KORE Poll 1: Trust and COVID-19

Thanks to everyone who participated in the first KORE Poll on Trust and COVID-19. A special thanks to the many Voter Choice Project panel participants who have decided to continue in the KORE Panel.

Trust

Questions were asked about trust in media, social media, and politicians. Generally, people trusted media and social media sometimes, and more often than they trusted politicians, but a majority thought it was important that they could trust politicians.

There weren’t a lot of surprises here.  The ABC was the most trusted news outlet, and Penny Wong was the most trusted politician with nominations from across the political spectrum. The two interesting numbers from this set were comparing trust of what people see on social media to trust in politicians. 62% usually or sometimes trusted what they saw on social media and being good information, while only 45% usually or sometimes trusted politicians.

That 54% of people rarely or never trust politicians is also interesting in light of the 83% of people who agree it is important to be able to trust politicians.

COVID-19

Respondents were asked a series of questions about the COVID-19 pandemic.

There aren’t many surprising results here either. People are scared.

When asked what comes to mind when they think about COVID-19, it’s mostly lockdowns, vaccines, the impact it’s having on people, and what the governments are doing.

Voter’s assessment of performance in managing COVID-19 in very depressing reading for Federal or NSW Government advisors, with only 18% saying the Federal Government have been good or very good in their handling of the pandemic. Only 23% of NSW voters rated the NSW government as good or very good. All other states and the ACT have over 50% thumbs up, although it is worth noting that Victoria and Queensland are highly polarised with a large percentage rating their State as very poor.

The sample size is not big enough to effectively do a geographic breakdown of the very good/very poor ratings in Queensland or Victoria, but the sentiments do seem to be most intense and polarised in the border regions.

At the last minute we threw in a question to see if people agreed with the shift in language from Morrison and Berejiklian that we needed to live with COVID-19. This was really interesting.

For those who aren’t familiar with the work of the Voter Choice Project, the methodology we use is a fused method where qualitative and quantitative questions are asked on the same topic in the same survey, and at the bottom of each page we put a text box so people can write whatever they like. This model gives a rich insight and helps with interpretation of the quantitative figures. This kind of question is where our fused model really shows its value… because if you looked at those figures, you’d think Morrison and Berejiklian were on a winner, right?

Wrong.

A huge number of people that agreed we have to learn to live with COVID commented, many furious, saying that we have to learn to live with it because Morrison and Berejiklian stuffed up. Many of the ‘other’ responses were along the same lines, we should be able to achieve elimination, but it cannot happen now.

By ignoring expert warnings and advice the totally unqualified politicians have welcomed an avoidable disaster
– Female voter, SA, 34-45

Weak leadership from Prime Minister. He lacks vision and has no plan to get us out of our hermit nation status.
– Male voter, Qld, 45-54

Start governing the country instead of gearing up for an election (Feds) and appeasing your supreme leader and cabinet (NSW)
– Female voter, NSW, 45-54

The country would be in a much better position without the pathetic leadership of the PM and NSW Premier
– Male voter, Tas, 45-54

We could have had elimination, but the federal and NSW governments incompetence stuffed it.
– Male voter, NSW, 55-64

It’s a disappointment. Feds have screwed it up royally and the focus on reopening and staying open led to delta blowing up.
– Male voter, NSW, 35-44

The Federal Government and NSW Government has completely bungled their handling of the pandemic
– Male voter, Qld, 25-34

In news that might make Morrison smile a little more than all of that, this election won’t be entirely about COVID-19. Only 4% of voters say it is the only thing that they will decide their vote on, and a third say it will be a significant factor. That leaves over half of voters saying COVID-19 management either won’t be a factor, or they will be considering other things as well.

Vote Intention

In the future we’ll have a full panel of vote intention numbers each month, as obviously that’s the real point of the exercise. However, it is evident that COVID-19 has shifted the baseline of sentiment towards voting in some fundamental ways, and the modelling and complex weighting used during the last election needs recalibrating, which will take a couple of polls to sort out.

So, we’re just giving you a taste of the different kinds of numbers we intend to produce so you can get an idea of what’s planned. These are only weighted by age and gender.

Effective vote is generated using the Hv6 (Hypothetical vote 6) question developed during the Voter Choice Project where respondents are asked to drag and drop 6 hypothetical candidates – Labor, Liberal, Green, Moderate Independent, Right Wing Minor Party and Left Wing Minor Party – into their preference order. Then the effective vote, that is where the vote will end up after the distribution of preferences, is calculated, including allowing for Greens or Independent/Other to be in the final two in those electorates that were not ALP v L/NP contests in the 2019 election. For example, in addition to the 6 seats held by independents and minor parties, Maranoa is LNP v PHON, New England is NAT v IND, Wills and Grayndler are ALP v Green, and so on – a 2PP figure is meaningless in those seats, you need a two candidate preferred (2CP) count. Thus 2PP is meaningless in trying to figure out what is going to happen nationally because you’re ignoring 10% of electorates, we need something else – this is one of the things we’re trying.

The utility of the Effective Vote calculation is yet to be proven, but the idea is that this may be a better indicator of what the likely seat share will be than two party preferred. Think of it like you would quotas for Senate vote share, with about 1.5% of the vote equalling a quota for a seat in the House. So in this example, it would equate to roughly 72 ALP seats, 65 Coalition seats, 9 on the cross bench and 5 seats worth of undecided votes still up for grabs. Note though, this has only been weighted by simple weights, so just take this as an indicator of what kind of figures to expect going forward, not a predictor of what will happen…

The Incumbent v Challenger model predicted from as early as January in 2019 that the Coalition would be returned – and I didn’t believe it. (I didn’t believe the Momentum Tracker either, we’ll talk about that next month.) This is pretty simple really; the effective preference vote is coded against who currently holds the seat the voter indicates they live in. If they’re voting for the party that currently holds the seat, then they’re coded incumbent, if they’re voting for anyone else on the ballot then they are coded challenger. If challenger figure is over 50% then the Government should change hands.

It’s close.

For those wondering, it’s 50.21 to 49.79 if you go down to the decimals.

We won’t be producing a 2PP. Get used to it.

KORE Poll 2 will be out in a couple of weeks; we’ll be testing the major issues of the last election to see if people still care about them. Join the KORE Panel here to take part.

---

Notes

KORE Poll 1 surveyed 1153 voters across Australia on their vote intention, trust of media and politicians, and what they think about the COVID-19 management strategy. It was conducted online between August 24 and 27, 2021, with participants drawn in equal measure from the Voter Choice Project panel – going forward the KORE Panel, and fresh participants recruited by social media advertising.

The comments in this blog post are for the benefit of participants and interested observers. They do not include the results of all questions and are not a formal report. Journalists wanting more detail can email team@korecsr.com.

Wait – what’s the Voter Choice Project?

The Voter Choice Project was my (Raphaella Kathryn Crosby) PhD research. It involved a large panel completing up to 16 surveys from June 2018 to June 2019 in order to track how people decided who to vote for and what was influencing those decisions.

The study involved experimentation with a number of new polling methods, in particular the Momentum Tracker – an interlocking series of questions to create a score for how soft or hard someone’s vote intention is, and alternative models to the misleading 2PP that most media coverage and online chatter obsesses on, the leading candidate being Incumbent v Challenger as a proxy for whether the Government was likely to change. In addition, a range of sampling and weighting variations were trialled with mixed outcomes.

This experimental work to find better polling models is incomplete, so, the KORE Poll picks up where the Voter Choice Project ended.

All members of the Voter Choice Project have been emailed and invited to join the KORE Panel. If you didn’t get your invite or are a new person who would like to join the panel, sign up here.