Monday 18 January 2010

Vowles on the 2008 NZ General Election

Liberation points to a chapter by Jack Vowles on the 2008 New Zealand election.
In the 2008 general election, half of voters (51%) thought there were only ‘minor differences’ between the parties during the campaign, while only 38% thought there were actually major differences between the parties. Furthermore, when survey respondents were asked to place the parties on the left-right spectrum, ‘A third could not place Labour or National’. These findings from the New Zealand Election Survey surely reflects the policy convergence of the parties, and are detailed in Jack Vowles’ new academic chapter about the election.
But in the 2005 NZES, 18.9% of respondents said "don't know" when asked National's ideology (with a further 6.7% leaving the question blank); similar proportions had problems identifying Labour's ideology. 23% of respondents either answered "don't know" or left the question blank for both of National and Labour. A further 11 percent of the sample placed National to the left of Labour (among those who did not answer either "don't know" or leave a blank for either National or Labour). So, a third of the 2005 sample could not place National relative to Labour. I wonder what proportion of the 2008 respondents placed National left of Labour....

He goes on to argue that the shifting of the parties to the center opened up room for valence issues. I certainly agree that 2008 was far less ideologically charged than 2005. For 2008, Vowles reports (on a left-right spectrum where 0 is left and 10 is right) that voters placed Labour at 3.7 and National at 6.7. But those registering an answer in 2005 didn't answer that much differently: Labour scored a 3.6 and National a 7.1. So the average gap closed from 3.5 points to 3 points. But if we look at individual respondents' reported distance between National and Labour in 2005 (for those registering an answer), the standard deviation of that difference variable is 4.8. Is a half point closing of the ideological gap really then significant?

No comments:

Post a Comment