There are new polls by ComRes, Opinium and YouGov in the Sunday papers. Toby Helm at the Observer has tweeted the Opinium results and John Rentoul has just posted up the ComRes results here.

Opinium in the Observer have topline figures of CON 27%(+1), LAB 36%(-1), LDEM 7%(+1), UKIP 20%(-1). Changes are from their poll a fortnight ago, and clearly don’t show any massive change.

ComRes for the Sunday Indy and Sunday Mirror meanwhile have topline figures of CON 26%(-3), LAB 35%(nc), LDEM 10%(+2), UKIP 19%(nc). Changes there are from the last online ComRes poll a month ago.

Here we have two online polls that still show UKIP up around 20% in contrast to the ICM and MORI telephone polls earlier this week that both showed UKIP at 12%. As I said when the ICM and MORI polls were published, the lower score for UKIP wasn’t a sign that their support has collapsed again, but the result of methodological differences between telephone and internet polls. For some reason most online polling companies (particularly newer online polls – YouGov tend to show levels of UKIP somewhere in the middle) tend to show higher levels of UKIP support. This could be due to the lack of interviewer effect, or could be a result of sampling problems for one or the other. Either way, its a reminder to take into account the house effects between different pollsters, and not to mistake differences in methodology for changes in support.

Looking at the other findings in the polls, ComRes also asked about whether people had favourable or unfavourable opinions of politicians and parties. They found Cameron was no longer more popular than his party, but neither was he a drag on them – 23% had a favourable view of Cameron, 23% had a favourable view of the Tories. Ed Miliband continues to trail behind Labour – 28% have a favourable view of Labour, compared to 20% for Miliband.

YouGov’s Sunday Times poll will likely surface tomorrow morning.


The fortnightly Opinium poll for the Observer is out tonight and has topline figures of CON 26%(-1), LAB 37%(nc), LDEM 6%(-1), UKIP 21%(+1). There is no significant change from a fortnight ago, though for the record the 21% is the highest that UKIP have recorded with Opinium and the 6% is the lowest the Liberal Democrats have recorded this Parliament*. Opinium do tend to produce some of the higher UKIP scores, something they have put down to not using political weighting in their polls.

The Opinium poll also asked about the situation in Syria, and as with the recent YouGov polling on the same subject found very little support for arming the Syrian rebels. Only 24% supported sending arms or military supplies.

(*Whenever the Liberal Democrats suffer a really low score someone will wheel out an anecdote about remembering when the Liberal Democrats were just “an asterisk” – what polls sometimes do to show a figure is less than 0.5%, but not actually zero. As far as I’ve been able to tell this isn’t actually true, or at least, has never been true in a national opinion poll. The lowest ever Liberal Democrat score I’ve managed to locate is 3% in an ICM poll for the Sunday Correspondent in 1989)

UPDATE: Just to re-emphasise, as I can see people already getting stupidly overexcited on twitter. The poll does not show any significant change or movement, it is just that Opinium’s methodology normally shows very high scores for UKIP and Other parties (probably due to their decision not to politically weight), and therefore relatively low scores for Con, Lab and Lib Dem.


There are new new polls in the Sunday papers, but both are from before the local election results (Opinium wholly before, YouGov mostly before) so don’t expect too see any impact from them yet. YouGov’s topline figures were CON 30%, LAB 40%, LD 11%, UKIP 12%; Opinium’s were CON 28%, LAB 35%, LDEM 9%, UKIP 17%. The last two YouGov polls looked as if the recent narrowing in the polls may be fading away again, but Monday’s polls will likely be affected by the local elections so we may never know for sure.

The first thing to note in the YouGov results is not one of the ad hoc questions, but the regular economic tracker of the “feel good factor” – that is, the proportion of people who expect their economic situation to get better minus those who expect it to get worse. In recent years this normally pootles along at around minus 40, extremely negative. Last week it perked up a bit to minus 35, equalling the highest since May 2010. This week it has risen again to minus 31, now the highest since 2010. It would probably be premature to talk of optimism, but perhaps the pessimism is softening. Keep an eye on it.

There were a brief couple of questions on UKIP, 49% of people think Ken Clarke’s description of UKIP as a collection of clowns was unfair, 31% was it was a fair description. Whether they agreed with it or not, a large majority (77%) thought it was bad tactics to describe them in that sort of language (an opinion the local election results would seem to have vindicated!)

YouGov also repeated a series of questions about how well Ed Miliband is doing as Labour leader, last asked in September 2012 (actually some were also asked just after the Labour conference too, but that had given Miliband a big boost in his ratings so would have been a bit of an unfair comparison). They don’t show any obvious improvement in perceptions of Miliband – only 22% think he has provided an effective opposition, only 22% think he has made it clear what he stands for and only 25% think he would be up to the job of Prime Minister.

Asked whether they think Miliband will ever be Prime Minister, 33% think it is likely, 54% unlikely. However, when YouGov ask what people think the result of the next election will be 49% say they expect a Labour-led government which, barring a surprise change of leadership suggests they do actually expect Ed Miliband to be PM! I suspect the apparent contrast between these two answers is an artefact of people just not seeing Miliband as Prime Ministerial… when they stop and think about the next election, they expect Labour to win, but their gut reaction to Miliband is that he doesn’t look like someone who is going to be PM.

Whenever I write about Ed Miliband’s negative ratings it provokes a lot of discussion so I may as well repeat what I’ve said earlier. However bad his ratings (and compared to many past leaders of the opposition they are bad), Labour do have a substantial lead in the polls, so they can’t be that much of an obstacle. Even if it he is drag on their support, right now it is clearly not preventing them getting enough support to win. The question, therefore, is whether it will become more of a factor in an election campaign. Are we currently seeing mid-term polls where people are just driven by opinion of the government and opinion of the opposition doesn’t matter, but nearer the election it will be more of a choice between two alternative governments and poor opinions of Miliband will matter more? Or have Miliband’s negative ratings already been “factored into the price” and won’t matter anymore come the election than they do now. It is beyond the ability of polls to tell, and most opinions I have seen on it so far have said far more about what the person saying it would like to be true than anything else.

Finally, on the back of the various operation Yewtree arrests YouGov asked about the naming of people who have been arrested or charged over offences, but not yet found guilty. There is a widespread belief that people who have been arrested but not charged with any crime should remain anonymous. For allegations of sexual assault 74% of people believe that those accused should remain anonymous, with only 15% thinking they should be named. Public opinion is more evenly split towards people who have been charged, but have not yet faced trial – 43% think those charged with sexual assault should remain anonymous until trial, 45% think they should be named. The figures shift slightly when asked about other crimes, with a clear majority of people thinking those charged with murder or terrorism should be named.


The fortnightly Opinium poll for the Observer has topline figures of CON 29%(+1), LAB 35%(-3), LDEM 8%(nc), UKIP 17(+1). I am always extremely cautious about reading movements into polls – more often than not they turn out to be no more than the result of random movement within the normal margin of error – however we do seem to be seeing a consistent trend. YouGov’s dailing polling for the Sun which normally shows Labour leads of around 10 points has produced leads of 7, 8, 11 and 7 this week, Ipsos MORI showed Labour’s lead dropping by four points, ICM by two points and now Opinium by four points.

Just as I’d advise caution in deciding whether or not there is a change in the polls, one should be equally cautious in assuming what the cause might be. Don’t just leap at the most apparent story in the news! Clearly one obvious explanation would be the coverage of the Thatcher funeral, but it doesn’t follow that this is automatically the cause (if it is, of course, then I would expect any narrowing to be very short lived. A bit of positive TV coverage of a leader from long ago is probably not going to lead to any long term shift).


The fortnightly Opinium poll for the Observer is out tonight and has topline figures of CON 28%(nc), LAB 38%(nc), LDEM 8%(-1), UKIP 17%(+1). Voting intentions are pretty much identical to a fortnight ago – the UKIP score looks startling but Opinium have had them this high for a month (they tend to prodce one of the higher scores for the party, which Opinium themselves put down to the fact they don’t use any past vote weighting.)