Tonight’s YouGov poll for the Sun has topline figures of CON 39%, LAB 40%, LDEM 8%. This follows figures of Con 40%, Lab 40%, Ld 9% yesterday, suggesting that the underlying picture in YouGov’s daily polling continues to be Labour and the Conservatives neck-and-neck with one another.

Earlier in the week I missed a discussion between Danny Finkelstein (probably behind a paywall), Will Straw and various others about the Sunday Times/YouGov poll and whether the question about Ed Miliband being too ugly to be Prime Minister was a legitimate thing to ask about.

For those wishing to save time, my short answer is “hell, yes!”

Now the long answer. In one sense, it’s not really up to the pollster to judge. A pollster might refuse questions that are grossly tasteless or risk contempt of court, but generally speaking we don’t judge. Sure, I might like every question to be a highbrow academic one, but newspapers commission polls for journalistic reasons, not out of the goodness of their heart, and that means our fair share of “which dog does this politician most resemble” questions.

However, questions about a politicians looks are entirely legitimate, since it is an important factor in their electability. In his post Danny Finkelstein is right to point to the work of Daniel Kahneman and Robert Cialdini, but there here is also more direct empirical evidence, such as this paper from last year which based on a large scale US survey found people with little knowledge of politics but who watched a lot of TV were likely to be influenced by hte candidates’ appearances (which also includes a summary of other relevant recent research), or this paper (Free, but requires annoying registration) that found people’s judgement of which of two unknown candidates *looked* more competent predicted who actually won at a much greater rate than chance would suggest.

We might wish that people voted purely on dry policy and ideological preferences, but all the evidence is that they don’t. If factors like what politicians look like actually are very important, then it is right and proper that polls ask about them – our role is to reflect public opinion as it is, not as we would like it to be and, by extention, we should be investigating the factors that actually drive voting intentions, not those that we think should. If anything there are too many questions asked about the minutae of policies and not enough about party image and perceptions of the leaders.

The downside of the particular question in the poll is that it didn’t, in the end, really answer the question very well. We know from the other questions in the poll that Miliband’s appearance *is* an issue – 70% of people agreed that Miliband does “not look or sound like a possible Prime Minister”, including 79% of those people who said that he has the “right policies”.

However, 72% of people said that Ed Miliband wasn’t “too ugly” to be Prime Minister, so the question doesn’t really tell us much about what those 70% think is the problem with him. I’m not that surprised – just because appearance is actually important for a politician, it doesn’t mean the public recognise this. I think you’d be hard pressed to find people who actually admit to pollsters, or to themselves, that they are influenced by it – we think we are influenced by policies, ideologies, perceptions of trust and competence… even if subconsciously we are biased towards the taller candidate, or the one with the nicest hair.

The fact that 70% recognise that there is such a thing as not looking like a possible Prime Minister does at least suggest that polling on it isn’t a complete write-off, but so far we don’t know anything about why. Is it Ed Miliband’s looks, his mannerisms, his voice, the way he articulates his arguments, or just looking goofy in photos? I don’t know, but it’s a problem that is important for him to solve if he wants to be a success – and therefore something that polling should be exploring.

(For clarity, normally the Sunday Times poll is written by me, but I’m currently off on leave, so I didn’t actually write the question I’m defending!)


Following the weekend polling on Scottish Independence, there is a new YouGov poll of Scotland for Channel 4 news out today. Full tabs are here. Topline voting intention figures show the SNP continuing to hold a healthy lead in Holyrood voting intentions and now also ahead of Labour in Westminster voting intentions.

Westminster: CON 16%, LAB 35%, LDEM 7%, SNP 37%
Holyrood constituency: CON 13%, LAB 32%, LDEM 7%, SNP 44%
Holyrood regional: CON 13%, LAB 31%, LDEM 7%, SNP 39%, Grn 5%

On the referendum, there is an pretty even split on whether it should be a straight in-out question (43%) or a three way referendum (46%). On the timings, 38% would like it this year or next, 33% would like it delayed until 2014.

Finally, on actual voting intention in a referendum, on “devo-max” 58% of those who gave an answer said they would vote in favour, 42% said they would vote against. On full independence, 39% said they would vote in favour, 61% said they would vote against.

The survey asked the questions as mock-ballot papers, which as ballot papers tend not to have abstain or don’t know boxes, means it gets answers from the vast majority of people. If you instead take only those who said they were 10/10 certain to vote it shifts things slightly in favour of extra devolution or independence, with respondents backing devo-max by 60-40 and opposing Scottish independence by 41-59.

Last night’s YouGov poll also included questions on Scottish independence, while ICM carried out parallel English and Scottish polls for the Sunday Telegraph. YouGov found strong support for a referendum on Scottish independence, with 61% of people supporting holding a referendum (including 76% of respondents in Scotland), although as we’ve noted before, referendums are intrinsically popular – I’ve yet to see a poll question showing people don’t want a referendum on something!

On timing, the balance of support in Britain as a whole seems to be for a referendum earlier than 2014. In YouGov’s poll 36% support a referendum this year or next year, 23% support a referendum in 2014, 7% later. ICM found a similar balance in England – 52% supporting a referendum as soon as possible compared to 25% who would prefer a referendum in 2014. In Scotland, however, there is more support for a delay to 2014 – ICM’s Scottish poll found 43% support a referendum as soon as possible, 41% support a delay. In YouGov’s Scottish subsample 25% would like a referendum this year or next year, 52% would refer a delay until 2014.

Asked about the referendum, neither poll had a straight “how would you vote” quesion. In YouGov’s poll they asked if people would support or oppose Scotland becoming a country independent from the UK. Overall 37% of people supported Scottish independence, 39% opposed it. Amongst respondents in Scotland the split was 45% support/ 45% oppose, amongst respondents in England the split was 36% support and 39% oppose.

ICM asked if people approved or disapproved of Scottish independence – in Scotland the split was similar to YouGov’s: 40% approved, 43% disapproved. In ICM’s English poll there was higher support, 43% of English respondents approved, 32% disapproved.

ICM also asked what people’s preference would be in a three question referendum – in Scotland 37% said they would prefer the status quo, 26% full tax and spending control, 26% full independence (Of course, if there was a 3 question referendum it looks as though it would be two seperate questions, not one three way question)

ICM found that 51% of English respondents thought that Scotland would be worse off if independent, with only 23% thinking they’d be better off. In Scotland the split was 38% of respondents who thought Scotland would be better off, 41% worse off. YouGov asked similar questions for both Scotland and England – again, a majority of respondents in England thought that Scotland would be worse off outside the union, with Scottish respondents more evenly split. For the English question, 36% of respondents in England thought that England would be better off without Scotland, 17% worse off. Most Scottish respondents thought that England would be worse off without Scotland.

In short, English respondents tend to think that Scotland gains more from the Union than England does, and in ICM’s poll at least this makes them more likely to support Scottish independence. Scottish respondents remain more divided about whether the Union benefits Scotland or not, and hence opinions on Scottish independence are also more evenly divided.

As we head towards a referendum there will no doubt be an awful lot more polling on Scottish independence. The figures on how people would vote in a referendum are probably not very meaningful right now – that’s one thing we should learn from the AV referendum – more important right now is understanding the broader opinions and concerns that lay behind those opinions and I expect we’ll have a lot more to mull over in the weeks and months to come.

Full tables for the YouGov/Sunday Times poll are now up here, covering Miliband’s leadership, Scottish independence, abortion, alcohol and shoplifting.

Last night I pondered whether the reason the polls were still close between Labour and the Conservatives was a lasting effect of the veto, or a reflection of Labour’s current troubles. The regular trackers in the YouGov poll would suggest the latter – government approval and David Cameron’s approval ratings are both falling back towards their pre-veto levels, government approval is back down to minus 26, David Cameron’s approval rating is back down to minus 10. In contrast, Ed Miliband’s figures get ever worse, dropping to minus 49 (from minus 46 a week ago). Amongst Labour’s own supporters only 46% think he is doing well, compared to 49% who think he is doing badly.

Asked if Miliband had the right policies to be Prime Minister and whether he looked or sounded like a Prime Minister, only 7% thought he both had the policies and the look/sound to be PM. 43% thought he had neither (including, as one would expect, most Tories). The interesting bit is the rest, only 4% thought he looked like a PM but didn’t have the right policies, 27% thought he had the right policies, but didn’t look or sound like a PM. Amongst Labour’s own supporters only 16% thought Miliband had the right policies and the right look/sound, 5% of Labour supporters thought he had the right look/sound but the wrong policies, 59% of Labour supporters thought he had the right policies but didn’t look or sound like a possible Prime Minister.

For all the discussion of Labour’s policy stance on the economy (though in the longer term, that will be extremely important too), this appears to be the ultimate problem with Ed Miliband – people don’t think he looks the part of Prime Minister. It is not, as John Humphrey’s suggested, anything as crude as Ed Miliband being “too ugly” to be Prime Minister (YouGov asked and only 10% agreed), but a general image. It backs up earllier findings like that in December when, despite Labour having been ahead in the polls for a year, only 17% of people and only 37% of Labour’s own supporters thought it likely that Ed Miliband would ever be Prime Minister. This is a real problem for Miliband – policies can be changed, it is extremely difficult to change the public’s perception of a leader once it has settled. Miliband’s ratings did get a good hike after hackgate last year, but it was purely temporary, Labour need to get something like that which sticks.

The rest of the poll covered Scottish independence (which I’ll do a seperate post on later),
attitudes towards alcohol pricing, lobbying, shoplifting and abortion.

On Alcohol pricing, people are pretty evenly split over cut-price promotions on alcohol – 47% think they are a good thing, 42% a bad thing. 53% oppose a minimum price on alcohol, 47% support it, although largely at at quite low levels. 30% would support a minimum pricing at the suggested 45p a unit or less (the equivalent of about £1 for a pint of beer), 17% would support a higher minimum price.

On shoplifting, 16% of people admitted that they had shoplifted at some point in their lives. 50% of people saw it as a less serious type of theft than burgulary or mugging, compared to 45% who thought it was about the same. Asked what the appropriate punishment should be for a first time shoplifter, 23% thought they should be given a caution, 30% a fine, 30% community service, 11% a jail sentence.

Finally YouGov asked about the time limit for abortion. 5% of people supported a higher limit, 34% supported the status quo of a 24 week limit, 37% supported reducing the time limit and 6% supported a total ban on abortions. As you often find on abortion questions, women were more likely than men to support a reduction in the time limit for abortion (49% of women supported a tighter limit, compared to 24% of men.)