December 8, 2015

What you do know that isn’t so

The Herald (and others) are reporting an international Ipsos-Mori poll on misperceptions about various national statistics.  Two of the questions are things I’ve written about before: crude wealth inequality and proportion of immigrants.

New Zealanders on average estimated that 37% of our population are immigrants.  That’s a lot — it’s more than New York or London. The truth is 25%, which is still higher than most of the other countries. Interestingly, the proportion of immigrants in Auckland is quite close to 37%, and a lot of immigration-related news seems to focus on Auckland.   I think the scoring system based on absolute differences is unfair to NZ here: saying 37% when the truth is 25% doesn’t seem as bad as saying 10% when the truth is 2% (as in Japan).

We also estimated that 1% of the NZ population own 50% of the wealth. Very similar estimates came from a lot of countries, so I don’t think this is because of coverage of inequality in New Zealand.  My guess is that we’re seeing the impact of the Credit Suisse reports (eg, in Stuff), which say 50% of the world’s wealth is owned by the top 1%.  Combined with the fact that crude wealth inequality is a bogus statistic anyway, the Credit Suisse reports really seem to do more harm than good for public knowledge.

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

    I’ve noticed it also claims that 37% of New Zealanders don’t affiliate with any religion, whereas the 2013 census found that number to be 42%.

    Considering that was up from 35% in 2006, it seems pretty unlikely that the trend will have reversed so drastically in the last two years.

    I had to follow a series of three links from their slides, each claiming to take me to the source of the “actual data”, before I was actually taken to the right document here: https://www.ipsos-mori.com/Assets/Docs/Polls/ipsos-perils-of-perception-sources-2015.pdf

    It seems their source of data for the number of religious New Zealanders comes from a Pew poll which, from its URL, appears to be from 2012 (although it uses 2010 population data). Why they used this instead of our more recent (and obviously more complete) census data isn’t clear, especially since they did use census data as the source for some of their other statistics.

    8 years ago

    • avatar
      Thomas Lumley

      There would be a case for using the same source for all countries for the religion question, but since they used census data for Great Britain, that doesn’t explain it either.

      8 years ago

    • avatar
      Richard Penny

      I get 38.5%. I got this from the number that ticked “no religion” divided by the total number for all ages.

      Caveats. One could argue that you should take it only for adults. Also you need to be consider the tick box “Object to answering” which is on the NZ Census form. Is that no religion or not?

      I also raise the Jedi Knight campaign we had a few censuses ago.

      And also the feedback we get EVERY time the census questions are reviewed that most of the boxes apply to denominations, not religions.

      8 years ago

      • avatar
        Thomas Lumley

        There’s a defensible range of values from the Census, from 38.5% at the low end through the 41.9% on the Census QuickStats page to 41.9+4.4=46.3% for “none” and “object” as a proportion of responders.

        Personally, I think I’d at least drop the substitute records from the denominator. The item non-response and object-to-answer are trickier.

        However, using a survey when there is more recent census data seems inefficient.

        8 years ago

    • avatar
      David Hood

      I vaguely recall reading something about the differences in question construction/categorisation with the Pew poll compared to the census, but the details escape me despite some quick googling.

      2006 (Census) Christian 2082942

      2010 (Pew) Christian 2490000

      2013 (Census) Christian 1906398

      2006 (Census) No Religion 1297104

      2010 (Pew) Unaffiliated 160000

      2013 (Census) No Religion 1635345

      2006 (Census) Muslim 36072

      2010 (Pew) Muslim 50000

      2013 (Census) Muslim 46149

      2006 (Census) Hindu 64392

      2010 (Pew) Hindu 90000

      2013 (Census) Hindu 89919

      So Pew’s methodology seems to favour reporting any religion (as all the religious raw numbers are above both the 2006 and 2013 census figures).

      2006 (Census) Total Population 4027947

      2010 (Pew) Total Population 4370000

      2010 (StatsNZ 2010 Projection) Total Population 4370000

      2013 (Census) Total Population 4242048

      As well as the “Objects”, One thing I am wondering is if Pew restricted people to one answer- assumed monotheism (because their respondents total was a little less than the projected population), when from the 2006 census we known the total religions stated was 2356491 while the total people stating religion were 2271921. This assumption is much more significant in the case of countries like Japan, where 80% of the population say religion is not important but 84% say they are Shinto, 79% say they are Buddhist, Christianity 2% and Other 8% (which is a lot more than 100% of the population if you are making a monotheism category error).

      8 years ago