Wednesday 1 January 2014

Roy Morgan Voting Intention 2012-2013 with Political Events

Before the 2014 election year gets really started, I thought it would be interesting to look at how the polls have changed over the two years since the last election in 2011. The chart shows polling data from Roy Morgan, who poll around 800-1000 electors roughly every fortnight, and a selection of political events/issues. The event/issue is marked (roughly) at the date when it was first reported in the news, as some issues will stay in the media for several weeks. The issues are also somewhat abbreviated; if I hadn't the word 'scandal' would appear far too often. Obviously it would be impossible to fit every single issue/event on the chart, so it is a selection of those that attracted the most media attention. I am sure that I have missed some important ones. Everyone is welcome to draw their own conclusions from the chart; mine are included below.

My take on this? 
Some issues affect polling and voting intention, some issues don't. The media would have us believe that every issue is the end of the world for either the government or the opposition, but a lot of the time people just aren't bothered or don't care (enough). Note that it is important to look at all the lines, not just the blue and red ones as many people often do.  Of course, the sample size is (in my opinion) rather small and very susceptible to all sorts of sampling errors (Roy Morgan assumes a simple random sample with an error of around 1.5%). Also, it would be fallacious to assume that voting intention is influenced by one political event/issue at a time; however, it is also true that one issue can have a non-negligible impact. The sample size is primarily made up of two groups: those who are ideologically deep-seated with a particular party and are unlikely to change their vote regardless of what the party does, and those who are "undecided" or "swing voters" who are influenced by the issues of the day. Hence, we should see some correlation between media coverage of political issues/events and voting intention as the "undecideds" get influenced.

So what does matter? 
Taking a very very simplistic view/analysis, voting intention is heavily influenced by the public's perception of individual people. I feel that individual people should have a small impact on voting intention in comparison to policy, but that appears to not be the case. The reason that so many individual names appear on the chart above is because whenever an individual falls from grace, is embroiled in some scandal, or is just outright revealed to be a bad person, the media jumps on them. An individual is a much easier target than an entire party, and hence Nick Smith, John Banks, Brendon Horan, Richard Prosser, Aaron Gilmore, Peter Dunne, David Shearer have each had a noticeable impact on the polls. The public needs to have confidence in their elected representatives, and hence those who are generally undecided are likely to move away from whoever they have least confidence in at the moment to the least bad alternative.

But government should be about policy?
That is certainly my belief. The culture of celebrity surrounding our politicians frustrates me. Ideally, the best solutions are selected objectively and executed flawlessly. Unfortunately, humans have to come into the equation somewhere, and therein lies many of our problems. We can't have democratic government without humans, and we can't have humans without their flaws. So every time an MP says something stupid, has a lapse of judgement, or makes a grievous mistake, that problem takes precedence, and the real problems facing the constituents get buried in the avalanche of flashing cameras and fluffy microphones. Sometimes this can be very entertaining, but entertaining doesn't get things done. Ultimately it's still about making sure that the right people are doing the right jobs. After all, without competent humans, you could argue and win a particular policy point, but still end up botching the execution.

So what does this mean?
Political events/issues involving individuals appear to have a disproportionately larger effect on voting intention than policy changes, which usually have a much wider real impact. In an ideal world, policy would be all that mattered. Unfortunately, we don't live in an ideal world.

I'm sure many will disagree with my analysis. There aren't many obvious, objective relationships appearing from the chart so that is to be expected. Feel free to comment below with your opinions and analyses; just remember to be nice.

No comments:

Post a Comment