September 14, 2013

Not how election polls work

The Dominion Post has a story (via @LewSOS) on the Wellington mayoral elections.

John Morrison is leading incumbent Celia Wade-Brown in the race for the Wellington mayoralty, according to a poll of Dominion Post readers.

Mr Morrison, who has been a city councillor for the past 15 years, had support from 27 per cent of the 635 readers surveyed last week – while Ms Wade-Brown trails on 17 per cent.

They don’t say how the survey was done — it’s not clear how you would get a representative sample of Dominion Post readers.  For all we can tell, it might just be a bogus poll.  It’s also not clear, on this topic, why Dominion Post readers are even the population you would want, since the story continues

Of those surveyed, 275 were eligible to vote in the Wellington City Council elections.

You’d at least expect that the voting preferences would be broken out by eligibility.

Sadly, no.

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