Monday, May 6, 2019

Two polls

This morning we are blessed with two new polls, both of which tell much the same story we have seen since the start of this election campaign: Almost every poll in the last couple of months has been this same 48/49 for the Coalition to 51/52 for Labor.

2019-05-03 Newspoll 38.0 36.0 9.0 5.0 12.0 49.0 51.0
2019-05-02 Ipsos 36.0 33.0 14.0 5.0 12.0 48.0 52.0

I must admit that the under-dispersion of the recent polls troubles me a little. If the polls were normally distributed, I would expect to see poll results outside of this one-point spread for each side. Because there is under-dispersion, I have wondered about the likelihood of a polling failure (in either direction). Has the under-dispersion come about randomly (unlikely but not impossible). Or is it an artefact of some process, such as online polling? Herding? Pollster self-censorship? Or some other process I have not identified?

Leaving aside my troubled thoughts, the Bayesian aggregation of the polls follows. Here we have the Coalition on 48.2 per cent of the national two-party preferred (TPP) vote share. Labor is on 51.8 per cent. With these numbers, you would expect Labor to win the Federal election on 18 May.

Turning to the ensemble of moving averages, we have pretty much the same story.

In the primary votes, we see more differences. As usual, Ipsos has the Greens well ahead of any other pollster, at the expense of Labor.

No comments:

Post a Comment