Updates from the other place

Things have been a little quiet here but I've had a few pieces up on the work site and there should be a new one every Monday.

Some recent highlights:

John Rentoul at the Independent on Sunday asked the heads of all the regular polling companies to give their predictions for the general election (a poll of pollsters). Ours is on the page for that article but there's a bit more explaining why we said what we said here. There's also an interesting post here by the team at Survation on their prediction.

At the start of February we rolled out some methodology changes for our general election polling. To overdramatise significantly we're moving from 'peacetime' (i.e. mid-term between elections) to 'wartime' (the informal and formal election campaign periods) so I wrote a little explainer of the changes here.

We did a little work on tactical voting and whether people had chosen their parties because they like them or dislike someone else. It's impossible to measure particularly closely but I think we get a flavour here of what voters would do in different situations where their party isn't really in contention.

I also had two related blog posts about the central conundrum in election polling at the moment. Namely that Labour lead in most polls despite trailing badly on the other traditional predictors of elections: leader ratings and who you trust to handle the economy. The first basically lays out that situation in more detail while the second comes after we first showed the Tories leading in voting intention (albeit within the margin of error). While the economy is affecting how people will vote, it's quite a nuanced subject so we've tried to come up with a way of measuring that in a way that, coincidentally, tracks vote share. More details in the post itself.

Of course voters prefer single-party governments!

For the political geek there were some interesting stats in this week's Opinium / Observer poll, centred around what should happen if there's another hung parliament in 2015.

We asked whether government made up of just one political party or governments made up of multiple parties in coalition best served Britain and the result was fairly overwhelming support for single-party governments by 55% to 15%.

There are a few explanations for this and one is that the majority of governments since our political system began its current incarnation in 1945 have been single party and all of the ones that are remembered generally as successful are single party including the icons of the left (Clement Attlee) and the right (Maggie). Aside from the national government during the second world war, which I expect most simply associate with Winston Churchill rather than parties, the only period of multi-party government was the Lib-Lab pact of the 1970's which few remember fondly.


Read more »

More people think the EU is a 'good thing' so why would they still vote to leave?

Here's a link to a post I put up on the Opinium site a few weeks ago. It looks at the divergence between the two EU questions we ask regularly on political polls. The first asks whether the UK being a member of the European Union is a 'good thing' or a 'bad thing' while the second (more widely reported) asks whether respondents would vote to leave or to stay if there was a referendum tomorrow.

Until recently these have tracked fairly closely but now the percentage saying it's a "good thing" has overtaken the figure for "bad thing" for the first time while more people would still vote to leave than stay.

Anyway, click the link to find out why.

A two tiered debate in 2015 looks more and more possible

I don't remember where I first read it but the most interesting solution to the "Farage" question when it comes to party leader debates in 2015 is to have two tiers or debates like they do in Germany and other countries with multi-party systems.

You take the two candidates most likely to be prime minister, in 2015 this will be David Cameron and Ed Miliband, and put them in a debate on their own and then take the leaders of all the other smaller parties like the Liberal Democrats and UKIP, potentially the SNP or Plaid depending on where it's being broadcast, and maybe the Greens and put them in another debate by themselves. This gives everyone exposure but doesn't create the sort of trainwreck that we saw with the Republican debates in the US in 2012 where you had one, maybe two, serious candidates and several others who were there to get exposure or sell books.

Obviously Nigel Farage would prefer a four-man (and it is men, how strange that 25 years after Thatcher resigned we'd still only have male leaders) debate with the leaders of the Conservatives, Labour, the Lib Dems and UKIP while Nick Clegg may want this or the same set-up as 2010. But it's hard to see David Cameron agreeing to it given his apparent regrets over the 2010 debates.

Having a separate Cameron vs. Miliband debate would no doubt work best for Cameron and the broadcasters would probably come around as well. The only sticking point would likely have been Miliband.

Until now.

In our polling for the Clegg-Farage debate last night we asked about whether voters would support or oppose this (it's near the bottom, just before we traumatised hundreds of people by forcing them to choose between Clegg and Farage for prime minister). The results were a bit mixed which suggests that people didn't really read it too carefully but as we get closer to 2015, the German option seems more and more like the only real debate compromise we'll get.

Cross post:ex-Lib Dems and Miliband's low ratings

I just posted this on the Opinium blog which is me asking whether the fact that Ed Miliband generally gets lower ratings among Labour voters than David Cameron does among Conservative voters can be explained by the fact that such a large proportion of current Labour voters are former Lib Dems who decided to vote Labour the minute Nick Clegg and David Cameron stepped out into the garden for their first Number 10 press conference.

David Cameron's ratings are generally pretty high (80-90% of current Conservative voters approve of him) partly because the Conservative vote has shrunk since 2010 and current Conservative voters are likely to be closer to the core party loyalists than more sceptical, floating voters.
Conversely, Labour's share is about 7-8 points higher than 2010 and therefore includes the core vote of about 29% with the rest being people who have moved to Labour since 2010, including the sizeable block of ex-2010 Lib Dems. Hence Ed Miliband's ratings among Labour voters are generally around 50-55% with far more being in the "Neither approve nor disapprove" category than you get for David Cameron.

Unfortunately the theory doesn't work and, if you read the Opinium post, you'll realise that the title is a QTWAIN. Reasons are after the jump:

Read more »

Treat yourself to a poll on Europe done by (mostly) me

Here's my festive, Christmas post and it's all about the EU!

The big project I've been working on for the last few months has been this (I wrote the report, not the article) which is all about the European Union and Britain's place within it.

Although there's far more data in the tables than we were able to include in the report, the basic conclusions were that British antipathy to the EU has three main sources:

• There's a hostility to the perceived impact it has on immigration which is why I think you can't rely on the "EU" figure being so low in things like the Ipsos MORI issues tracker. Immigration is seen as much more important than "Europe / EU" but the two issues are intertwined as our study shows. Immigration overwhelms the positive effects people think the EU has on areas like travel and the advantages for businesses
• There's a cultural barrier between the UK and the rest of Europe. We're far more likely to feel closer to other English speaking countries than those physically close to us and therefore things the EU does are viewed from a default position of hostility and suspicion regardless of the actual effects it has. It's not so much what the EU does but the fact that it's doing anything at all
• People don't think leaving the EU would be a particularly bad thing for Britain. In most areas (e.g. trade, relations with other countries) they think it will have a neutral or positive effect and therefore the answer to the "in or out" referendum question becomes more of an "eh, why not leave" rather than a "you can never take our freedom!". This second group does exist and is very large and vocal but not a majority and there is a large portion of the population who don't really think about Europe very much and are reacting to referendum polling questions with "well I suppose on balance it'd be better to leave"

The whole thing is worth a read (although I would say that) and there's plenty more data that we weren't able to fit into the report so we'll be coming out with some more tidbits and interesting nuggets as time goes on.

If you're wondering what connection a poll on the EU has with Christmas then that's fair. We published it in early December so for the last few months my to-do list has had the following items next to each other: "Write Europe report. Publish Europe report. Buy Christmas presents. Wrap Christmas presents..." and the two concepts have largely merged in my mind.

If you're looking for something to get you into a proper festive spirit, I suggest this followed by the 6th greatest Christmas film of all time.

More updates from the other place

Again, I've been more active on the work site than I have here so a bit more cross-posting this week. Most recently there's been some polling on whether the BBC has a left wing or right wing bias. The results are here.

The most interesting part was that more Conservatives and UKIP voters thought the BBC was left wing whereas Labour and Lib Dem voters thought it was neutral.

There's a bit of analysis here and a potentially more interesting piece about false equivalence, selective retention and information feedback loops that I'll try to find the time to write as well.

In the meantime we run an Observer poll every other weekend (next one will be out on Saturday 16th November) and we follow that up with top line results and a bit of analysis on the Opinium news page.

Updates from the other place

Just a couple of links really:

After a piece a little while ago on what Labour voters thought of Ed Miliband and what Conservative voters thought of David Cameron, both leaders saw a boost among their own supporters after the party conferences. Full piece is here.

The other one is a bit longer and about the difficulties you find with polling on policies.
We did some work on free schools a little while ago and the experience of condensing entire BBC or Wikipedia articles on the government's flagship education policy into a short, three sentence paragraph is a strange one so the blog post just talks about things to bear in mind when you see polling results on particular policies. Full piece is here.

Things that have happened in recent polling

We've had a few interesting things in Opinium polling over the last few weeks.

The first is that we asked who voters think is likely to win the next general election with all the realistic options being shown (i.e. Conservative majority, hung parliament with Conservatives as the largest party, hung parliament with Labour as the largest party, Labour majority). We asked this back in March and again in September and the major difference is that the Conservatives are a lot more confident of victory than they were six months ago.

Counting a majority or being the largest party in a hung parliament as a "win" for each side, 54% in March expected a Labour win in 2015 vs. just 24% for the Conservatives. In September the figures were 42% and 37% respectively with the main difference coming from the fact that more Conservative voters expect a win.

We also had a look at intra-party approval ratings (e.g. how many supporters of a party approve / disapprove of their leader) and why Ed Miliband's ratings have tumbled recently. My theory about why Labour voters were suddenly less positive than usual about their leader was that this was because of media criticism of Miliband by people from his own side (e.g. John Prescott) which wouldn't make Labour voters any less likely to vote Labour but would make them less likely to approve of their leader.

It also mentions why I thought Ed Miliband's own-party ratings were lower than those for David Cameron and, given that we've just had a Labour party conference where the main announcements were policies that the Labour base will love, it'll be interesting to see what happens to his figures in the weeks ahead.

Some filler to occupy you while I'm out

Bit of a drought recently due to a combination of a crazy-busy period at work and me going on holiday but we have had a few interesting pieces on the Opinium site:

We did a bit of heat wave based polling basically pointing out that while the media use Fahrenheit to emphasise how hot the weather has been, most people actually think in Celsius instead.

The other thing was asking voters of each party what they thought of the others with the most interesting result being that Tory voters are reasonably positive about UKIP but not vice versa. Results are here.

Otherwise I'm off to rural France to remind myself what life was like before mobile phones and internet access.

Nobody wants Margaret Thatcher Day! And other points on the Alternative Queen's Speech

The other week Opinium did a slightly larger than usual selection of topical polling and had a few questions on the "Alternative Queen's Speech", the manifesto of radical (read: vote losing) policies suggested by Phillip Hollobone, Peter Bone and a few other right wing Tory MPs.

It's basically a list of proposals to ban other things that annoy right wingers like Burkhas or forcing schools to remain open during "adverse weather conditions" and various other things that frequently appear in the comments under MailOnline articles.

Predictably enough most people were broadly in favour of various anti immigration measures (such as continuing restrictions on Bulgarian and Romanian nationals) and there wasn't much enthusiasm for privatising the BBC but slightly more for forcing schools to stay open in the snow.

What was depressing was the surprisingly high level of support for banning the Burkha (though I suppose this is more a case of 'well since you ask...' rather than a burning top of mind desire) but arguably less surprising was that only 12%of likely voters wanted to bring in Margaret Thatcher Day while 67% were opposed.

Given the wall to wall coverage of Thatcher's funeral and the deification campaign by various right wing media outlets and politicians it's an interesting antidote. Public views on Thatcher have been well covered, but I think there's general resistance to elevating her beyond a politician, there was general disapproval of giving her a full state funeral and this result comes from the same place.

The full set of results is available at the bottom of the page here.

Maybe people just like the idea of referendums?

I've got another blog post on the Opinium site (link here) about whether people really want a referendum on the EU or just like the idea of referendums generally with a bit of polling data to back it up.

The reason for looking at this is that we regularly see commentators and politicians citing polls that say the public are in favour of a referendum on EU membership. Before this they were polls saying the public wanted a referendum on the Lisbon Treaty and before that it was the EU Constitutional Treaty and the Euro.

But the flaw in these polls is that they don't measure intensity. Our questions ask using an agreement scale (Yes - definitely, Yes - probably, No - probably not, No - definitely not, Don't know/no opinion) but you can only really measure intensity by asking people to prioritise. YouGov do a "pick the top three" most important issues question and even among UKIP voters, 'Europe' is only selected by 32% and 20% of Conservative voters.

Read more »

Polling on Syria

We did some polling last week on the situation in Syria and whether the UK and other countries should intervene.

People are generally opposed the idea of sending weapons to the Syrian rebels or putting troops on the ground but are broadly in favour of humanitarian aid and, surprisingly, enforcing a no-fly zone over Syria.

I've got a quick piece on the Opinium site about possible reasons for this (found here).

The blog section is starting to become a bit more active now so hopefully there'll be some more interesting content coming over the next few weeks. The Syria polling also included some questions on Britain's foreign policy generally, whether the UK is overstretched and what our obligations are so there's analysis of that to look forward to as well as other topical issues.

About that UKIP number (continued)

Earlier this week I put a quick article on the Opinium website just explaining why we think our UKIP number is frequently higher than others. The link is here but the basic idea was that it's down to the fact that we don't do any sort of political weighting beyond whether people are likely to vote.

Just to explain, weighing means counting some people in a survey as slightly more or slightly less than one so that the overall demographics (i.e. the proportion of men vs. women or the percentage of the sample coming from each region) match up with those of the country as a whole.

Opinium figures are weighted to national demographics (that we use in all our commercial polling work as well) and then filtered to those who said they would be likely to vote and told us which party they would vote for.

Filtering to those likely to vote is our way of making sure that we aren't giving prominence to the views of people who aren't going to vote anyway but most other companies also apply a form of political weighting to make sure of this. This either takes the form of asking people how they voted in 2010 and making sure it matches the proportions of that election or giving more or less weight to peoples' answers depending on whether they voted before or not.

The other thing is that, like most companies I think, to see UKIP as an option you have to have said you'd vote for "some other party". In theory surely we'd get the most accurate results by showing the options as they appear on a ballot paper (like Survation do) but the argument against this is that presenting the other parties equally gives them a degree of false prominence given that an actual election will come at the end of a campaign dominated by coverage of the bigger parties. We can't predict what the 2015 campaign will look like exactly but it seems unlikely that UKIP will get as much coverage as they did for the last few weeks of 2012.

Anyway, our approach has worked well in the past and we're confident it's accurate now but we are experimenting with a number of different methods, particularly with forms of political weighting and turnout filtering, to see where we can improve it in the future.

Regional figures don't matter!

On a side note, there was some cheering in UKIP circles at the news that the purple party apparently pulled ahead of the Conservatives in the North West. I haven't checked the exact poll in question but it's worth pointing out that regional splits in national polls are best ignored, particularly for political ones. The sample size is much smaller and the margin of error therefore much higher.
The sample is also designed to match the country's demographics at a national level but may not match those of each region if that makes sense. Put more simply, we make sure that there are the right proportion of, say, 35-54 year olds in the sample but it's perfectly possibly that there will be too many of these in one region and too few in another. The only reason regions are included are because it's standard practise but the figures within them are indicative only and far more prone to being outliers.

UPDATE: I've checked and UKIP and the Conservatives both have 28 people selecting them in the North West. UKIP are on 18% and the Conservatives 17%. This is difference is due to weighting and certainly not significant in any way.

Obligatory "2012 in review" post

So it's that time of year when newspapers and websites fill the gap in actual news over Christmas by reminding us all of what happened in the year we're about to finish.

This is my (mini) attempt at something similar from a political polling perspective and for me 2012 (compared to this time in 2011) has really been about two things:

1. Ed Miliband has gone from being "that guys brother who accidentally became Labour leader" to possibly the most secure of the major party leaders.
While David Cameron and Nick Clegg both have to worry about "kings across the water" in the forms of Boris Johnson and Vince Cable, rumblings about Miliband's leadership have quietened down as his performances at PMQs improve and his lack of close ties with the Murdoch empire saw him immunised against the revelations of the Leveson Inquiry.
As the government (and David Cameron) have seen their ratings decline, Miliband's rise is more a case of looking good by comparison than any consequential achievements on his part. But after setbacks in 2011, Labour have performed the role of opposition effectively with by-election and local election victories as well as Miliband's well received "One Nation Labour" speech and other signs that the party is trying to put together a serious answer to the question "what would you do differently?"
This is also related to the more general point that since the Budget, Labour have opened up a traditional mid-term opposition lead of about 10 points. While Labour people should be a little nervous that their lead is only 10 points in a time of double dip recession and government screw-ups, it is at least an improvement on 2011 when the two main parties were much closer and, in the wake of David Cameron's EU veto in December 2011, the Tories actually pulled ahead briefly.
(There's a useful chart illustrating the change in approval ratings that I'll be adding as soon as I finish compiling it but the voting intention chart is here)

2. The (polling) rise of UKIP.
As the voting intention chart shows, UKIP started 2012 on something like 7% but they end it on just over twice that with 15% in the latest Opinium/Observer poll.

UKIP is an odd one. On the one hand they came in second in the 2009 European Parliament elections (and may come first in the 2014 round) with 16.5% of the vote, beating Labour and the Lib Dems into 3rd and 4th place respectively. Yet in the general election a year later, they were back down to 3% of the vote, no MPs and Nigel Farage failed to win Buckingham despite it being almost tailor made for a UKIP breakthrough.

It's tempting to think of the current UKIP surge as being fuelled by former Conservative voters disillusioned with the government's performance on Europe and other issues. The chart shows some correlation between the drop in the blue line and the rise in the purple one but while 2010 Conservative voters make up a large block of current UKIP supporters, they are not the majority.
Another explanation comes from (Lord) Michael Ashcroft. Say what you will about Lord Ashcroft, he does produce a lot of (sometimes useful) polling data and has conducted another of his super-sized jumbo polls on UKIP's threat to the Conservative party. The results suggest that UKIP considerers, from all parties, are driven by more than just the EU and are more generally dissatisfied with modern Britain in general.
Particular topics that come up tend to be immigration and political correctness as well as Europe with 80% of UKIP "considerers" agreeing with the statement that "UKIP seem to want to take Britain back to a time when things were done more sensibly".
With gay marriage set to become law and the census revealing that Britain is getting less religious and less white, UKIP seem to be turning into the vehicle for resisting these changes and this is probably quite a natural response given that the three main parties are all lead by metropolitan (social) liberals.

We'll have to wait and see whether UKIP have reached a natural peak or whether this surge will continue. With the EU likely to be in the news A LOT in 32013, my money is on the latter.

I tend to be a bit of a grinch when it comes to new year's eve (overcrowded bars and lots of spilt drinks made even worse by the fact that they cost twice as much as usual) so my evening is going to be spent with a can of Fosters and possibly crisps in a bowl if I really want to celebrate.
2012 has been a fairly decent year for me personally so here's hoping 2013 will be pretty decent as well.

Happy new year.

Would 56% of the UK really vote to leave the EU?

So the main finding of this week's Opinium/Observer poll is that if presented with a hypothetical in/out referendum on the UK's membership of the European Union, most people would vote to leave.

The Observer have presented this very nicely as always with a patriotic looking Nigel Farage standing in front of a union flag but there are always a couple of caveats about polls involving referendums, Europe and particularly those involving both.

The first, from Anthony Wells, is that when a polls asks if people want a referendum on any topic the answer tends to be "yes". This is because even if people have a modicum of respect for their own local MP, they are distrustful of MPs in the aggregate and will always choose to have their own say rather than delegate authority to politicians. However, as we saw with the AV referendum and more recently with the Police and Crime Commissioner elections, wanting more direct democracy in the abstract doesn't necessarily translate into a willingness to take part if a vote actually happens.
For this reason we didn't think it was worth while even asking whether there was widespread public demand for a referendum on EU membership as the numbers would have come back as expected and in any case public demand rarely features in the decision to hold a referendum or not.

The second is from Mike Smithson which is that in the two UK wide referendums to have been held (on EEC membership in 1975 and AV in 2011) the vote for maintaining the status quo has been twice as large as the vote for change, 67% to 33% in 1975 and 68% to 32% in 2011. The implication here is that the UK tends to vote for the status quo with the campaign often focusing more on the cost of making the change (as was certainly the case with AV) than the pros and cons of the current arrangement. If a referendum had been held on whether to join the EEC before the Heath government did so then I have always thought that the public's answer would have been 'no'.

Although I have no doubt that the Opinium/Observer poll is an accurate representation of how the public would respond to such a question if asked now, it's important to remember that any actual vote would come at the end of a long and, no doubt impassioned, campaign of voter education.

When we ran the question we did discuss the possibility of giving respondents some background to help them make an informed choice as we do with work for clients when the questions are on a topic that people may have limited knowledge of. However we decided against it not only because we couldn't possibly hope to include all of the information required but also because we have no idea how the campaign would turn out and what information voters would ultimately be responding to.

With the standard voting intention question it's understood that people are responding to the current state of events and that the numbers are a reflection of how each party is doing at the moment, something which could change during an actual election campaign but where most voters can be said to be receiving a reasonably balanced mix of news praising and criticising the government.

With the European Union, there is effectively no "yes" campaign. At the moment the news that the public receives about the EU is almost entirely negative with virulent criticism from the most popular newspapers (save only the Guardian, Independent and Daily Mirror) along with news of impending financial apocalypse and of Greeks and Spaniards rioting.


Until recently I would have expected an in/out referendum on the UK's EU membership to produce a status quo result similar to the other UK wide referendums as the costs of leaving the EU became apparent and the 'leave' campaign had to sell the public on benefits which necessarily remain hypothetical. The business community would make the point about potential tariffs and the effect on exports while the basic unfairness of having to abide by EU rules but having no say in how they are made would ultimately lead more people to stick with the devil they know.

However, the seemingly never ending crisis in the Euro-zone and the possibility of dramatic changes in the institutional makeup of the EU mean that the "status quo" may no longer be the safe and benign "devil you know" and be morphing into something that fundementally changes Britain's relationship with the EU. If that happens then the structural advantage enjoyed by the status quo in any referendum starts to disappear and factors like the majority of the popular press having a very clear agenda on one side start to make a difference.