The monthly ComRes online poll, conducted for the Independent on Sunday and Sunday Mirror, is out tonight and has topline voting intention figures of CON 29%(-1), LAB 35%(-3), LDEM 8%(nc), UKIP 19%(+4). Changes are from ComRes’s previous online poll in mid-April. The changes are in line with other polls we’ve seen since the local elections, a slight narrowing of the Labour lead over the Conservatives and growing support for UKIP. The 19% is the highest figure that UKIP have scored in any poll so far.

The other questions in the poll asked best Prime Minister (I think the first time ComRes have asked it recently, and a welcome break from their tyranny of agree/disagree statements) with Cameron on 32%, Miliband on 24%, Clegg on 6%.

There is also a question where 49% of people agreed that “If a party wants my support at the next general election, it is important to me that they offer a referendum on Britain’s membership of the EU”. The question of whether people actually care about Europe and referendums and whether it will actually change votes is a key issue at at the moment. Unfortunately is it not as easy as this to actually answer it – but that it were!.

Regular readers will know that questions attempting to measure the salience of particular issues at election are a particular bug-bear of mine, one that I’ve written about in the past. There are multiple problems with questions like this. The first is that questions like this take an issue out of context and give it false prominence – that is, if you ask about an issue in isolation (and sometimes on the same page as a grid of other statements asking about the same issue!) people may think they are important, but come an actual election there are all sorts of other issues like the economy, crime, the NHS, pensions, taxes and so on that people may see as even more important. The second is that respondents to surveys are not stupid – they know that such poll questions are used by papers to show unhappiness with a policy and will use it to register their support or opposition to a policy regardless of whether it would actually change their vote. Thirdly, and most importantly, is that as people we are often not very good at actually understanding the drivers behind the decisions we make, normally rather overestimating how rational and calculating we are. Key driver analysis of British Election Study data tells us that things like party identification, perceptions of the leaders and perceived competence are the things that drive votes… not policies on individual issues.

The simplistic view of how policies affect voting intention – if people like a policy it wins votes, if they don’t like a policy it loses votes is just that – simplistic. Individual votes don’t win or lose votes. However, they presumably do play into wider perceptions of parties and leaders and how people rate them. So the issue of Europe may well have an impact in terms of whether parties and leaders are seen as caring about the same issues as the public do, being willing to listen to the people, whether they are strong or weak leaders with a vision or purpose, whether parties are united or competent. How it might affect people’s vote is not a question that can be easily answered, let alone with a single question.

UPDATE: There is also an ICM poll in the Sunday Telegraph. They don’t have normal voting intention (instead having the ICM wisdom index thingy that asks people to predict the shares of the vote rather than ask how they themselves would vote – the figures this month are Conservative 29%, Labour 32%, Lib Dem 16%, UKIP 15%) but do have EU referendum voting intention, asked using the wording in the Conservative party’s draft Bill. 46% say they would vote NO (to leave), 30% would vote YES (to stay).

UPDATE2: The fortnightly Opinium poll for the Observer bucks the trend of a narrowing Labour lead, but does also have its own UKIP high. Topline figures are CON 27%(-1), LAB 37%(+2), LDEM 7%(-2), UKIP 20%(+3)

The monthly ICM poll for the Guardian is out tonight. Topline voting intentions stand at CON 32%(+1), LAB 38%(-1), LDEM 15%(nc), UKIP 9%(+2).

The Labour lead of only six points is significantly lower than other polls are showing (indeed, it’s the lowest we’ve seen since February), but this is due to methodological reasons not some sudden Tory recovery. ICM typically show lower Labour leads and higher levels of Lib Dem support than other companies because of the way they reallocate people who say don’t know to the party they say they voted for last time and weight down the answers of people who didn’t vote last time. The figures are, therefore, pretty much in line with what we’d expect, with the changes from last month’s ICM poll all within the normal margin of error.

The Church of England have released a poll they claim shows the vast majority of people believe in the power of prayer, when it does no such thing. There is nothing at all wrong with ICM’s actual polling, which asks people “Irrespective of whether you currently pray or not, if you were to pray for something at the moment, What would it be for?” (emphasis is mine). A perfectly reasonable question, asking people what they would pray for, if they were the sort of person who did pray.

However, the Church of England have gone rather rogue in interpreting the results, deciding that everyone who gave an answer to ICM’s hypothetical question of what people would pray for if they prayed must therefore believe in prayer – putting out a press release claiming that “Four out of five British adults believe in the power of prayer”. The Telegraph has gone on a similar flight of fancy, declaring “Six out of seven people still believe that prayers can be answered despite a dramatic drop in formal religious observance, a study has found.”

In a population where only around half of people believe in a god at all, any claim that 80% of people believe prayer works should ring alarm bells anyway. For the record the last poll I can find that actually asked whether people believed that prayer worked was by YouGov for the Sun in 2012. That found 31% of people believed that prayer worked in some way (that is they thought prayers were heard by God, or were physically answered in some other way), compared to 45% who did not and 25% who weren’t sure.

Hat tip to Alex Hern at the New Statesman for spotting it – his own mockery is here.

Meanwhile this morning’s YouGov poll for the Sun had topline figures of CON 30%, LAB 41%, LDEM 13%, UKIP 11%. The thirteen points for the Liberal Democrats is the highest that YouGov have shown them since November 2010. While all the usual caveats about individual polls apply, it is indicitative of a broader underlying trend – since the end of last year there has been a definite uptick in levels of Lib Dem support in YouGov’s daily polling. Last autumn YouGov were typically showing them at 8-10%, in recent weeks they have typically been showing them at around 11-12%.