Search correlations with StateStats

posted by Jason Kottke Dec 03, 2008

StateStats is hours of fun. It tracks the popularity of Google searches per state and then correlates the results to a variety of metrics. For instance:

Mittens - big in Vermont, Maine, and Minnesota, moderate positive correlation with life expectancy, and moderate negative correlation with violent crime. (Difficult to commit crimes while wearing mittens?)

Nascar - popular in North and South Carolinas, strong positive correlation with obesity, and and moderate negative correlation with same sex couples and income.

Sushi - big in NY and CA, moderate positive correlation with votes for Obama, and moderate negative correlation with votes for Bush.

Gun - moderate positive correlation with suicide and moderate negative correlation with votes for Obama. (Obama is gonna take away your guns but, hey, you'll live.)

Calender (misspelled) - moderate positive correlation with illiteracy and rainfall and moderate negative correlation with suicide.

Diet - moderate positive correlation with obesity and infant mortality and moderate negative correlation with high school graduation rates.

Kottke - popular in WI and MN, moderate positive correlation with votes for Obama, and moderate negative correlation with votes for Bush.

Cuisine - This was my best attempt at a word with strong correlations but wasn't overly clustered in an obvious way (e.g. blue/red states, urban/rural, etc.). Strong positive correlation with same sex couples and votes for Obama and strong negative correlation with energy consumption and votes for Bush.

I could do this all day. A note on the site about correlation vs. causality:

Be careful drawing conclusions from this data. For example, the fact that walmart shows a moderate correlation with "Obesity" does not imply that people who search for "walmart" are obese! It only means that states with a high obesity rate tend to have a high rate of users searching for walmart, and vice versa. You should not infer causality from this tool: In the walmart example, the high correlation is driven partly by the fact that both obesity and Walmart stores are prevalent in the southeastern U.S., and these two facts may have independent explanations.

Can you find any searches that show some interesting results? Strong correlations are not that easy to find (although foie gras is a good one). (thx, ben)

Reader comments

katyaDec 03, 2008 at 9:19AM

the results for "brunch" are sort of similar to those of foie gras: http://statestats.appspot.com/?q=brunch

Benjamin DohertyDec 03, 2008 at 9:21AM


TimDec 03, 2008 at 9:25AM

"Subway" has a strong positive correlation with population density and a strong negative correlation with suicide.

LarryDec 03, 2008 at 9:27AM

"apple" (thinking mac users):

VotedForObama 0.75 (Positive, strong)
Income 0.63 (Positive, moderate)
SameSexCouples 0.53 (Positive, moderate)

Suicide -0.57 (Negative, moderate)
EnergyConsumption -0.72 (Negative, strong)
VotedForBush -0.75 (Negative, strong)

AndrewDec 03, 2008 at 9:34AM

swamp music has an obvious map.

And then there's vegan

VotedForObama 0.71 (Positive, strong)
SameSexCouples 0.69 (Positive, moderate)

VotedForBush -0.72 (Negative, strong)
EnergyConsumption -0.76 (Negative, strong)

SmartAssDec 03, 2008 at 9:34AM

Great tool. Very amusing.

Being from North Carolina (and half of a same sex couple) I thought the NASCAR results were quite telling.

Currently in FL, so no surprise that HURRICANE leads to both FL and LA.

I thought it was fun to compare YOGA and GAY MARRIAGE.

They both show up in almost the same states, while YOGA has the strongest same-sex couple correlation and then GAY MARRIAGE almost a non-existent same-sex correlation. The strongest correlation for GAY MARRIAGE was life expectancy.

The gays are looking for yoga instructors and the old people are worried about gay marriage?

r0btDec 03, 2008 at 9:34AM

I tried terms that have nothing to do with Americans. Well some of them have. Try these:

Ali G: has a strong correlation with Obama (75%) and who are for same sex marriage (68%). It has a very negative correlation with Bush (-75%).

Zidane: strong positive correlation with Obama (72%) and strong negative with Bush (-73%)

God: strong positive correlation with obesity (84%).

BBC: strong positive correlation with Obama (73%).

Restaurant has a strong positive correlation with Obama, same sex couples, urban density, and strong negative correlation with Bush.

GregDec 03, 2008 at 9:35AM

Stimulus: strong positive correlation with obesity, moderate negative correlation with several indicators of liberalism.

Jeopardy: strong positive correlation with longitude.

Old GaijiinDec 03, 2008 at 9:44AM

Vasectomy looked kind of strange.

JackieDec 03, 2008 at 9:55AM

"Gardening" gets a .48 positive with "suicide."

ChrisDec 03, 2008 at 9:56AM

Density, .74 (postive, strong).

DRDec 03, 2008 at 9:58AM

Cool tool. After cruising a few terms, I found it kind of interesting to just search states (e.g. California). Though things like Area, and Latitude become meaningless, it's kind of interesting to just compare states' "baseline" data.

SaiDec 03, 2008 at 9:58AM

food stamps


weather : (+) Latitude

: (+) Unemployment

Also see google vs. Yahoo and MySpace vs. Facebook

I had earlier done similar economy-related searches, but in a much more crude fashion. And I was looking for more temporal trends (link). Most of the terms that show up on Google Trends don't seem to have enough data here though.

Bevan HoustonDec 03, 2008 at 9:59AM

I was amused at how few people from outside of my state queried "Oklahoma."
Oklahoma 100.0
Kansas 4.0
Arkansas 3.0
Texas 3.0
It doesn't surprise me that we're most associated with "VotedforBush" being that OK was the only state that McCain carried every single county...

ctoyosDec 03, 2008 at 10:04AM

heh, 4chan: Unemployment, Illiteracy, ViolentCrime, Obesity

ACDec 03, 2008 at 10:10AM

"two girls one cup"

Obesity 0.32 (Positive, weak)
Unemployment 0.28 (Positive, weak)
VotedForObama -0.23 (Negative, weak)
HighSchoolGrad -0.31 (Negative, weak)

- 100.00 in Kentucky, 45.00 in Utah

John PDec 03, 2008 at 10:19AM

Its rather discouraging as a conservative that those who would vote in that direction consistently reveal a sense of tackiness and ignorance- because conservative intellectual thought has a strong and cohesive tradition. So while Nascar is trumpeted, conservatives are rarely identified (intellectually) as

libertarians (see reason. com),
economic conservatives (see WSJ editorial page),
national strength conservatives (see Weekly Standard) or even
social/religious intellectual thought (see First Things journal).

But what is ... is what is.

tobyDec 03, 2008 at 10:24AM

saab is quite interesting
wind power (Latitiude and Frost vs Violent Crime and Illiteracy)
heroin = voted for Obama

sloDec 03, 2008 at 10:30AM

This is phenomenal!!

(of course, most popular in TN)
Obesity 0.63 (Positive, moderate)
InfantMortality 0.46 (Positive, moderate)
Unemployment 0.41 (Positive, moderate)

Latitude -0.5 (Negative, moderate)
HighSchoolGrad -0.54 (Negative, moderate)

ElizabethDec 03, 2008 at 10:37AM

Also from NC, I was proud that NC trumps Texas by such a strong margin forbarbecue.

LondonMDec 03, 2008 at 11:20AM

What about ambiguity? The pattern for Paris is interesting: Paris, Texas vs Paris, France vs Paris, Hilton?

(assuming Paris, of Troy and Paris, Plaster of (etc etc) are negligible)

SmartAssDec 03, 2008 at 11:24AM

Things that make you go hmmmmmm. Palin highest in Alaska, positive for Same Sex Couples and negative for VotedforBush?

I could spend all day here ...

JasonDec 03, 2008 at 11:25AM

Candle has an interesting split along the continental divide.

MargaretDec 03, 2008 at 11:27AM

Just as I thought, only New York Publishing Types care about Roberto Bolano.

MargaretDec 03, 2008 at 11:29AM

@dane: yes, but that may just show Utah's prudery: everyone else doesn't search for pornography, they just search for porn.

gDec 03, 2008 at 11:42AM

Sarah Palin
Hustler, Playboy
Divorce, check this one! LOL
Puppies, interesting how different from searching Kittens
Inflation, interesting one

CDec 03, 2008 at 11:44AM

@dane -- search for porn gets red everywhere, and has its strongest correlation with unemployment...

gDec 03, 2008 at 11:48AM

New York

gDec 03, 2008 at 11:52AM

Interesting ones:


gDec 03, 2008 at 11:56AM

nailin palin
nailin paylin

Joshua WorksDec 03, 2008 at 12:00PM

The geography of these religious terms were expected, but like the correlations, too:

Baptist - very Southern, and strongly correlated to obesity and infant mortality. Negative correlation for lots of things, like voting for Obama, income, HS grad, and life expectancy.

Catholic - not very strongly correlated, but moderate negative with suicide, as expected.

Methodist - Again, strongly correlated with obesity.

Lutheran - Lake Wobegon territory, here.

Muslim - Positively correlated to violent crime, but negatively to suicide.

Hindu - Strong negative correlation with suicide.

Buddhist - Strong negative correlation with obesity. Nice!

Athiest - Oddly correlated with obesity and voting for Bush, but I suspect it's used more often as a derogatory term.

Voodoo - Almost singularly correlated with Lousiana and violent crime.

matt sDec 03, 2008 at 12:13PM

N-word Not entirely "the south", and has HighSchoolGrad -0.77 (Negative, strong)

Other curse words and slurs are interesting in general, they tend to be common in few localities.

Bible The map does a pretty good job of outlining the bible belt.

Jon MayDec 03, 2008 at 12:37PM

arugula is similar to and even more polarizing than fois gras and yet the two are really not alike at all. I can understand the love/hate with fois gras, but arugula is really just bitter lettuce! Come on, flyovers, try some!

new york times is much more polarizing than fox news. Is the opposite of the times really limbaugh? Perish the thought!

CharlieDec 03, 2008 at 1:06PM

apparently it's mainly literate people who search for duct tape. then again, maybe only the literate people know that it's spelled "duct" and not "duck."

richard laniganDec 03, 2008 at 1:14PM

'cum'. the universality of spoo.

JeffDec 03, 2008 at 1:18PM

"Rap" is strong/positive with obesity and strong/negative with hoghschoolgrad.

LauraDec 03, 2008 at 1:24PM

gay, dating, snow, cunnilingus, Twilight, Flickr, blog, 4chan, marriage, news, new york

ChrisDec 03, 2008 at 1:30PM

The "Kottke" that is more popular in MN and WI is probably guitarist Leo Kottke, who went to college in St. Cloud, MN, settled in the Twin Cities, and has an honorary degree from UW-Milwaukee.

SamanthaDec 03, 2008 at 1:55PM

"love" has a strong positive correlation with density.

jkottkeDec 03, 2008 at 2:11PM

Chris, it's probably more that Kottke is a popular family name in WI/MN, not because of Leo (or me).

MaaikeDec 03, 2008 at 2:18PM

Asia shows a strong correlation with Voted for Obama and Income. Similar results with other continents, except for Antarctica which has a strong correlation with Latitude and Highschool Grad :-)

JohnnyLADec 03, 2008 at 2:51PM

Gas - VotedForBush 0.55 (Positive, moderate)
Obesity 0.48 (Positive, moderate)
EnergyConsumption 0.4 (Positive, weak)

video games - Unemployment 0.29 (Positive, weak)
Obesity 0.23 (Positive, weak)
LifeExpectancy 0.22 (Positive, weak)

(and I work in the industry, yikes!)

Jazz -SameSexCouples 0.53 (Positive, moderate)
Income 0.43 (Positive, moderate)
VotedForObama 0.4 (Positive, moderate)
Density 0.37 (Positive, weak)

Noam LovinskyDec 03, 2008 at 3:15PM

New York is almost always #1 for any sexually transmitted diseases.

MaureenDec 03, 2008 at 3:26PM

"Rifle" has a very strong correlation with both Voted for Bush and Suicide, and a very strong negative correlation with Voted for Obama and Density.

lauraDec 03, 2008 at 4:11PM

"Interracial" has a strong positive correlation with infant mortality and a strong negative correlation with high school grad and is most often searched among the Southeastern and Midwest states. I just wonder in what context they are searching under.....


ericDec 03, 2008 at 5:20PM

Booty. It's a southern thing. It doesn't read well and seeks damp, violent places.

DaveDec 03, 2008 at 5:35PM

Try "Proposition 8"--you'll see that the searches for that term in Utah blow every other state out of the water, including California.

TylerHDec 03, 2008 at 8:03PM

Polygamy- Very strong correlation with Utah! Utah seems to be the most common outlier from the phrases people have picked. It's probably because it has the most homogeneous population.

jack bierlowDec 04, 2008 at 2:13AM

France splits along party lines,
but Conspiracy is universal.
What do they Fear in NC?
and why do they put up Fences in VT?
(to keep the Phish fans in?)

The choices in Montana are Meth and/or Fly Fishing.

robert ivanDec 04, 2008 at 3:56AM

Penis Pumps are HUGE in Mississippi. Not so much in VT.

ramkiDec 04, 2008 at 4:07AM

This point is not something new, but has been repeatedly hammered at in the book 'Freakonomics' by Levitt and Dubner

This thread is closed to new comments. Thanks to everyone who responded.