You Say Tomato, I Say Soda. Or Is it Pop?

Here’s a nifty new way of looking at some old regional differences in dialect. NC State graduate student Joshua Katz was looking for an idea for his end-of-year statistics project, when he came across interesting linguistics data from Dr. Bert Vaux of Cambridge University. Vaux had collected data on regional dialects in the U.S. via a 120-question survey that asked about everything from the pronunciation of “caught” to whether “y’all” is preferred over “you.” Then he had mapped that data, to indicate where particular dialect differences were more prevalent.

“Dr. Vaux’s maps showed each response as a single color-coded point, so you could see individual instances of each answer,” Katz says. “I wondered if there was a way to take the existing data and create maps that gave a more complete picture of national dialect differences.”

Katz created a statistical algorithm that weighted the responses around a particular location. The approach let Katz create colorful maps that show how regional dialects are spread across the country.

So what do you call a sweetened, carbonated beverage?  According to the map, coastal residents are predominately soda people, the pop drinkers are concentrated in the upper Midwest, and a few solid coke holdouts dwell in the southern states.

Beyond the soda question, Katz has created an interactive map which lets you explore the responses to each of Vaux’s 120 questions.  Katz has also created a map with a pull-down menu which lets you see the “aggregate dialect difference” in selected cities. Basically, “aggregate dialect difference” just means, “Who else in the U.S. talks like me?”

Katz hopes to continue collaborating with Vaux on future projects. As for why a statistician decided to tackle language? “I’ve always found variations in dialect fascinating – language says so much about who a person is,” Katz says. “To me, dialect is a badge of pride – it’s something that says, ‘This is who I am; this is where I come from.’”