With each election comes the promise of change. Voters can throw a government out, install a new one, or give those in power a new mandate to tackle the issues the country faces. Millions of dollars are spent, names are dragged through the mud, and, in the end, about four out of every five people we elected last time are sent back to Ottawa again.
The rest of the article can be read on The Globe and Mail website. The Hill Times study mentioned in the article can be found here.
I've been getting used to using the resources on the Elections Canada website, and have gotten quite fond of them. This is an incredibly useful site managed by the Parliament of Canada, and has electoral results stretching back to 1867.
I had approached this article with the intent of showing that veteran MPs are harder to knock off than sophomores. It seemed like an obvious hypothesis. But my analysis showed that there really isn't much difference between a four-term MP and a one-term MP.
You might be wondering why I didn't do a complete study of all 308 ridings. Time constraint is the reason, but my sample was representative. The re-election rate of incumbents in the 192 ridings I analyzed for 2006 and 2008 was only a tiny bit higher than the actual rate (89.1% to the actual 87.4%).
While looking into this, I also tracked other factors related to incumbency. I'll have to finish studying the ridings that remain and crunch some numbers, but it appears that in the vast majority of cases an incumbent's rate of increase or decrease in vote share from one election to the next is less than that of the incumbent's party in the region as a whole. If this turns out to be the case, it will be useful for my projections. The test I am running on the new projection model is good, out-performing other methods of projecting, but seems to be hung-up on over-achieving incumbents. With the results of this analysis I think I will be able to achieve a higher rate of accuracy. More to come when I have completed my tests.