03/09/2010: "polls - how we are conned"

a classic example today, posted in the times newspaper about the gap in support between conservatives and labour is narrowing.

the headline shows neck and neck in marginal seats between tory and labour
now on first read, it gives the clear impression that the gap is closing, which is a fair point as people make their minds up as the election on may 6th gets closer.

on further inspection however, the "story" starts to fall apart:

the 100 seats/areas they used for the polling in DOES NOT include the top 50 seats, where the majority for the sitting labour MP is only very small - why not?
id have thought that this would give a clearer picture? why did they ignore these seats?

the poll also doesnt include seats where the liberal democrats are the sitting MP with a small majority, again, why not?

classic example of mis-leading the public
nice story, no doubt full of opinion and counter comment, nice headline that would attract many with a political interest, factual basis is very poor.

theres a warning here for pollsters and the government - believe your own polls at your peril!
many times in the past, come elections, the polls and estimates have been way off the mark.
it should be an interesting may 6th!