Sunday Polls

Lord Ashcroft has another of his big polls out today (thought the fieldwork is about two months old), this time an update of his previous look at the Conservative vote. As ever, it’s an interesting read and I’ll leave it to you to read yourselves here. Ashcroft’s conclusions don’t show any radical changes from the previous wave, often a problem with good research. Public opinion doesn’t actually often shift around and show exciting new patterns, so good solid research will often confirm things we already guessed at anyway, or that things haven’t really changed much.

Ashcroft’s key points are on page 9 of his report. They say about 23% of people who would vote are Tories who have remained loyal from the last election, a further 6% are people who didn’t vote Tory last time, but would now. On top of that there are 3% of people who DIDN’T vote Tory last time but might consider it, and a further chunk of people DID vote Tory last time, wouldn’t now, but may or may not consider doing so at the next election. That’s basically the battleground the next 16 months will be fought over. I should add a warning not to take the boundaries of that battleground as being set in concrete – people are not that good at predicting their future behaviour, some people who say they’ll definitely vote X won’t, some people who say they’d never vote Y will, and so on. Treat things as a good guide, but remember the boundaries are fuzzy and changeable.

Those former Tories are still fairly Toryish (after all, they voted for them last time) and most want to see the Conservatives win the next election, prefer the Tories to Labour on policy issues and (despite being dissatisfied with him in general) think Cameron would be the best PM. How things pan out at the next election will largely depend on how many of those people will go back to the Tories (many are currently UKIP, but many others are don’t knows) and at the other end of the scale, how well Labour hold on to the support they’ve gained since the election. In 2013 neither party did fantastically well at their respective challenge! The Conservatives gained a meagre one point, Labour managed to lose three points.

More generally 28% of respondents as a whole are satisfied with Cameron as PM, 32% are dissatisfied but prefer him to Miliband, 40% are dissatisfied and would prefer Miliband. In a forced choice 57% of people trust Cameron & Osborne more on the economy, 43% Miliband & Balls. Labour lead on the cost of living, and perhaps most interestingly, on balance people think that they themselves would be better off if Labour won, but think that the country as a whole would be better off if the Conservatives won (the obvious implication is that some people think that the country would be better off if the Tories won, but that the benefits would go to people richer than them.)

All in all, the picture is as we’d expect it. Labour are still ahead (the fieldwork itself was done back in November, so it’s not new anyway), Cameron is preferred to Miliband as Prime Minister, the Conservatives lead on economic competence and national growth, but Labour are more trusted on cost of living issues.

I should make a comment about some of the media reporting of the finding that the Conservatives have lost 37% of their voters since the last election. This is accurate, but far less exciting than it sounds. A lot of it is just down to people saying don’t know, it happens in all polls, other bits are just the natural churn you get in all directions between parties. To put it in context, Labour have lost 22% of their vote since the last election, which would make a nice headline of “Labour lost one in five of their voters!” if one wanted to spin a poll badly for Labour. The more salient facts are that the Tories have gone from 37% at the last election to 30% in this poll, Labour have gone from 30% to 39%. The underlying churn and back and forth between parties is good for understanding what has driven this but is bad for creating non-misleading headlines! There is churn in all elections, even a party on a roll will lose some of its previous supporters, the key is gaining more than you lose.

Meanwhile there were two polls with new fieldwork out in the Sunday papers. The fortnightly Opinium poll for the Observer had topline figures of CON 30%(nc), LAB 37%(nc), LDEM 8%(nc), UKIP 17%(+1) (tabs here). A Survation poll in the Mail on Sunday had topline figures of CON 31%(+2), LAB 35%(nc), LDEM 11%(-2), UKIP 16%(-1) and European election voting intentions of CON 23%, LAB 32%, LDEM 9%, UKIP 26%.


Elections in 2014

After the Christmas and New Year break we should be getting some polling tonight – Lord Ashcroft at least is promising a new batch of research (and perhaps Opinium will start up their fortnightly Observer polls again – YouGov’s daily polling doesn’t start again till Monday).

In the meantime, I’ve updated the site to reflect the elections in the year ahead:

Polls on the European election so far are here, though note the experiences of 2009 when early polls for the European election bore very little resemblence at all to the actual result. We don’t know if it was because people didn’t really consider European voting intention until much closer to the time, or the increased publicity UKIP got in the run up to the European elections, or because the expenses scandal shook things up, but one way or the other UKIP support massively increased in the run up to the European elections in 2009 and polls conducted more than a month or so before were of little use in predicting them.

On the election guide part of the site I’ve also added information on the candidates standing so far (Scotland, North East, North West, Yorkshire, West Midlands, East Midlands, Wales, Eastern, South West, South East, London)

Aside from the locals on the same day as the European election, the other big election is the Scottish Independence referendum on the 18th September (my birthday incidently – what better present can one get for a psephologist than a referendum for your birthday?). Polls on the referendum so far are all collected here.


-->