I think, the fact that the meaning of google itself has changed over time.

Author: andrewperrin

University of North Carolina, Chapel Hill

One thought on “correlation, causation, and so on”

  1. I do use a couple of their examples (“mittens”) to make exactly the point your title suggests: to teach students about correlation and causation.

    I drew a distribution I know without thinking about the meaning of the axes and that yielded nonsense. There are not a lot of problems I care about that have “a count of web searches” and “over the last decade” as relevant characteristics. I suppose that’s what you’re getting at by pointing to the importance of an interest in “the meaning of google.”


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: