Tories set to hold Newark


The Conservatives are on course to hold onto Newark in the by-election on Thursday, according to a poll I have conducted in the constituency. I found the Tories on 42%, with UKIP second on 27% and Labour third on 20%.

This represents a fall in vote share for all three established parties since the last general election: the Conservatives are down 12 points, Labour down 2 and the Liberal Democrats down 14 points on their 2010 score.

Two thirds of Newark voters said either that they were satisfied with the job David Cameron was doing as Prime Minister (34%) or that even though they were dissatisfied they preferred him to Ed Miliband (31%). Only one fifth, including less than a quarter of UKIP voters and only two thirds of Labour supporters, said they would rather see Miliband in Number Ten.

The Conservatives led comfortably on all policy issues – including managing the economy, tackling the cost of living and Britain’s relationship with the European Union – with the exception of “improving public services like schools and the NHS”, on which they were level with Labour.

Around four in ten expected the economy as a whole to improve over the next year for the country as a whole, including 62% of Conservative voters and around a third of Labour and UKIP supporters. Only 16% thought things would get worse, a similar proportion to that saying they expected things to get harder for themselves and their families. While a quarter expected their personal circumstances to improve, more than half said they would probably stay the same.

Cameron and George Osborne were more trusted to run the economy than Miliband and Ed Balls by a margin of 62% to 24%. A majority of UKIP voters and a quarter of Labour supporters said the same.

This poll shows a bigger Tory lead than was found by Survation in their by-election poll published at the weekend. Interestingly, though, the two polls put UKIP within a point of each other. This would seem to suggest that the “spiral of silence adjustment”, used in my poll but not in Survation’s, made little difference in UKIP’s case. The adjustment works by re-allocating a proportion of those who refuse to state or claim they don’t know how they will vote to the party they voted for at the last election. It was introduced to help account for “shy” voters who were reluctant to admit their allegiance, a problem which had the effect of seriously skewing polls at previous elections. The similarity of the UKIP share in polls that did and did not use this adjustment suggests that there is nothing shy about the party’s voters; they do not coyly claim to be undecided.

This makes sense when you consider that most UKIP supporters say the chance to register their discontent is one of the main reasons for their decision. Seven in ten of those planning to vote for Roger Helmer on Thursday say they are making a general protest to show they are unhappy with all the parties; only 16% of them say this was not a factor. More than six in ten UKIP supporters say they were sending a message that they were unhappy with their usual party – more than twice the proportion of Lib Dem voters (and three times the proportion of Labour voters) saying the same.

A by-election two weeks after the European election means Newark’s electors have enjoyed the prolonged attention of the parties, a privilege for which they must surely be grateful. The evidence from my poll is that the Tories have had the better of an intense ground war and have by no means taken the seat for granted. More than nine out of ten voters say they have heard from the Conservatives locally, including 81% who have had literature through the door; nearly half have received personally addressed mail. Eight in ten say they have heard from UKIP; the party is reported to be slightly more active than Labour in all elements of the local campaign.

The poll was conducted in the week before polling day, and just under a fifth of voters say they may yet change their mind. Despite this, it looks clear that the next MP for Newark will be Robert Jenrick.