Once again, Gordon shows worrying statisitcal ignorance

May 02, 2005 15:13

Following on from their obvious recent error (a vote for the Lib Dems is not a vote for the Conservatives by the back door. TWO votes, maybe)- New Labour have come up with the statistic that "If one in 10 Labour voters don't vote, the Tories win". Unveiled by Tony and Gordon fresh today.

Now hang on just a second.

Picking holes in election statements could all too easily take up all of anyone's time, but I can't help wondering where they got this from.

1) What is our starting point?

*2001? (a) people have switched views already and (b)Demographics- Tory voters are significantly older (albeit richer) on average- and overall have higher mortality as a result. Hence, you actually need some drift from Labour to Tory (over the course of a lifetime) for the figures to remain stable.
*Opinion Polls? a) which ones, b) how far back in time can you go and still claim that they're credible, c) what counts as a "Labour voter", d) You're basing this on a sample in any case. Tch.

2)What are we counting as a win here? most votes? or most seats? For example, if the 1 in 10 all live in the South East of England then no-one will care, or even notice. If the Conservatives gain the largest number of seats but can't form the next government, is that a win?

If Gordon needs someone to explain the hard stuff for him in between looking after his young family and taking over as PM, you know where to call.

PS I hate German Kezboards. At least Iäm not trzing to tzpe a treatise on the use of the zyzygy.

PPS If you vote postally and then exit the voting population (eg die or are convicted of voting fraud) before the election date, does your vote still count?
Previous post Next post
Up