January 26, 2013

Selma and Stonewall

Today’s fascinating survey time series: for fifty-five years Gallup has been asking people in the US if they approve of interracial marriage. (via Paul Krugman, in the NY Times)

Gallup, via New York Times

On the one hand: Yay, progress! On the other hand, one in seven people still not only don’t approve but are prepared to admit this to a random caller on the phone.

 

 

avatar

Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »

Comments

  • avatar
    Margaret

    People might say they approve, but their actions speak louder than their words…

    “In 2007, 4.6% of all married Blacks in the United States were wed to a White partner, and 0.4% of all Whites were married to a Black partner.”
    http://en.wikipedia.org/wiki/Interracial_marriage_in_the_United_States#Black_and_White

    It looks like reality is still in 1958.

    11 years ago

    • avatar
      Thomas Lumley

      It’s not the same thing at all. The change is that people don’t disapprove of other people’s interracial marriages. There’s a difference between assortative mating and intolerance.

      That’s the point behind the title of the post.

      11 years ago

  • avatar
    Margaret

    There is also a difference between what people say to pollsters about what they think, and what they really think. It is often called “revealed preferences” in economics … people say they love something that is perceived as good (eg going to church or giving to charity) but they don’t really (so they don’t go to church or give to charity…etc).

    What I am saying is that the actual figures reveal a preference which is quite different to the stated one. Assortive mating doesn’t do it … there are many whites and blacks with the same characteristics other than skin colour eg education, interests …. but people are still showing a revealed preference to mate solely along colour lines.

    The difference between stated preference and revealed preference is also common in New Zealand on issues of race …

    11 years ago

    • avatar
      Thomas Lumley

      I still don’t think this contradicts the results. It’s not just a revealed preference vs stated preference issue.

      The distinction is clearer with same-sex marriage. I expect everyone would agree that “I approve of same-sex marriage” (nearly 60% in NZ) is very different from “I am likely to marry someone of the same sex” (perhaps 1% in NZ). There is (or will be after the law passes) a stated vs revealed preference issue, but it obviously isn’t the main reason for the discrepancy.

      The same issue is present in the interracial marriage. Certainly it’s true that racism is part of the low rate of interracial marriage, and that is revealed preference data. But it’s not revealed preference data about the same question that was on the survey. It used to be the case that most people in the USA disapproved of other people marrying those with different colored skin, and were willing to say so (and go much, much further than just saying so). Now they mostly don’t.

      11 years ago

  • avatar
    Margaret

    Or say they don’t but really do … and so don’t ever consider marital partners. Not many – if any – Pakeha politician would have survived saying the inverse of what Hone said … but that doesn’t mean they don’t think it!

    11 years ago

  • avatar
    Margaret

    BTW I regarded your same-sex analogy as a red-herring. There is nothing inherently equivalent stopping interracial marriages — as can be seen from the very high level of interracial marriages in NZ.

    11 years ago

    • avatar
      Thomas Lumley

      It’s not quite the same, but it really isn’t a red herring.

      Let me try this once more without any analogies.

      There is a difference between disapproving of the existence of interracial couples and not personally being willing to consider being part of one. You can make a good case that both are racist, but the difference is still real and important.

      While you can conclude that people who are in an interracial marriage approve of the existence of interracial marriages, you cannot conclude that people who are not in an interracial marriage disapprove of the existence of interracial marriage. That’s more than revealed preference is capable of revealing.

      The distinction between one’s own behaviour and one’s approval of other people’s behaviour isn’t just a matter of private opinions, either. There are public consequences. It was only 1967, after all, when the US Supreme Court overturned laws against interracial marriage (Loving vs Virginia). And Alabama kept its (unenforceable) laws on the books until 2000, and even then 40% of votes were against eliminating the law. And historically it hasn’t just been legal interference that interracial couples have had to worry about.

      11 years ago