So, I was going to write a really important post tonight.  It was going to deal with philosophy of science, and it was going to be a shoe-in for Open Lab 2011.  It was gonna make you question how you thought about Life, The Universe, And Everything.

And then Google Labs released a new tool.

And I was all, “WOO SHINY NEW TOY!”



That was how I found out about ultimate frisbee and insects.

Google Correlate is an experimental new tool on Google Labs which enables you to find queries with a similar pattern to a target data series. The target can either be a real-world trend that you provide (e.g., a data set of event counts over time) or a query that you enter.”

Basically, Google takes a search term that you enter (“insects”) and examines search patterns for other search terms in its database to calculate a correlation coefficient (r).  It’s an extension of Google Trends; it’s looking to see which search terms trend together.

In case you barely remember that statistics class from your misspent youth, the correlation coefficient is a value between -1 and +1.  The closer to ±1 the r value is, the more closely correlated the patterns are.  The closer the value to zero, the less the two patterns are related.

Of course, we’ve all heard the “correlation is not causation” trope a million times. It’s especially true here; when you don’t even have a hypothesis about a relationship, the data points are just amusing.

So for your amusement and edification:

In addition to frisbee, “insects” is also strongly correlated with the search terms “snake photo” and “lizards”.

“Insect” (non-plural) is most strongly correlated with “aluminum siding,” “dunking booth,” and “frisbee” (non-ultimate).

“Ants” is most strongly correlated with “string trimmer.”
“Bees” is most strongly correlated with “Tool Rental.”  “Honey Bee” is correlated with “raptor cam“.
“Roaches” is most strongly correlated with “warts,” as well as “5 year anniversary.”

Lice” is strongly correlated with “dragon fruit“, but also “literacy stations” and “cheer routine.”  In fact, several cheerleading terms show up in the correlation list for “lice.”  “Head Lice” is strongly correlated with “tackle football“, and “Nits“are correlated with “cheerleading bows“, so perhaps football season mirrors lice season?

And, of course, you know I had to go there.

“Crab Lice” is most strongly correlated with….”Lighthouses“?

“Pubic Lice” is most strongly correlated with….a host of civil war search terms.

I am frankly rather baffled about why these search terms should be seasonally correlated, unless Civil War Re-enactors are taking things a little too seriously in their search for authenticity.

Give it a spin–what fun correlations can you discover?

Posted by Gwen Pearson

Writer. Nerd. Insect Evangelist. Have you heard the good news? BUGS!


  1. Bonus points if you spot the reference to Douglas Adams on Towel Day :)

  2. Ha! The only response, I guess is:

    “Nice Marmot Correlation, Dude.”

  3. Bird correlations are all over the place:

    0.8185 types of birds
    0.8173 sea star
    0.8157 assassination of jfk
    0.8121 clerihew
    0.8105 excretory
    0.8090 vietnam war songs
    0.8087 excretory system
    0.8063 famous people from utah
    0.8061 blue morpho
    0.8047 chickadee

    Same with birds:

    0.8810 birds windows
    0.8494 the dandelion
    0.8447 eliza doolittle
    0.8424 2554
    0.8084 mexican lemonade
    0.8066 usa road trip
    0.7920 lily bloom
    0.7800 spring pad
    0.7782 what is a composite bat
    0.7768 the world is a beautiful place

    I like the last correlation for birds. Well, for most birds, anyway.

  4. I got ‘Adwords’ and ‘detention centre’ – take that Google Adwords! This is ultimately the downside of the tool Google Correlate, it pretty much has no useful applications. The correlations seem to be entirely spurious or totally obvious that you would have figured it out anyway. There is no ga in knowledge that requires us to know that insects and ultimate frisbee are correlated. This is just going to lead to people trying to draw tenuous conclusions about relationships between search terms. For every one correlation that is interesting and lucid, there will be at least 10 that are not!

Comments are closed.