A statistical gripe

Statistics are great tools, so long as you don’t lie about them. I just came across a number that seems to have taken hold in international circles, which is the “Share of individuals that cast a ballot during an election, as a percentage of the voting-age population”. This is referred to as “Voter Turnout” by the OECD and the Swedish International Institute for Democracy and Electoral Assistance (IDEA).

Given that voting is compulsory in Australia, you might be a tad surprised to know that the folks in Paris and Stockholm reckon our “voter turnout” is below 80%, and has fallen substantially since 1945. Surprised, that is, until you realise that immigrants are in the denominator. Clearly it was too tough for those compiling the data to divide voters by the number of citizens, so they figured no-one would notice if they just divided by the number of adults. That’s fine if you’re a country that keeps foreigners out, but as a country with large-scale migration (and an increase in the migrant share since WWII), it’s no great surprise that this makes it look like voter turnout in Australia is low and declining.

By contrast, when you look at AEC data on the share of registered voters who turn up to the polling booth in Australian elections, the figure is about 95%. And since the ratio of citizens to registered voters is about one to one, we can safely say that Australian voter turnout is around 95%.

(xposted @ al.com)

3 thoughts on “A statistical gripe”

  1. So the OECD are liars? and bigots too for keeping out foreigners.  🙂 


  2. This has got me worried – who’s in the denominator of the OECDs standardised unemployment rates?


  3. I’m not sure if I entirely agree with your citzens and registered voters are about 1:1 – particularly at the young non immigrant end of the spectrum, there are quite a few people who aren’t registered to vote. But the AEC measure is surely a better one than not excluding immigrants.


Comments are closed.

%d bloggers like this: