Monday, 3 May 2010

Behavioural politics, day 24 of 30

The leaders' debate tonight showed the immense power of the phenomenon of confirmation bias.

The debate was interesting but neutral in terms of campaign impact - in my judgement, the candidates performed about as well as each other. And yet watching it on twitter was very informative yet again, primarily because of the reactions when the post-debate polls were released.

Nearly all the polls showed Cameron as the "winner", with Clegg and Brown in second and third place or vice versa. A typical poll (see writeup here) gave:
Cameron 35; Clegg 33; Brown 26
But when you look at the voting intentions of the people who answered, what do you get?
Conservative 35; Lib Dem 36; Labour 24
Spot anything? The numbers are almost identical.

When there are no clear distinctions between the quality of the performances, people who already agree with David Cameron will think he's talking sense; Labour supporters will think Brown is the only one who isn't lying; and those who support Nick will agree with Nick.

So it's no wonder that Cameron, who's in the lead in most of the regular polls, will come out as having won the debates. This is more true the further we go into the campaign.

But there are still some undecided voters, and the Angus Reid poll splits them out. About 37% went for Clegg, 25% for Cameron and 22% for Brown. Even this doesn't tell us that much, because undecided voters are likely to have a specific profile, and I'm not surprised that they would be pro-third-party.

A revealing detail is that while Tory partisans clearly chose Cameron as top performer, Labour and Lib Dem supporters were more likely to "swap", picking Clegg or Brown respectively as the winner. This is likely to indicate that the parties are drawing from a common base of shared support. Tory and Lib Dem supporters have a smaller swap tendency, showing that there are some swing voters between those two parties; while only 4.5% of Labour supporters thought Cameron best, and only 3.5% of Tories rated Brown.

Labour and the Lib Dems were also more likely to think nobody won; I wonder if that might indicate weaker support of their party, or a lower likelihood of voting.

Given all this, it's surprising to hear today that a third of voters are undecided (I think what Doug Alexander means by this is that a third have said they might still change their minds). This debate hasn't changed much in itself, but the timing of it has managed three important things:

  • moved the news cycle on from yesterday's "bigot" news (the intensity of yesterday's coverage, and Brown's instant apology, also made it easier to move on today). This is a critical save for Labour, as they came close to annihilation yesterday, so they get the best ratings today.
  • sustained the Lib Dems' momentum for yet one more day, making it just that bit more believable.
  • helped the Tories firm up their own momentum, which will become more important over the next week.
Ratings: Labour 7/10, Conservative 6/10, Liberal Democrats 6/10.

1 comment:

JLTan said...

It is a pity.

Many of my British friends here are willing to vote, but are finding it too difficult.