August 12, 2013 9:00am by David Farrar

The Campbell Live GCSB poll

Campbell Live is trying to run what they call the biggest opinion poll in NZ history.

That neutral observer on such issues, is promoting it of course:

However expert pollster Grumpollie points out that the so called poll is meaningless in terms of accuracy:

Here’s a really important point for anyone wanting to judge the accuracy of this poll –big numbers do not provide representative data!

Here’s why…

This poll commits the ultimate sin of survey research – it uses a self-selecting sample.

People choose to take part based on the topic. This means that the poll only represents the New Zealanders who feel strongly enough about the GCSB bill to take part in the poll.

He points out that having a self-selecting sample, rather than a random sample, makes a huge difference. For example all the polls on same sex marriage with random samples showed a majority or plurality in favour. But the Campbell Live text in poll found only 22% in favour and 78% against. A result that has no scientific usefulness. Note that particular poll also had a huge number of responses. What counts is whether the sample is random – now how many people take part.

But the problems are even worse than that for the Campbell Live so called poll. Thomas Lumley at Stats Chat points out that to vote through their website you need to give Campbell Live your name, e-mail address and postal address. He points out:

Wouldn’t you expect that people unhappy with the prospect of increased (legal) surveillance of New Zealanders might be less willing to give all their personal details with their vote?

But you can understand why a Mediaworks show goes with an unscientific poll than a scientific poll. A scientific poll actually costs money to do. But those who use the text option for their bogus polls end up paying money to Mediaworks, so Mediaworks makes money out of the bogus poll.

Comments (47)

Login to comment or vote

Add a Comment