POLLSTERS ICM seem to be hedging their bets on the outcome of the EU referendum with two new polls giving completely different results.

The polls, carried out for the Guardian, showed those who were spoken to by ICM on the phone would vote to remain in by 47 per cent to 39 per cent. However, those who took part in the internet poll said they would vote to leave 47 per cent to 43 per cent.

All those taking part were asked the same questions over the same period of time. And the method of analysing the date is similar. Puzzled pollster Martin Boon from ICM admits he doesn’t know why that difference exists.

“If you want to ask me, which is unlikely, the answer you’d get is ‘I just don’t know’. I can see reasons why phone polls overstate Remain shares, and reasons why online polls overstate Leave shares.”

People might then expect the reality to be somewhere in the middle, but that, says Boon, is not borne out by the evidence. “Polling has often depended on hidden error cancelling itself out, but it seems increasingly unlikely that pollsters can depend on that on this occasion.” In ICM’s phone poll when the 14 per cent of undecideds are removed remain looks set for a clear 10-point lead, by 55 per cent to 45 per cent.

When the don’t knows are removed from the online survey, however, the result is 52 per cent to 48 per cent in favour of Brexit.

ICM Unlimited interviewed 1,002 people by phone, and 2,048 people online on 13-15th May 2016.

Professor John Curtice said the new data from the ICM poll left his poll of polls unchanged, with both Remain and Leave on 50 per cent each.