CON 40 (-1) LAB 30 (+3) LD 18 (nc)

What a difference from ICM

This is quite a dramatic poll which is very unexpected simply because Populus operates in a very similar manner to ICM and indeed usually uses the firm for its field-work.

So the variance from the ICM survey published only yesterday is quite remarkable,

One thing I will be looking at when the data is published is the scale of the “spiral of silence adjustment” where Populus allocate a proportion of the won’t says and don’t knows in line with how they said they voted last time.

You should only compare a survey with one from the same firm so it’s hard to draw any firm conclusions.

But if other polls from other pollsters start showing that the gap is closing then we might have to start reviewing our view of the election.

A 10% lead might not be sufficient to give Cameron a majority.

Mike Smithson