Volatility in British poltics
Peter C challanged me a few weeks back on my statement that the concept of 'record swing' was less important as political trends had got more volatile. He noted that parties seemed to retain power for longer now than they used to. I'm still not backing down, but my initial attempt to prove this hasn't worked. Using this excellent resource of Guardian ICM polls back to 1984, and filling in a few monthly gaps by interpolation (only about four polls in the mid 1980s and strangely February this year) the following chart shows the average monthly change in the Conservative share of the vote over the past year. I've used absolute values, so if the Conservatives over the past year went up 1%, down 1%, up 1%, down 1% and so on, that would average at 1%, not 0%.Anyway in the year to Feb 2009 the Conservative vote share has been more volatile on this measure than at any time since May 1989, but on the other hand in Feb 2006 and then before in Feb 02 it was less volatile than any other period, and if there is a trend it probably is towards less volatility (dividing the % change by their vote share doesn't make much of a difference, by the way). I suppose this is partly because a lot of this period they have been down to their core support, but that of course is not helpful to my theory either.
[Larger image here]
posted by Matthew @ 6/22/2009 07:28:00 AM
Comments: 1
