The point I discussed this morning was that, when purely random choices are asked for, it is better to be in the middle of the ballot paper rather than at the top or bottom.
As Mark Pack pointed out, research in the past has suggested that it is better to be first on the paper. I raised this with Andy and he said that there is some evidence for this from studies of real elections in America where the order of candidates on the paper is randomised rather than alphabetic.
This is because we use different strategies to decide how to vote when the choice is random and when it is a real election. As you would expect, the more informed someone's choice is, the less the order of the candidates on the paper matters.
But the reason I am returning to this research is that, having held further mock elections with the names of real parties on the ballot paper, Andy made the suggestion that it is a greater advantage for a Liberal Democrat to be first than it is for a Labour or Conservative candidate.
He suggested this is because it reminds the voter of our existence before he comes across the candidates of the other two main parties. I seem to recall that research on opinion poll methodology has shown that we do better in polls where all the parties are named before the question about who people will vote for is put.
This is early research and Andy is not claiming any deep statistical significance for this finding yet, but it is an interesting trend.
I asked if the same thing holds true for Green candidates, and Andy said it did but to a lesser extent. Perhaps Green voters are more motivated to begin with?
As I said, it is early days for this research and these findings are tentative, but I shall keep my eyes open for further developments.
No comments:
Post a Comment