It asks you all sorts of things, about everything from vowel pronunciation to word usage to idiomatic expressions to which syllables in a given word you stress, and when you're done, you get a lovely map with big blobs of color showing the areas of greatest --- and least --- similarity to your own speech.
Here is mine:
 |
| My dialect map! I am clearly an Upper Midwesterner, but the rest of the Midwest and Mountain West aren't too far off. The South and New England, though? Might as well be another country. |
My best match is Grand Rapids, Michigan, followed by a four-way tie between Lansing, Michigan; Rockford, Illinois; Flint, Michigan*; and Cedar Rapids, Iowa. (The lattermost city is actually one I've spent a lot of time in. I've been to Michigan --- went to science camp on the Upper Peninsula of Michigan in high school --- but haven't lived there).
My worst matches are mostly in the Deep South: Huntsville, Alabama; New Orleans, Louisiana; Metairie, Lousiana**; and Birmingham, Alabama. The one exception is Plymouth, Massachusetts.
That map is particularly visually striking in its division of the country so neatly in half --- it looks like someone drew a line diagonally from the Michigan/Ohio border on the western shore of Lake Erie to the southwest corner of New Mexico, and dyed the top half red and the bottom half blue. There's a bit of an intermediate belt, especially toward the West Coast, but all across the central part of the map the red and the blue are right up next to each other.
They give you a numerical value for each city on either list; that number refers to the probability of any random person in that city giving the same answer to any random question on the quiz that you gave. The map makes the differences seem starker than they are --- the spread between my highest- and lowest-scoring cities was only about eight points. This makes sense when you consider how young a country the US is, and how relatively uniform American English is compared with, say, British English*** or other European languages like Dutch, German, French, Italian, Spanish and probably zillions of others, which have dialects so different from one another (and from the standard language, which American English doesn't really have --- probably because we don't need one yet) that a speaker of, say, standard German would be unable to understand a German-speaking person who speaks Low German or any of the High Franconian or Upper German dialects.
At the end of the quiz, the ask you where you live now and where you spent most of your childhood; they also ask if you are a native English speaker, so maybe if you're not they will ask you about your native language. The data it draws on are restricted to continental American English (i.e., the US minus Alaska and Hawaii), so I don't know how enlightening it will be to someone who speaks any other form of English. I could see it being of interest to non-native English speakers who learned American English; maybe it could tell you something about where the person who taught you English was from!
*So, if people reading this blog ever try to imagine my voice, you could probably do worse than imagining my words read by Michael Moore. Except for the part where Michael Moore is a guy, but whatever.
**I didn't know there was a Metairie, Lousiana. It has a beautiful name, even if it is apparently one of the five places in the nation where I am most likely to have serious trouble making myself understood.
***I would expect Canadian and Australian English also to have regional dialects that are still pretty similar to each other overall, what with their having a similar history of being a British colony at around the same time. I don't know this, though, so it would be nice to have confirmation --- or contradiction, for that matter --- from any Canadian or Australian readers I might have.
