A Shot at Linguistic Profiling

Reading through the previous blogs about taking the profiling quiz, what struck me most was how various everyone's levels of success were. People's scores ranged from 100% correct to nothing better than random chance. That seems to imply that one's score is heavily dependent on their background and personal exposure to these accents. I grew up in a very white suburb, but I went to a school that was relatively racially diverse and my town was right next to a low-income, Hispanic neighborhood. I don't think I was ever especially exposed to accents much different from my own.

That said, I got 5-6 correct on my first try. (I didn't keep count, which was stupid in retrospect, but the number was definitely around there.) The ones I remember I got right immediately were the white speakers, the Indian speaker, and the Middle Eastern speaker. I had trouble with the black and Hispanic speakers, although by the end I was able to pick up some differences, such as an overall softer vowel and consonant quality, especially at the start of words, in the Hispanic speakers. With some speakers I was completely off-base, cycling through literally all other options before typing in the correct one.

Ultimately, I think linguistic profiling is legitimate (not in a moral sense but in that people are able to correctly profile people through their voices), but it depends to a large degree on the background of the person doing the profiling. Since I had a relatively homogenous childhood, I didn't do very well, but others should and did have more success.