Democrats pin hopes on sushi

by Ethan on November 01, 2007

I love a juicy steak, but I get withdrawal pangs if I can't have sushi at least once every couple of weeks.

The other day we were sitting around Urbanspoon World Headquarters discussing the relative merits of these two wonderful cuisines, and we began to wonder if what we eat relates to the way we vote.

Naturally, as scientists, we had to dig deeper. After careful and extensive investigation it appears there is in fact a strong correlation between politics and food.*

We calculated the ratio of steak to sushi restaurants -- what we're calling the Steak/Sushi Index -- for cities around the country. We then compared that to the election results from Bush vs. Kerry in 2004 (measured by the spread in the percentage of the vote).

Cities in general lean Democratic, but there's still a clear relationship here -- the more sushi, the more liberal. Somehow this feels intuitively obvious. You'd expect more red meat in red states.

But once you've established a connection between voting, steak and sushi, you can't help but try to apply that knowledge. As my statistics professor used to say, "correlation is causation". If we could only see how public interest in sushi vs. steak is changing over time, we could predict which way the political winds are blowing.

But we can! Thanks to Google Trends, we can get a glimpse by measuring search traffic for the terms "Steak" and "Sushi":

By this measure, you can see why Kerry lost in '04 (sushi down), and why the Democrats took the majority in '06 (steak down). If this trend continues, it looks like a Democratic sweep in '08!

But wait, before liberals get too smug and conservatives despondent, online data can be misleading. The netroots are powerful, but it's possible that there are lots of people living in this country who don't search for steak and sushi on Google. We looked at the growth rate of cities, and found another stark correlation.

Steak-eating, Republican-voting cities are getting bigger faster than their sushi-loving, blue state brethren.

The race is still on.

By the way, we pored over lots of other statistics in our quest for "meaningful" correlations -- like the fattest vs. fittest cities (nope, steak doesn't necessarily make you fatter), the most wired cities (no real connection between internet access and the food you eat), and the cities with the most attractive people (it appears that eating sushi makes you slightly better looking, statistically speaking).

It is also fitting that a new restaurant has recently opened in Washington DC that focuses on offering great steak AND sushi. Perhaps a sign of bi-partisan times to come?

Calculating the Steak/Sushi Index (SSI)

We found the Steak/Sushi Index (SSI) with a sophisticated algorithm -- we divided the total number of steak restaurants by the total number of sushi restaurants. So cities with values greater than one have more steak (shown in red), and those with less than one favor sushi (in blue).

* Reading the politics and food graph

I realize the numbers on the axes of the graph may be a little unclear, so here's a little bit more explanation of how to read them.

On the left side of the graph you see the ratio of steak restaurants to sushi restaurants - so Dallas has three times as many steakhouses as sushi spots, whereas San Francisco has less than half as many. That is tracked by the blue line.

On the right side, we're showing the point spread between Bush and Kerry. So in Dallas Bush was up by 20 points (e.g. 60% vs. 40%) and in San Francisco Bush was down by 50 points (e.g. 75% vs 25%). That's tracked by the red line.

Photo available from flickr user sparktography under Creative Commons license.