Wednesday, May 27, 2009

the "hispanic" fallacy

so, while this doesn't make nate silver any less one of my imaginary boyfriends, i though it worthwhile to point out how even the smartest analysts are at the mercy of the problematic "hispanic" fallacy.

i was reading some fivethirtyeight posts about sotomayor's nomination, and i came across this post, where nate debunks the idea that "non-white hispanic" voters are more opposed to marriage equality than white voters.

so, although i really like the point silver's making, and wouldn't be able to question his actual math even if i wanted to, i (legitimately or not) decided to cherry pick this to bring up a major issue in the way we think about voting blocs and race as a whole. maybe this is just my inherent qualitative (anthropological) bent, but isn't a far more interesting and telling question the ways that opinions among "hispanics" differ? the category "non-white hispanic" is almost useless, really, for predictive purposes. there is such a world of difference among the experience and political calculus of, say, a first generation chicano voter in southern california, a second generation voter of cuban descent in miami, and a self-identified newyorican whose grandparents moved to the states 60 years ago, that it seems pretty inadequate to talk about some mythical category of "non-white hispanics" as a coherent voting bloc.

to be fair, it isn't like it's silver's fault. it's what he's got to work with. race in the united states is a deeply-ingrained, all-encompassing, and oversimplified category-creating discourse perpetuated by everything from talking heads on CNN to the way we apply for our driver's licenses. it just seems that in a nation where, for instance, the "non-white hispanic" population is increasing in size and diversity, that we can't rely on these overly broad categories to accurately tell us anything important about the way we and our fellow citizens actually behave. or perhaps more to the point, our continued reliance on these categories of understanding tells us a lot, and none of it comforting, about the way we continue to behave towards our fellow citizens.

7 comments:

  1. I'm actually going to have to refute Molly. Against all odds, there is a large divide between Hispanics who identify as white and those who do not. It affects political views, social views, and an very sharp divide between whom he or she marries.

    The irony of the whole thing, is that a White Hispanic and a Non-White Hispanic may actually look exactly the same, physically speaking. That's what makes it so predictive. It's not an external, phenotypical assertion, but an internal self-identification.

    For more:
    http://pewhispanic.org/reports/report.php?ReportID=35

    ReplyDelete
  2. i'll buy that. i probably oversold my case (splitting my attention between annual report writing, vaccination schedule and blogging probably doesn't help with nuance haha). my point wasn't that self-ascribed racial identity isn't ever salient, but rather that i think, in thsi case, it's obscured the need to ask equally important questions about internal differentiation and identity in a qualitative way (my point re: qualitative and quantitative being obscured by my use of "predictive"...should have been "analytically rigorous" or similar). this is true for other dominant modes of social understanding in other places as well (for instance, there's arguably a strong analogy between race in the US and religion in the middle east as the way people intrinsically understand identity and social relations among groups and individuals). at the end of the day, this really has to do with my mistrust of quantitative analysis without qualitative interrogation, which, by necessity draws on broad categorical strokes (which i almost always find more interesting to break down further).

    more thoughts after i leave work and can read the pew thing.

    ReplyDelete
  3. also should have been more consistent "hispanic" vs. "non-white hispanic". silver just uses "hispanic" as his variable which elides even more significantly important internal differentiation, a la luis' comment.

    ReplyDelete
  4. Totally fair. I think good reports on U.S. Hispanics need to isolate for country of origin, as well as self-described race, and then analyze both within and between those blocs. That's the only way to really understand the demographic. Hopefully, by the time "Hispanic" becomes an ethnic plurality, we'll have had the good sense to discard the term and will be understood by national origin, at the very least.

    ReplyDelete
  5. This comment has been removed by a blog administrator.

    ReplyDelete
  6. This comment has been removed by the author.

    ReplyDelete
  7. This comment has been removed by the author.

    ReplyDelete