Well we had one clear month without a by-election… the September wave of the Voter Choice Project returned us again to multiple versions of the survey, with a Wentworth by-election study now underway. The Wentworth first survey is open until the end of the month (patience…) and Wentworth voters have been removed from the numbers in our September results.
We’ve also made another change to our numbers (the fun part of being an academic study is you can tinker with things to get it to work better). We continue to have a problem with getting Liberal Party voters into the main panel, but they’re just as willing as any other voter to do the short surveys. So whereas previously short survey respondents were excluded from polling numbers, those who gave us their email address – and thus a manner in which to identify them as a unique respondent – were included to see if that would result in a more balanced sample. It did. The weighting on age, gender and 2016 vote is still in place to get the responses more reflective of the broader population, and we will continue to recruit (beg) Liberal voters to take part. Sign up here if you’d like to join the main panel.
We’re singing the same song as most other pollsters: Labor is now clearly ahead, and the change in PM sent many scattering to protest votes but also brought others back off the bench to kinda cancel each other out. Ok, there’s a very large hit to the Coalition primary vote, but many of those are inidicating they are not committed to that decision and they still want the Coalition to win government, so they’re likely to come back. Remember, our major party primary votes are always lower as we don’t exclude the informals, don’t knows, and have more minor parties in the pick list (and our panel are generally an entertaining collection of unhappy voters), but Essential also had the Greens and Independents/others jumping up, while Centre Alliance and PHON went down, in their figures released yesterday; Others also jumped up in Newspoll released Monday so – while avoiding group think – I have high confidence in the sample.
As those who have been following us know, the point of the VCP is to study change in vote decision, and reasons for those decisions. The vote retention heat map again shows a great deal of instability in the vote from August to September (panel members only). Read along the rows to see where voters have moved to, and down the columns to see where votes have been gained from. Click on the image for a bigger version.
No, I don’t know what happened in PHON, and why their left leaning supporters are leaving. I missed something somewhere, perhaps deep inside the web of Facebook groups…
This month we asked panel members about their political activities, party involvement and the like. This information is mostly useful in the broader context of the study rather than producing interesting numbers. However, one of the questions was which party do voters normally support, regardless of who they intend to vote for at this election. Combining this information with current vote intention again shows the volatility and movement in the electorate.
We also asked again about the change in Prime Minister. This was to test the findings of the snap poll we did in spill week to see what was going on now that things have settled in a bit. We asked whether it was the right decision – remember that these results exclude Wentworth voters.
Malcolm good, Liberal Party bad, and a lot of consistency in the response to those questions.
We also asked what level of confidence people had in Scott Morrison as Prime Minister and Josh Frydenberg as Treasurer.
Well Frydenberg is getting a chance, but Scomo has work to do.
We asked an identical question to the short survey, which is whether the change in PM inpacts your vote. For convenience, both results are presented here.
This would indicate that as the shock wore off and the reality set in, things have gotten worse, not better in terms of instability. Particularly note the jump from 19 to 26% reconsidering who to vote for, and the slight drop in people who have locked in their vote.
Lets see what October brings.
In the change of PM first short survey there was a very clear message from the comments that people wanted an election now. We asked the question directly, and the result is not surprising. (Unless your surprised at the 10% who don’t care.)
Results of the Greens short survey will be posted separately once I’ve figured out how to code all this colourful language.