My dialect — the sound, vocabulary, and grammatical structure of the way I speak English — is most similar to the dialect spoken by people in Topeka, Kansas. That's according to a popular survey and data visualization making the rounds on social media, and the result makes sense. I was born in Topeka and spent the first 12 years of my life there.

It's pretty cool that a dataset can figure out that connection based solely on my answers to a series of 25 questions. Do I pronounce "caught" and "cot" the same? (Yes.) Do I think it's acceptable to say something like, "I write exclusively about science anymore"? (Yes. Haters can hate.) Do I call carbonated beverages "soda", "pop", or "coke"? (I think I say "soda" sometimes and "pop" other times and I'm not sure why.)

But those questions — and our collective answers — are good for a lot more than simply performing "guess your hometown" parlor tricks. You can learn a lot about where we're going and how we're changing as a society. Because, here's the thing, contrary to what you might think, the United States isn't losing its dialects. We're not all speaking more similarly to one another. In fact, sociolinguists say the opposite is true. Even in a world where people in Topeka, Kansas and Brooklyn, New York listen to the same music, watch the same movies, and share words instantly, the way those people talk isn't merging into a single, consumer-culture voice.

If that fact seems surprising, it's probably because the myth — that distinct regional and local dialects faded over the course of the 20th century and are on their way to oblivion — used to be what experts believed, too. When Bert Vaux, a linguist at Cambridge University, was working on his Ph.D., he was taught that American dialects had largely disappeared. Mass media and mass culture were creating a mass speaking voice.

Then, about 10 years ago, while working as a professor at Harvard, Vaux put together a survey. Using 122 different speech variations — some having to do with vocabulary, some with syntax, and some with pronunciation — he asked volunteers a series of 140 different questions and linked their answers to their hometowns. Finished in 2003, the Harvard Dialect Survey forms the basis of the more-recent online quizzes you've probably taken through Facebook and Twitter. It showed that regional variations in dialect first mapped in the early part of the 20th century still existed.

That's a big deal because those regional dialect families actually date back to the first migrations of English-speaking immigrants into North America. The Northern dialects — which actually stretch from Boston to Eastern Minnesota — have their origins in southeastern England and Puritan settlements. Southern dialects began with immigrants for southwestern Great Britain, including Wales. The third dialect family, the Midlands, comes from a wave of immigrants originating in northern England, Ireland, and Scotland. It spread west horizontally from the Philadelphia area. My tendency to use "anymore" as a positive part of speech — I do "x" anymore — as opposed to purely using it in the negative — I don't do "x" anymore — is part of the Midlands dialect.

Everyone knew these families once existed. Dialect surveys in the 1930s and 40s had identified them and matched them to historical immigration patterns. But nobody expected them to still exist. Turns out, Vaux told me, that's because researchers had been associating the dialect families too heavily with colloquialisms that had dropped out of common speech — things like the phrases farmers used to call in their cows at night. That stuff really had vanished. But the dialect families remained, they were simply now united around different things — like what we call tennis shoes (or "sneakers" for some of you), or how we pronounce certain vowel sounds.

Vaux's discovery corresponds to work done by other sociolinguists, who say that even beyond the maintenance of these classical dialect families, Americans really aren't speaking more similarly to one another. In fact, there are actually new dialects in the process of emerging.

For instance, the West — which you can think of as pretty much everything west of the Mississippi — has long been a mishmash of dialects. It was settled much later than the rest of the United States. Its residents came from a much more diverse set of places. It never really had the kind of coherent city dialects you can hear on the East Coast, or even larger regional dialects, on the scale of states. Instead, if you look at maps like the ones produced by the Atlas of North American English, the West looks big and open and wild — characterized by language features that span half a continent. The tendency to pronounce "caught" and "cot" as though they are exactly the same (and, more importantly, to not really grok that they could be pronounced differently) is a major part of the Western voice.

But that could be changing. Over the last 20 years, some researchers have started building cases for regionalization within the vast expanse of the West. Scientists have identified the emergence of several dialects that fall under a larger Pacific Northwest English dialect family, said Walt Wolfram, a sociolinguist at North Carolina State University. Portlandian, it seems, may actually be on it's way to becoming a thing.

Researchers say this because the way younger residents of Portland speak is becoming more distinct, and it's different from how older residents speak. The dialect includes aspects of the larger Western dialect — including the caught/cot merger. But that used to be something only some Portlanders did, writes Portland State University linguist Jeffrey Conn. Starting with the Baby Boomers, it's become much more the norm. Younger Portlanders are also picking up a tendency to end their declarative sentences on an upswing of the voice, making statements sound kind of like questions.

Dialects abide and new dialects form because the way we speak is a major part of how we tell the world about our identities, says William Labov, a linguist at the University of Pennsylvania and the man behind the Atlas of North American English. American English has always been unstable because our regional identities are younger and more open to change, he told me. If your parents move to a new community when you're a child — or if you move to a new community yourself as a young adult — your dialect doesn't stay tied to that of your parents. It changes to match your community.

Of course, that means not all dialects will last. In the game of how we speak, you win or you die. When we think that Americans are starting to all speak the same, what we're often actually seeing is the loss of a few "famous" dialects, without taking into account the strengthening or emergence of others. Southern dialects, for instance, are leveling out, Wolfram and Labov told me. The younger they are, the less "Southern" a Southerner is likely to sound. But, at the same time, Northern dialects, especially those associated with cities surrounding the Great Lakes, are becoming more distinctive.

All of this ends up influencing what you see when you take a quiz that shows you how your dialect matches up to dialects around the country. Yes, my #1 match was the city in which I was born … but that was just a 60% match. And I was almost as strongly matched with several cities in northern California and Nevada. Places I've never lived and, in some cases, never been.

Why don't I match any better to my hometown? Because dialects change with social identity. I've lived in a lot of other places since I was 12 and my dialect has changed along with those moves and shifts in the communities I identify with.

Why would I match so well with places I've never lived? Because there's a lot of similarities in the dialects of the West overall. It's not too surprising that my Western dialect (via Kansas) would have a decent amount in common with other Western dialects. Those similarities reflect a shared history of recent migration and a shared identity that's only now starting to separate out into regionalisms.

More importantly, given another decade, my results could end up becoming entirely different. I could end up sounding less like my hometown dialect and more like the place I now live — Minneapolis. My hometown dialect could end up sounding more distinct than it did when I last lived there. Language is change. A single map only gives you a snapshot.