John Koontz, a linguist at the University of Colorado, has a website full of information about Siouan and Other Native American Languages, with a particularly interesting page about etymologies (including Kemosabe and Tonto, an entry that manages to cite both Aeschylus and the publication glitches of the Eastern Ojibwa-Chippewa-Ottawa Dictionary). The beginning of the Nebraska entry will give you an idea of the level of detail:
The state is named for the Platte River, which is called in Omaha-Ponca NiNbdhaska (=khe) ‘(the) Platte River’; literally ‘(the) Flatwater’, or in Ioway-Otoe N^iNbraske (or, more recently, -brahke or –brat^ke) [all with the same meaning].
My suspicion is that the actual source was Ioway-Otoe. This comes from two factors. First, during much of the later 1700s and 1800s, the Otoe were situated at the mouth of the Platte, in a position to present their own name for the stream to visitors. Second, Nebraska looks to me like a collapsed syllable spelling Ne-bras-ka, probably intended to represent what I would write as in the Lewis & Clark Phonetic Alphabet (LCPA) as Nee-BROSS-kay. That is, I suspect “ka” was intended to represent phonetic (NetSiouan) /ke/, not /ka/ (LCPA kay, not kah), and that would have to be the Ioway-Otoe version. My feeling is that real phonetic /ka/ would have been written “kar,” cf. “Mahar” (this really is a Lewis & Clark spelling) for UmaNhaN ‘Omaha’ or “kah.” The Dhegiha languages retain ska from *ska (LCPA skah) in final position while Ioway-Otoe converts it to ske (LCPA sk ay).
Once the word was written as a lump “Nebraska” and subjected to pronunciation by English speakers who hadn’t heard the original, the final syllable was changed to phonetic (NetSiouan) /ka/ (LCPA kah), or, actually, /k
/ (LCPA kuh). In the same way the initial “ne” acquired a lax (short) e (LCPA neh) or schwa (LCPA nuh) pronunciation instead of i (LCPA ee) (long e in English terms) pronunciation, and the medial a in -bras- was fronted to the low front a of American cat (instead of the low central a of American father).
Of course, early popular transcriptions are incredibly imprecise, and I don’t have any information on the early history of the word in English. Maybe final “ka” did represent phonetic /ka/ (LCPA kah), in which case, it would have to be a Dhegiha form something like the Omaha-Ponca version that was the source. In fact, with this word any of the Dhegiha languages would produce pretty much the same effect on English ears. While Omaha-Ponca would seem the most likely suspect because the Omaha and Ponca were conveniently nearby, the Kansa and Osage were also originally both below the Platte along the Missouri and their languages are also plausible sources for the names of major tributaries upstream…