May 13, 2013

Your guess is as good as ours

There’s currently discussion in NZ about whether to change the 5-yearly census.  North America is providing some examples of what not to do.

Canada decided a while back that they were going to chop most of the questions off the census and put them in a new survey.  The new survey is still sent to everyone, but is voluntary — the worst of both worlds, since a much smaller survey would allow for more effort per respondent in follow-up. Frances Woolley compares the race/ethnicity data from the 2006 Census and the new survey: the survey is dramatically overcounting minorities.

In the USA, a Republican congressman has proposed a bill that would stop the Department of Commerce and the Census Bureau from collecting basically anything other than the census.  That would wipe out the American Community Survey, the detailed 1%/year sample that provides a wide range of regional data. It would also wipe out the Current Population Survey, used to estimate the unemployment rate.  Fortunately for the US economy, there’s no chance of this bill becoming law: the business community hates it, and Senate will never pass it.  It’s still worrying that there’s a public-opinion advantage in pretending you want to abolish the government’s economic data collection.

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

    “Canada decided a while back … ”

    Here’s a bit of background on this one. The decision was entirely driven top-down by the Conservative (our right-wing Republican-lite political party, currently in power with a majority) government, directly over the objections of statisticians, Statistics Canada, and most users of census data. To this day, no-one is really sure _why_ they actually did it; the reasons given were laughably insincere.

    The head of Stats Can resigned in protest after this decision, saying:

    I want to take this opportunity to comment on a technical statistical issue which has become the subject of media discussion. This relates to the question of whether a voluntary survey can become a substitute for a mandatory census.

    It can not.

    —Munir Sheikh

    My best guess is that as soon as the Conservatives are voted out of power, the long-form will be reinstated, as the other two large political parties in Canada were strongly against the decision.

    11 years ago

    • avatar
      Richard Penny

      What I heard, with caveats as to hearsay, is that it was a way for the government to say “we are worried about your privacy”. A result would be that people spend less time investigating the government’s other instrusions that materially affect the privacy of their information.

      Another example was Michelle Bachman, an American congresswoman, who wanted no census long form to “protect” her privacy. Seems all the stuff she was trying to protect she had freely made available for anyone with access to internet.

      To be realistic, defending collecting data is very hard. Unless of course you collect it without making it obvious (e.g. customer databases, web exhaust) and make money from it.

      Reminds me of that old joke. “I don’t care what the world thinks of me, as long as mother doesn’t find out”

      11 years ago

    • avatar
      Thomas Lumley

      Yes, we’ve quoted Dr Sheikh’s statement a couple of times before on StatsChat.

      11 years ago

  • avatar
    Joseph Delaney

    Would it be bad to ask whether undercounting minorities might actually be the point of the Canadian decision? It is the sort of thing that you’d never want to say as a politician (unless you like being hated by a segment of the population). But it could led to understating the magnitude of some chronic and difficult to address social justice issues (especially among the First Nations).

    11 years ago

    • avatar
      Thomas Lumley

      The survey is overcounting most minorities, so if that was an aim it isn’t working.

      11 years ago

      • avatar
        Joseph Delaney

        There goes a fine hypothesis, slain by exposure to actual data.

        This just mystifies me then. If there is no political advantage to this decision then why trash the data for no reason at all??

        11 years ago

        • avatar
          Thomas Lumley

          There is a political advantage, just not that one.

          The census, as an example of government meddling in the affairs of free citizens, is unpopular with a certain subset of right-wing voters. Only a subset, because big business likes having the data, but an important subset for the current Canadian government. It’s the same motivation as the US bill to get rid of the CPS and ACS, only in the US this sort of law won’t actually get passed, because the Senate can’t even pass things that are supported by huge majorities. The real danger in the US is defunding of the surveys as part of budget negotiations.

          11 years ago