Poll predictions

A Mori poll carried out between 21 and 25 April for The Financial Times suggests that once intention actually to vote is factored in, a ten point lead shrinks to a mere two points, with Labour getting 36 percent of the vote and the Conservatives 34 percent.

Then, The Guardian today reports that the two parties are "neck and neck in key marginals", facing strong voter scepticism and a "disciplined Conservative attack”. This has reduced Labour's lead to two percent or less in key constituencies.

It seems here that the polls are beginning to reflect the reality on the streets. This is certainly what we are finding. During our canvassing in some areas where historical data have shown mixed responses, it has been difficult to find anyone to admit an intention to vote Labour.

Other constituencies with which we have conferred are reporting the same experience and, all together, our impression is that this contest is much closer than it appears to the pollsters.

Another of our findings is that "Europe" is raised spontaneously "on the doorstep" far more often than the poll ratings would indicate, although in a complex manner. People will talk about the "headline" issues but then bring up the effect of EU policies in a variety of ways.

Talking to some truckers yesterday, there is holy war about the Working Time Directive, with some drivers losing hundreds of pounds in wages. They are not at all happy and there are well over half a million of them, all thinking hard about their vote.

The indications are, therefore, that the polls are no better in identifying voters’ concerns than they are in predicting them more straightforward question of which party is in the lead. The politicos would do well to be cautious about poll predctions.