Those of you who have been with me from the beginning will have noticed that there’s been a few changes to the Voter Choice Project. Mainly, instead of studying three elections, we are now only doing one – the federal election.
To explain this change a little of the background of the Voter Choice Project (VCP) may be useful.
Planning for this study began in April 2016, after more than a year of searching for supervisors willing to come on the ride with me. We don’t study voter behaviour much at all in Australia, and no one has used the theory and methodology that I’m using, so it was always going to be a tough ask finding the right uni and supervisors – if you are familiar with the university world you know such a search is not unusual.
When I was shopping this study to almost every uni in Sydney (and a few beyond), I proposed studying a single electorate in the federal election in detail, having an open online panel where you too could play at home as a control panel, and doing a pilot study on the Queensland election. In early discussions with my supervisors we added the Victorian election into the mix as a kind of super-pilot – so a very small pilot in Queensland in late 2017 or early 2018, adjust and test the model in Victoria with its set date election in November 2018, and then run the full study on the federal, which at that point would have been due in the second half of 2019. That’s huge for a PhD, but the elections were well spaced and it was do-able (it’s certainly no harder than running a national political party…).
In early April 2016 when UTS said they’d let me apply, the double dissolution still seemed like the wishful thinking of those who like a good political drama. And even when it happened, it took me until a few weeks afterwards for the consequence to sink in. The window for the federal election had moved forward a full year, and was now a 10 month window from August 2018 to May 2019, rather than just a few months in late 2019.
The risk for a study like the VCP with such a huge variance of the election date is simply enormous. The problem with panel studies is attrition – people dropping out of the study. I was now facing a situation where people might be in the panel for 3 months – less than desired, but ok – or a full year. Attrition over a full year may be so great that I have to junk the study (or at least the quant part of it). That might give you an insight as to why a study like this has never been done before.
At the same time, Queensland became jumpy, with talk about them going to the polls a year early. There was no way I could get a pilot study designed and through ethics in time for such mishigas.
So in mid 2017 we changed it. We added the NSW election to the Victorian election, the two together accounting for half the nation’s voters, made that the main study, and scaled down the federal to just a tiny complementary study on time of vote decision. Still absolutely enormous for a PhD, but no surprises were expected in either state election, they are both fixed date, only a tiny overlap in the questionnaire schedules… the federal arena can do its thing, it won’t affect the thesis, but we’ll collect the full data set while we’re going for a lovely goldmine of post-doctoral work. Plan in place, work hard, insert ‘you can doo eet’ type affirmations here.
To say that politics has been volatile in Australia in recent times is one phenomenal understatement.
That volatility has spread to NSW and Victoria, and they aren’t looking like they will be the stable predictable elections they should have been.
I watch a number of indicators that voters might be getting spooked – scandal, controversy, signs a leader might dumped, stuff like that, and particularly how that kind of information is received and resonates on radio talkback, Twitter, Facebook groups and letters to the editor. (I don’t look at state wide 2PP polling, it’s irrelevant, particularly this far away from the election.) I’ve been saying for sometime that the Liberals don’t have a lock on NSW with multiple major policy backflips and the local government forced amalgamation anger yet to wash through, and all Labor really needs to do is switch from Luke Foley to a leader who has a chance in the preferred premier stakes and we could have an interesting contest.
But Victoria was supposed to be a veritable coronation with a generally happy and unengaged electorate – until Labor gets busted for using tax payer dollars for campaigning. And then the Libs pull some strange stunt on an Easter holiday sitting. Wait, WHY were they sitting on Good Friday? There’s a number of other grumbles, and the federal Liberal contingent seems very determined to muck with Victorian state affairs, from the various crime commentary to Turnbull announcing money for rail infrastructure without so much as a quick phone call to the Premier before he told media, but you get the idea.
Voters don’t like volatility. They start paying attention, and tend to get angry. Angry voters yack at the less engaged citizens, who just gather that something is wrong – and if enough people yack at them, they too start paying attention. It’s a politically deadly cycle.
I can study one volatile election, but not three. Particularly given the heat and intensity of said volatility; a whole lot of claims, counter claims, fake news this, witch hunt that… I’m ordering popcorn by the carton for the likely “is this candidate eligible” section 44 tomfoolery.
The start for Victoria had already been delayed and the entire VCP was being reviewed anyway (due to a change in one of my supervisors and some other stuff that’s not at all interesting), so it was an easy decision to switch it back to just being a federal study. Those who have signed up to a state study will get an email in the next day or so about whether they want to be deleted or switched to the federal study.
If there are by-elections, we may run a 3-wave pilot on those, but no promises. Other than that, there won’t be any more changes, because it starts in 6 weeks.
It will be fun! Like having a long conversation with several thousand voters about their election experience and listening to them tell me what they think of it all. However, to properly execute the study I need lots and lots of people to sign up and participate from every corner of the country, and stick with me until the end!
Thanks for joining me for the adventure of the Voter Choice Project. I’ll do my best keep this blog updated with project news and posting things of interest for fellow election nerds.