Rubbery Poll Interpretation – I blame the GFC. Check Centrebet next time

There has been much ado about the interpretation of Galaxy pre-election opinion poll on the Queensland election which predicted a win for the LNP and the actual poll result.

The folowing should be noted

  • No margin of error was not reported.
  • The method of sampling was not reported.
  • The reported sample size was 800.
  • No break down of the sample in age or urban/rural reported
  • The electorate of Queensland is bigger than the city of Melbourne and is spread over a couple of multiples of size of the state of Victoria.
  • Brisbane is closer to Melbourne than Townsville.
    Update

Bob on Unleashed at abc.net.au

There are many ways a poll, if manipulated, can get you a false, convenient result. You can take the poll on a holiday weekend when many younger, and therefore Labor-leaning voters, or Green-leaning voters, aren’t at home. You can take the poll in early January when (as always) the Labor vote and the Green vote will go dramatically down for this reason, when many, many voters are in a caravan park or a holiday motel, or overseas. You can ring up on Friday night when only the old, the ill, the poor, the loveless and the mad are at home, these being more likely Liberal voters or National voters or One Nation voters or Independents. You can ring on Saturday morning when the prosperous are out shopping. You get more Liberal voters at home that way, saying Lawrence Springborg for me.

This is the way, I suggest, that Newspoll gives its principal client Rupert Murdoch what he wants and why Newspoll is widely known as “the Bill O’Reilly of statistics”.

Newspoll is very careful to go where the ambiguity is, and show no-one its raw figures. Why otherwise its raw figures should be secret is hard to say. These figures are “adjusted”, we are told, “to reflect population distribution”. I’ll bet they are.

Newspoll, moreover, rings no mobile phones, just those people who happen to be home on Friday nights and not, say, at the pub. Just those people who possess home phones (many young people don’t any more) and don’t mind giving up half an hour of their evening when they might be cooking dinner. Those who do mind are listed as “refused”. Only those who speak English, and speak English well, are persisted with, I am told; otherwise, I am told, Newspoll hangs up, and lists them as “refused”.

Advertisements

One Response

  1. I am pretty down on the poll stories pre-election. Not for the actual poll results themselves, but (a) their usefulness for making predictions, which is (as we can see) enormously overstated, and (b) the commentators’ groupthink that led to them adopting one interpretation and sticking with it.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: