This quiz pinpoints your American dialect down to the town

There are a bunch of quizzes out there that purport to tell you what American dialect you speak. But now there's one that tells you what city your accent and dialect is from. For some of you, it's an amazing thing that pinpoints your hometown exactly. For others, it'll tell you that, for whatever reason, you don't sound like anyone else around.

The above map (where you learn that the northeast pronounces "centaur" differently from everyone else) is from NC State PhD student Joshua Katz's project "Beyond 'Soda, Pop, or Coke.'" The project is a slick visualization of Bert Vaux's dialect survey, and lets you look at maps of the results of 122 different dialect questions, either as a composite showing the variation across the country or each individual dialect's prevalence across the country. You can also see the exact results of a number of cities.

The project is described this way on its website:

Using data from Bert Vaux's dialect survey, we examine regional dialect variation in the continental United States. Each observation can be thought of as a realization of a categorical random variable with a particular parameter vector that is a function of location—our goal was to interpolate among these points in order to estimate these parameter vectors at a given location, making use of a combination of kernel density estimation and non-parametric smoothing techniques. Results in a smooth field of parameter estimates over the prediction region. Using these results, a method for mapping aggregate dialect distance is developed.

Most recently, the project's added a dialect quiz. You can take either the full 140-question version or a random 25-question version. I've taken both, and got the same results. Which look liked this:

This quiz pinpoints your American dialect down to the town

The quiz explained the results this way:

Based on your responses, the map at right shows the overlap between your speech and the various dialects of American English, as measured by data from the Harvard Dialect Survey, conducted by Bert Vaux and Scott Golder. The numbers next to the most/least similar cities (which correspond to the colors displayed in the heatmap) are estimates of the probability that a randomly-selected person in that city would respond to a randomly-selected survey question the same way that you did.

It's pretty interesting, except that I think my refusal to call ANY place "the City" (and marking "other" instead of L.A., NYC, Boston, or Chicago) is the reason I keep getting Bay Area cities rather than my hometown of Los Angeles.

So did anyone else take it? Was it spot-on or way off? And, out of curiosity, what results are people for whom English is a second language getting?