Polls apart from reality

Peter Hatemi, an associate professor at Pennsylvania State University, claims that genetics “account for a substantial proportion of individual differences in political traits”. Hatemi’s professorship encompasses the unlikely mixture of political science, microbiology and biochemistry. Some might unkindly argue that he has a vested interest in discovering a link between politics and chemistry to underpin his own hybrid expertise. Perish the thought.

In any event, I remain fonder, if still a little sceptical, of an earlier theory which came out of the University College London in late 2010. A study of 90 students’ brains uncovered a correlation between political leaning and the size of part of the brain. In particular, the amygdala was apparently more pronounced in those on the right of the political spectrum. At the time, I put forward a rigorous explanation for this theory based in evolutionary science (“Evolution explained”, Quadrant Online 2, January 2011). Perplexingly, this hasn’t been picked up by the scientific community or the mainstream media. My explanation therefore languishes unheralded; waiting to be discovered.

In the meantime, the recent fluctuating opinion polls in Australia cry out for a scientific explanation. Can they be explained by unstable human genomes in large numbers of individuals; or, alternatively, by frequent inflations and deflations of the amygdala in those same individuals. Is any of this painful might also be a compassionate question to ask?

I will concentrate on the Newspoll, though the Nielson poll could equally be used. On 27 March it was reported that only 28% of voters would give their first preference vote to the Labor Party compared with 38% at the federal election in August 2010. This meant that roughly 1.3 million voters had deserted the Labor Party. You might think that this strikes at the very heart of either scientific theory of political allegiance. On the other hand, the sheer weight of incompetence and duplicity on the part of the Gillard government may have overwhelmed genetic and/or amygdala tendencies. After all, as Slipper, Windsor, and Oakeshott have taught us, political allegiance is susceptible to naked self-interest and opportunism. Why not also to deep disenchantment?

The difficulty comes with the latest poll showing that the Labor vote by 21 August had rebounded to 35 per cent. Over 900, 000 of the 1.3 million disenchanted voters had switched back again. What happened in the interim? What happened was more of the same. You will recall Julia Gillard lying about Bob Carr’s appointment only a little before the March poll. We have since had the implementation of the iniquitous carbon tax, the back flip on the boats, and the unbelievable spin on why Slater & Gordon parted company with Gillard. There is nothing there to change rational minds; so what happened to the 900,000. Were they visited in the night by Bill Shorten and some union heavies?

There would appear to be two plausible explanations. One is that the 900,000 voters suffered a temporary aberrant shift from their true nature and the August poll signifies that their genetic or amygdala disposition has reasserted itself. The other is that the polls are completely unreliable – at least within a wide margin. Call me anti-scientific; I plump for the latter.

The Newspoll samples only around 1100 voters by telephone. The Nielson poll a few more. The Newspoll quotes a margin of error of (+ or –) 3 percentage points. This, of course, might mean that the March poll was really 31 per cent not 28 per cent and the August poll 32 per cent not 35 per cent. In other words, maybe few voters actually changed their minds between the two polls.

But it might be more complex than that. Numbers of people may give answers to pollsters that they think they should in politically-charged circumstances. It was probably never the case that only 28 or even 31 per cent of voters would give their first preference to Labor. The recent Wisconsin recall election is instructive. Pre-election polling showed it would be tight. Governor Scott Walker romped in. Why? Because numbers of people disguised their voting intentions. The political climate in Australia has quietened since Rudd challenged Gillard in late February. Some of those rusted-on Labor voters, embarrassed into giving misleading responses earlier in the piece, now perhaps feel more relaxed about nailing their colours to the mast.

What does this mean? It means, in all likelihood, that the Coalition is more or less in the same winning position it has been since Gillard did her dirty and disreputable deals with the Greens.

Peter Smith, a frequent Quadrant Online contributor, is the author of Bad Economics