on the basis of historical research, ethnography, discourse analysis, and sociolinguistic interviews, [to] describe how a set of linguistic features that were once not noticed at all, then used and heard primarily as markers of socioeconomic class, have come to be linked increasingly to place and ‘enregistered’ . . . as a dialect.The topic has since led to a special issue of American Speech (here, and the quote above was pulled from Michael Paul Adams' intro to the issue). Along with work on Pittburghese and other cool stuff, the issue included work on Wisconsin and the Upper Peninsula of Michigan, giving evidence that regional speech patterns have been recognized in Wisconsin increasingly over the last century, especially in recent decades. Instead of features that marked class, in the Upper Midwest it's typically features that were once associated with immigrant populations, like 'stopping' (using a d sound for 'th' sounds, for instance).
One bit of evidence bearing on this question hasn't, to my knowledge, been talked about: We've got ever-better corpora which we can search, like the Google NGram Viewer, which we've played with here in the past. If you search for 'Wisconsin dialect' you don't really get anything to speak of, but 'Wisconsin accent' yields an interesting pattern, shown here compared to 'Minnesota accent' (I start with 1940 because nothing shows for 'Wisconsin accent' before then; as always, click to embiggen):
So, our western neighbor's speech pattern was apparently talked about somewhat earlier and somewhat more in books than ours. But these numbers are all relative, so compare this to a state of roughly similar size (i.e. in population):
If we generalize to 'Southern accent' you see some real presence (and btw, 'Upper Midwest accent' doesn't get you anything really):
It's not surprising that Wisconsin (and Minnesota) lag behind the South in this regard, just given the historical demographics of European settlement.
Oh yeah, just because:
Update, 3:30: Was kinda in a hurry before and didn't check the Library of Congress "Chronicling America: Historic American Newspapers". Checking 1836-1922 to see if yielded earlier attestation, I found only a few. A couple had 'Wisconsin' and a form of the verb 'accept' (i.e., these were OCR glitches). The others, it looks like, refer specifically to Wisconsin people with 'foreign', even specifically 'German' accents. That's highly consistent with the association of Wisconsin with immigrants and with immigrants who didn't speak English or, in this case, spoke it with an accent. Here's an early example, from the Omaha Bee:
Gee, there might be a little story in here.
Johnstone, Barbara, Jennifer Andrus, and Andrew E. Danielson. 2006. “Mobility, Indexicality, and the Enregisterment of ‘Pittsburghese.’” Journal of English Linguistics 34: 77–101.