I've said it before, and i'll say it again...

Our newspapers really have to stop distorting the opinion polls they commission.

Take today's Guardian and its new ICM poll for example.

The basic voting intention figures were: Conservatives - 42 percent; Labour - 30 percent; and the Lib Dems on 18 percent.

The Guardian reported that the Tories were down 2, as were Labour, and the Lib Dems were up 2.

Given how several recent polls (3 out of 5) had shown a swing to Labour from the Lib Dems, one might have concluded from reading the Guardian that this swing had continued.

But this was not what happened.

In fact, the Tories were up 2, so were Labour, and the Lib Dems were actually down 4!

So what happened? The Guardian, like all the other newspapers, compared their new poll with the last time they commissioned a poll, i.e. a month ago. But in the intervening period, there had been a further ICM poll - hence the disparity.

Newspaper editors really need to stop distorting the opinion poll situation. It is for this reason that when a new poll is released i ignore the newspaper reports and get my polling news from the excellent UK Polling Report and PoliticalBetting.com.