|Populus for the Times||Apr 7||Feb 7|
|LAB to CON swing from 2005||5.4%||6.5%|
But why record the figures without rounding?
Tonight’s poll by Populus for the Times is hugely significant because the firm’s methodology is based on ICM and generally the fieldwork is carried out from ICM call centres.
So it provides us with something that’s quite similar to compare the recent ICM poll with – that’s the one that was carried out over the Easter weekend for the Guardian where the Tory lead was down to just 4%.
At the time even the commissioning newspaper was mentioning the traditional problems phone pollsters can have with bank holiday polling.
The Populus poll is also the first national voting intention survey from the firm for more than two months and the comparisons are from early February.
The feature I can’t understand is why the shares are shown to one decimal point – I can’t recall that happening in modern times.
So a poll that’s broadly in line with other firms with numbers that should produce a small overall majority for the Tories taking the marginals into account.