If you ask people how likely they are to change their voting intention a majority will often say there is some chance of doing so (exactly how much depends on the question). Take for example the PoliticsHome marginals survey – 61% of people said their was some chance of them changing their vote. It’s also worth looking in that poll at what parties those Lib Dem waverers might consider switching to. Of Lib Dem waverers 45% said they would consider voting Labour. Assuming that whatever figures Labour are looking at show roughly the same thing, you can see where they are coming from. If two thirds of Lib Dem voters might consider switching, and half of those might consider voting Labour, there’s your extra five points.

Real life, however, doesn’t work like that. In most cases people saying they might change their vote is a “never say never” answer, people who really are pretty certain of voting a certain way but don’t want to commit themselves totally. In the case of that PoliticsHome poll, the 61% included 33% who said “Unlikely – I may yet change my mind, but I would be surprised if I didn’t end up voting for this party”, people who I think are realistically very unlikely to change.

Polls normally show the Lib Dem vote as being the most “uncertain”, but I suspect this is more a result of the type of person who votes Lib Dem: more likely to see themselves as a floating voter dissatisfied with the big two. Certainly it is a regular finding in polls, but never translates into the Lib Dem vote collapsing at election time (indeed, more often they gain support in election campaigns, though it’s not the given some assume). While Lib Dem voters might claim uncertainty to pollsters, I would be more than surprised if Lib Dem support suddenly dropped by a third over the next 6 months.

It’s also worth pointing out that while that PoliticsHome poll showed 45% of potential Lib Dem waverers might vote Labour, it also showed 33% might vote Tory. If the Lib Dems were horribly squeezed, votes could go both ways.

More interesting is actually the fate of “other” voters. While it seems implausible to expect the Lib Dem vote to drop by 5 points, a drop of three points in support for others from their current high sounds more likely. Whether these voters would shift en masse to Labour seems less likely, especially when it comes to UKIP voters.

I’m sure both Labour and the Conservatives could gather some support from Lib Dem waverers, but 5 points worth is just silly. If Labour are to close the gap with the Conservatives, it’s more likely to be because people switch back from the Tories.

On one front the poll has good news for them. It reveals a very positive party image, more people see them as caring than the other two parties, more people see them understanding the way people live their lives, being honest and principled and sharing people’s values. On other measures like being united, having a good team of leaders and being competent they trail behind the Conservatives, but are ahead of Labour.

Less good, and one reason why the positive image doesn’t translate into more votes, is that 65% of people agreed with the statement that “It doesn’t make any difference what policies the Lib Dems put forward because they have no realistic chance of ever putting them into practice, so ultimately they’ll always be just a protest vote party at national level.”. This included 37% of Lib Dem voters. 32% of people thought that “really strong distinctive policies on important issues, it will have a chance of making a big advance by gaining more MPs at the next election, because neither of the other parties is particularly appealing to most people”.

In considering why such positive perceptions of party aren’t currently translating into more support in the polls, it’s also worth looking back a bit into the past. While these are good figures and in many cases show a large positive shift when compared to the results of the same survey a year ago (particularly in being seen as honest & principled, caring about ordinary people and understanding the way people live their lives) in most cases they are still below the figures the Lib Dems recorded back in 2005 at the end of the Charles Kennedy era – 34% think they have a good team of leaders now, but 52% did in 2005; 50% think they are united now, but 71% did in 2005; 37% think they are competent now, but 46% did in 2005. The Lib Dems have a positive image, but not as positive as it was at the last election.

The presentation includes the results of questions about the effect knowing or thinking (it’s sadly not made clear exactly what the question wording was) that the Lib Dems would form a coalition with David Cameron and the Conservatives or Gordon Brown and Labour would have on people’s likelihood to vote Lib Dem. In both cases just under 4/10 people said it would make no difference.

Overall the figures were not hugely different. 29% of people would be more likely to vote Lib Dem if they were going into coalition with the Tories, 31% less likely. 30% more likely if they were headed into coalition with Labour, 25% less likely.

Broken down by party, unsurprisingly if the Lib Dems allied themselves with Labour Conservative supporters would be drastically less likely to support them – 61% would be less likely, including 38% who would be “much less likely”. There is a mirror image for Labour supporters – 59% of whom would be less likely if the Lib Dems allied themselves with the Tories. No surprises there, though it underlines the importance for the Lib Dems of maintaining a neutral stance, there are plenty of supporters of both other parties who vote tactically for the Lib Dems to keep the other one out, and they can’t afford to alienate half of them.

More interestingly, amongst current Lib Dem supporters attitudes are far more positive towards a Brown alliance than a Cameron one. 34% of current Lib Dem supporters would be more likely to vote for the party if they allied themselves with Brown, with 24% against. Only 23% were more likely to vote for the party if they allied themselves with Cameron, while 34% were against.

An important caveat is that the polling is a year out of date now, so public attitudes towards David Cameron and Gordon Brown have probably changed. The Lib Dem presentation goes on to make the sound point that even if there is a hung Parliament, the decision will probably be made by the Parliamentary arithmatic, but – back in Spring 2007 at least – it looks as though Lib Dem supporters would have been much happier to see their party supporting a Brown government rather than a Cameron one.

In ICM’s poll support for a referendum on the treaty amongst Lib Dems stands at 67%, with 33% opposed. The MORI survey found support for a referendum on EU membership was at 49% amongst Lib Dem voters, though the questions were asked in a very different way so a straight comparison is somewhat unfair.

ICM’s poll suggests only 25% of Lib Dem voters supported the idea that there should be only a one question referendum on EU membership, but 70% would prefer a two-question referendum that asked about both EU membership and the Lisbon treaty.

This appears to be a direct contradiction to the Lib Dem commissioned MORI poll, which amongst Lib Dem voters found 30% wanted a referendum on the treaty, 37% wanted a referendum on membership and 7% wanted both. The reason is the different questions asked, or more specifically, the different prompts people were given.

In ICM’s question Lib Dem voters were asked to chose between a one question referendum on just membership or a two-question referendum on both membership and the treaty. In MORI’s question people were prompted with the options of a referendum on the treaty OR a referendum on membership. While some people said both and MORI dutifully recorded it, it wasn’t actually given as an option, so many people would have assumed that had to give one of the options presented to them and that both was not up for grabs.

It would appear therefore that, while in forced choice between referendums on just the treaty or just EU membership Lib Dem voters would go for the latter, if presented with the chance to have both they’d opt for that.

Of course, this is largely a debate about the number of angels dancing on the head of a pin. Why pay so much attention to exactly what type of referendum Liberal Democrat voters would like? Especially since, as I’ve said in the past, questions on referendums nearly always show majority support – “Do you want a vote on this issue, or are you a nitwit who shouldn’t be allowed a say?”. The only real impact is if it has an affect on Lib Dem support. On that question ICM asked Lib Dem voters (or at least, 2005 Lib Dem voters…remember some of these people will already be backing other parties) how they would react if the Liberal Democrats voted against a referendum. 32% said it would make them less likely to vote for the Liberal Democrats next time round (including 10% would said they would definitely not vote Lib Dem), 7% said it would make them more likely to vote Lib Dem.

I am always highly sceptical about questions like this – we don’t know how many of those would not votes are people who wouldn’t vote Lib Dem anyway. Given this is a sample of former Lib Dem voters there shouldn’t be that many such people. The second issue is that people may be susing the question just to register their dislike of the policy and, when push comes to shove, will still vote for their party. Even with those caveats in mind, it does suggest some disquiet amongst Lib Dem supporters. The probability though is that, in reality, Europe isn’t the main thing driving the voting intention of most Lib Dem supporters – something underlined by the last question.

ICM also asked Lib Dem voters what their general attitude towards Europe was. 11% wanted more European integration, 33% supported the status quo, 46% thought Europe should have less power and 10% supported leaving the EU altogether. Obviously it would be more useful if we had the answers from supporters of other parties to put these answers in context, but the balance of opinion seems to be towarsd Euroscepticism, and it does appear that many Lib Dem supporters back the party despite having very different views on Europe.

Asked if people would prefer a referendum on the current EU treaty or membership as a whole, people much prefered a referendum on EU membership (by 38% to 18%, with 10% not wanting a referendum at all, 8% saying they’d like both and 26% don’t know).

Interestingly enough, the least supportive towards the Lib Dem policy were their own voters – they supported a membership referendum over one on just the treaty by only 37% to 30%, and only 49% supported a membership referendum per se, with 42% opposed.

Incidentally, before someone asks it is clear from the crossbreaks provided that a voting intention question was asked. Unfortunately, there are no details of a likelihood of voting question so we cannot extrapolate comparable topline voting intention figures from that. The corssbreak figures work out at Conservative 34%, Labour 44%, Lib Dem 15%, which is within a percentage point of the unfiltered voting intention figures in MORI’s January political monitor.

The only time questions are really asked about them is in Populus’s annual poll for the party conference season, which asks some questions on every party – this year 69% thought the Lib Dems were just a protest vote party, and 68% thought they “seem decent people, but their policies probably don’t really add up”. The polls occassioned by the Lib Dem leadership election underlined why polls don’t have much to say for them – invariably showing that most people didn’t have any opinion whatsoever over whether it should be Huhne or Clegg, and often had no idea who they were.

For the last year the Lib Dems have languished in the polls – though exactly how bad things were differed between pollsters. The level of Liberal Democrat support in the polls is the most variable between the pollsters, while Populus tend to be the most favourable to Labour and ComRes to the Tories the differences aren’t actually that great – everyone had the Tories up around 40% or so, everyone had Labour down to the low 30s. With the Liberal Democrats it’s different, looking at the average of the monthly polls this year from February till November – months when I have comparable figures from 4 pollsters – YouGov had the Lib Dems at 15.2%, Populus at 17%, ComRes at 17.6% and ICM at 19.4% (Ipsos MORI didn’t have their regular monitor for two of these months, but for the record the average in the remaining months was 16.1%). I posted a while back about what I thought some of the reasons might be. In recent months though even ICM have showed them falling to the mid-teens, so while it might be open to debate how bad things got, it’s pretty indisputable that things were bad.

What happened? There is an absence of polling evidence here, so I am afraid what follows is largely my own personal opinion and is far shorter than the other two posts. Having looked at Labour and the Conservatives though, I didn’t want to start the year without also looking at the Lib Dems.

The casuality of the Lib Dems’ poor performance was Sir Menzies Cambpell. To some extent he wasn’t really the problem, at least, he wasn’t a negative for the Liberal Democrats, only the an absence of a positive where they needed one to alleviate a problem that was not their own doing. Polls didn’t show anyone disliking him, and if they did show that people thought he was too old for the job, it probably wasn’t dragging the party down: a Lib Dem leader doesn’t have to be a potential Prime Minister, people know he isn’t going to be one. Ming’s problem was that he didn’t have any impact at all, and he was filling a role that needed to be carried out by someone who did.

In the last post I wrote that the Conservative advance in 2007 was largely down to Labour’s failings, rather than anything they did for themselves. That goes double for the Liberal Democrats. Lib Dems anxiously worrying about ratings and blaming their leader for not doing better should accept that they are not necessarily masters of their own fate – a certain proportion of Lib Dem votes are always going to be negative votes against the two main parties, if the reasons for the protest against the main parties receeds, so will that vote.

A large proportion of people who voted Lib Dem at the last election were people who identify themselves as Labour supporters but who voted Lib Dem either for tactical reasons or in protest over Iraq or something else that Tony Blair has done. While the former may remain, Iraq and Tony Blair as recruiting serjeants for Lib Dem protest votes have faded. Equally, whereas in 2001 and 2005 many people would have wanted to vote against Labour but would have found the Conservatives too toxic to contemplate, as David Cameron improves the Tory image the Liberal Democrats now have to share an anti-Labour vote they would once have been the obvious home for.

One strategy for the Liberal Democrats could be to try to hold back the tide, reblacken the Tory name or fight to keep Brown linked to Iraq and the Blair government. More realistically though they need to adapt to the changed circumstances. Their positioning at the last election was perfectly in tune with the political environment of the time – an unpopular government with an opposition that was distrusted – the Lib Dem slogan was “the Real Alternative”, a narrative that was values and mission free, it didn’t involve standing for anything, just not being the other two parties. It worked well for them and meant they could win on both fronts. With a detoxified Tory party it won’t chime in the same way, not least because in a more competitive election the very real alternative to a Brown government will be a Cameron one.

To prevent themselves being squeezed the Liberal Democrats need to present a new narrative that tells people what their purpose is, they need to differentiate themselves far more clearly as standing for something distinct from the other two parties. Rather than claiming to be the “real alternative”, they need to paint a coherent picture of what a “liberal alternative” is, so they can build more of a positive vote for them, making up for the inevitable loss of some of the negative vote they got last time round. Nick Clegg’s initial comments after being elected leader about Britain being a Liberal country that doesn’t yet vote Liberal perhaps points towards this sort of strategy – a view that while there are people who would vote for liberal politics, the Liberal Democrats haven’t necessarily managed to clearly identify themselves with it in people’s minds.

Nick Clegg appears to be a far more media-savvy and charismatic leader than Ming Campbell. Just by being leader he isn’t going to suddenly make the political environment any friendlier for the Liberal Democrat party, there are tough market conditions out there for them, but he does at least have the potential to be better at keeping them in the public eye and that alone would improve things somewhat. Putting forward a coherent and distinct narrative that gives people a really positive reason to vote Liberal Democrat, rather than them just being the nice people who aren’t one of the other two, will be harder – the other parties will copy popular policies and say ‘me too’ to popular values – but that’s what the Lib Dems need to do to avoid being sidelined in the first really competitive election since 1992.

Polls of party members can be tricky to do – they are hard to sample (there are no publically available lists of party members to ring up or email) and hard to weight, since no one really knows the demographic make up of the party memberships. If there is no real demographic division in support hat doesn’t matter – for example, in the last Conservative leadership election YouGov’s polls suggested that Cameron’s lead over David Davis was much the same whether members were young or old, activist or armchair supporter. The record of polls of Lib Dem members in the past is hard to judge – YouGov overestimated Chris Huhne’s support in their poll during the last contest and showed him ahead of Campbell, when in reality Campbell beat him by 13% on the first round. However, the poll was conducted several weeks before the end of the contest, so it is impossible to say whether Campbell simply picked up support and overtook Huhne in the final weeks, whether Lib Dem members on YouGov’s panel were more likely to support Huhne or whether the survey (which included several questions prior to the ones on voting intention) has biased the sample in favour of Huhne.

Either way, since many members have already voted this poll should be close enough to give an idea of the final result and – if it is representative – it looks likely that Nick Clegg will be the next Lib Dem leader.

There is also a new ICM poll out tomorrow, presumably for either the Sunday Telegraph or the News of the World. No full figures yet, but it reportedly shows an 11 point Conservative lead, with the Tories on 41% and Labour on 30%. The last ICM poll had figures of CON 37%, LAB 31%, LDEM 21%, so this would suggest a significant increase in Conservative support and a drop in Labour support.

UPDATE: Another interesting finding from the poll, given that a hung Parliament is relatively likely next time round. 44% of Lib Dem members would oppose a coalition with the Tories under any circumstances; 26% would oppose a similar deal with Labour. I’m sure some Lib Dem readers will be able to tell me, but I have the perception from coalition wranglings in Wales and Scotland that the Lib Dem leadership need the approval of Lib Dem members to actually agree to any coalition deal. In the event of David Cameron being the largest party and trying to cut a deal with the Liberal Democrats, it looks like it would be very difficult for (presumably) Nick Clegg to win Lib Dem members approval for all deal he secured.

UPDATE 2: Over at Lib Dem Voice they are pondering whether the polls’s sample is too young. YouGov’s weighted figures had 35% of Lib Dem members under the age of 40, 32% between 40 and 60 and 33% over 60. Looking at the breaks in the table under 40s were far more likely to back Clegg than over 60s – Clegg has a 32 point lead amongst under 40, but is neck and neck with Huhne amongst over 60s.

As I mentioned above, polls of party members are tricky because there are no accurate demographic figures to weight to. I don’t know where YouGov got their 35/32/33 figures from, but if they are incorrect it would skew the figures. In short, if Lib Dem members are actually older than that, the poll would have underestimated Huhne, if they are younger than that, it would have underestimated Clegg. Either way, Clegg will still be ahead, since even if the party was entirely over 60, which it isn’t, Huhne could only manage to be equal with Clegg.

UPDATE 3: The YouGov target weights were based on comparing the age of Lib Dem members with the age of the panel as a whole – effectively giving Lib Dem members an average age of 51. The only academic study of party membership is that done by Seyd & Whiteley which was well over 10 years ago (I can’t remember the exact date, it came out in several different volumes concentrating on the different parties) when they had an average age of 59. That age profile would have put Clegg 10 points ahead, rather than 12, so no great difference.

The graphs below shows the voting intention of respondents in ICM’s polls for respondents who said they voted for each of the main parties in 2005*. When I started this post I was going to look just at the Lib Dems, but actually it is as interesting to look at the others too. Hopefully it’s self explanatory which is which – white is those who told ICM their chance of voting was less than 7/10, grey the don’t knows and won’t says, purple other parties and gold, red, blue are Lib Dem, Labour and Conservative. Looking at the individual breaks on each graph can’t tell us much – the sample size is small and the figures erratic – so these are rolling averages of the three most recent polls.

Looking at those people who told ICM they voted Tory in 2005 they remain very loyal. Over 70% consistently say they would vote Tory again, what leakage there is goes mostly to don’t knows and not sure to votes – very few people who voted Tory in 2005 say they’ll vote for other parties, and when they do it has a tendency to be for “other” parties like UKIP. Even at the height of the Brown bounce few would vote for Labour. The 2005 Tory vote is solid.

Moving to Labour we should add a caveat. People who tell ICM they voted Labour in 2005 are not necessarily the same as people who actually voted Labour in 2005. Because of false recall many of those people who say they voted Labour are actually people who didn’t vote at all, or people who actually voted tactically (or as a protest) for the Lib Dems. Hence the reason why former Labour voters look less likely to vote may just be because they didn’t actually vote last time either!

Looking at how people who say they voted Labour say they will vote now we can see a substantial wedge don’t know or aren’t sure if they will. While Tony Blair was Prime Minister the proportion of Labour voters staying loyal was only just above 50% (if that sounds low, remember again that the chances are not all these people really voted Labour), with a big wedge of former Labour voters defecting to the Conservatives, Lib Dems and others. There is an obvious step change at the point when Gordon Brown takes over as PM: since Brown became Prime Minister he has attracted back a significant proportion of those disloyal Labour voters, putting Labour’s vote retention back into the region of 60%. However, having been split between Lib Dems, Conservatives and others, in the three months since then though the remaining ‘disloyal’ Labour voters have coalesced around the Conservatives.

Finally we come to the Lib Dems. Whereas with those people who said they voted Labour in 2005 probably includes people who didn’t, people tend to under-remember voting Lib Dem – this sample probably misses out many Lib Dems. Looking at the graph the Lib Dems are only retaining around half of those people who claim to have voted Lib Dem in 2005, and since Gordon Brown became leader that has fallen further, now down into the low 40s. The interesting bit is looking at the group of former Lib Dem voters now backing other parties. This was around 18% or so of 2005 Lib Dem voters when Blair was PM, and they were mostly backing the Tories. It’s grown since then to almost a quarter, but those voters have been swinging about, first backing Labour and then moving back towards the Conservatives.

So looking at these breakdowns what can we tell? Firstly past conservative voters are solid. Past Labour voters swung back to Labour after Brown became leader, and that renewed support hasn’t faded, Labour might be behind again in the polls but the past Labour supporters Brown won back haven’t gone anywhere, they are still sticking with him. Unless that changes Labour aren’t going back to the depths they experienced prior to Brown becoming PM. The movement and the recent Conservative recovery has all been amongst the floating voters who backed Labour or Lib Dem in 2005 but aren’t anymore. Those former Labour voters have gathered behind the Tories in the last few months, 2005 Lib Dem voters looking for someone else to support have backed first Labour and then the Tories.

So back to the question of what happens if the Lib Dems recover, who will they take support from? Well, amongst past Lib Dem voters far more have defected to the Tories than Labour, if people came back in direct proportion to who they are backing now it would damage the Conservatives more. It doesn’t necessarily work like that though – the people who have drifted away from the Lib Dems in the last few months may be more likely to come back, and they would appear to be more likely to be backing Labour. It may also depend up who the Lib Dems select- conventional wisdom is that Nick Clegg would be better at attracting Tory votes, Huhne better at attracting Labour votes. My own suspicion is that it probably doesn’t make much difference where the Lib Dems are placed ideologically, it will be more important how successful the new leader is at getting media coverage and once again becoming part of the mainstream media agenda, rather than being rather sidelined as the Lib Dems seemed to be under Ming Campbell.

*Not all the figures are on ICM’s tables, so I’ve assumed that the difference between the number of people who gave their likelihood of voting at 7 or above and the total of those who did give voting intentions is made up of don’t knows and won’t says. The proportions supporting each party would be much the same anyway.

Like ICM’s recent poll this suggests the Lib Dems are recovering from their awful ratings last month, presumably thanks to the publicity of their leadership contest.

What it doesn’t show is Labour closing the gap, Ben Brogan! I don’t, it has to said, have particularly high expectations of press coverage of polls, but Ben Brogan’s blog is normally one of the best, so slapped wrists Ben. Unlike all the other companies Populus have not shown the Conservatives back in the lead since the election, their methodology tends to produce figures that are slightly more favourable to Labour than other companies, and this is actually a relative advance for the Tories compared to Labour – though clearly the Lib Dems seem to gaining from both of them.

Peter Riddell’s article suggests that the polls may be returning to the sort of equilibrium they’d reached before Gordon Brown became Labour leader. I think it’s still too early to draw that conclusion, it’s possible they will, and the Conservative vote does indeed seem to be streadying at about the same level, but Labour and Lib Dem support is still on the move, and we still don’t know what the lasting effect of the Lib Dem leadership change will eventually be. I’m going to be very cautious about concluding that the polls are starting to be steady again.

UPDATE: I’ve had a chance to look at the rest of the poll’s findings. As with other recent results Populus have found a drop in perceptions of Gordon Brown. His figures here haven’t fallen off a cliff as they had in some measures on YouGov’s Sunday Times poll, but then, these haven’t tracked concepts of decisiveness which was where Brown had suffered the most. Populus have found a steady decline in the percentage of people who think Brown has what it takes to be a good PM (49%, down from 54% last month and 57% at the height of the Brown boost), those who see him as a strong leader (58% down from 60%, at the height of the Brown boost 77% thought him “strong”) and those who think he understands the problems facing ordinary people (47%, down from 49% last month and 61% during the boost). Meanwhile David Cameron’s figures are creeping upwards, 40% now think he has what it takes to be a good PM (up from 37% last month and 32% at the height of Brown’s popularity) and 42% think he is strong.

Despite the more hostile media narrative these days, these are actually all still pretty positive figures for Gordon Brown. The trend however is downwards and, while he still leads David Cameron on nearly all the measures Populus asked about, the lead is narrowing. Back in July he had a 24 point advantage in terms being seen as having what it took to be a good PM, now it’s only 9 points. Since a lot of the Labour’s increase in the polls has been Brown’s increase (satisfaction with the government increase only marginally during the handover while satisfaction with the PM rocketed), I suspect we won’t have stable figures for voting intention until perceptions of Brown have stabilised.

While looking for the past figures on leader perceptions I also found this poll Populus conducted for the Daily Politics last month on the Lib Dem leadership contest. As Mike Smithson commented yesterday, we are really flying blind on the Lib Dem leadership race – there has been no polling of Lib Dem members voting intentions, and the two candidates are so little known we can’t even really see what the wider public think of them. The Populus poll asked people if they would vote Lib Dem with Nick Clegg as leader, and if they’d vote Lib Dem with Chris Huhne as leader. 11% said they would vote Lib Dem with Clegg, 12% with Huhne – so no obvious difference. I should add that these low figures don’t represent some collapse of the vote with either man, it’s because 35% and 36% respectively said they didn’t know or hadn’t heard of them.

Several people have asked what the reason might be – including a post my Mark Park on Libdemvoice here – and which figure (if any) is right? Unfortunately these are no easy answers to who is right, but I can at least flag up some potential reasons:

Mode of questioning. ICM and Populus ask people over the phone, YouGov ask people online. In lots of cases people might give an answer to an anonymous computer screen that they would be embarrassed to give to a live interviewer on the other end of a phone line. Obvious cases are extremist parties like the BNP, but small fringe parties suffer in general because people are unsure about going out on a limb. With a mainstream party like the Lib Dems this shouldn’t be a factor…but actually it could be having a knock-on effect. Some support for the Liberal Democrats isn’t positive support for their ideas or policies, but a “neither of the above” vote. Potentially some of the higher support for the Lib Dems in phone polls could be “neither of above” voters who might really be tending towards fringe or extreme parties but are unsure about naming a fringe party to an interviewer. Given that YouGov tend to have a higher “other” score than other pollsters though, this might well be a factor.

Don’t knows. ICM and Populus both do an adjustment to their figures to take account of don’t knows. Based on past studies they assume that a proportion of don’t knows will actually end up voting for the party they did last time round. People normally think of this as an adjustment for “shy Tories”, but the net effect these days is rarely if ever to help the Conservatives. For the last couple of years it has normally helped Labour. The way it works though will help any party who finds that some of their past supporters have drifted away and are now telling pollsters they don’t know how they’ll vote. In ICM polls 50% of previous Lib Dem voters who now say “don’t know” are added onto the Lib Dem score, in Populus polls they re-allocate 30% of previous Lib Dem voters. Does this increase the level of Lib Dem support in ICM polls? Because ICM publish the figures before and after the adjustment this is one area where we can quantify the difference – some observers have taken the difference as having emerged after December 2005, so taking the 18 monthly ICM polls since then, the adjustment has been large enough to increase the Lib Dem score by 1 point six times, and has reduced it by 1 once – so on average it increases Lib Dem support by 0.28 of a percentage point, leaving another 2.72% difference to be explained by other factors.

Likelihood to vote. Uniquely amongst the pollsters YouGov do not filter or weight by likelihood to vote. If Lib Dem supporters were actually more likely to vote than supporters of other parties then this could explain some of the difference. In fact, it’s normally Tory supporters who are most likely to vote, followed by the Lib Dems with Labour supporters further behind – so perhaps this could be contributing to a lower level of Lib Dem support in YouGov polls? Again, by looking at the figures before weighting by turnout in ICM’s detailled tables we can quantify this – since December 2005 ICM have published the breakdown for likelihood to vote from 11 of their Guardian polls. If you compare what the rounded figure would have been without turnout weighting, and what it actually was afterwards, 2 times it increased the LD score by a point, 2 times it reduced it…so it has no overall effect at all. Even with MORI’s very harsh filter by turnout it makes only a minimal difference.

Sampling. Could there be a difference from the sampling techniques? Could the people who pop up in a telephone poll be more likely to be Lib Dem voters than people who join an online polling panel? In terms of political activists it probably works the other way if anything, but when it comes to normal voters who knows? It is possible, but given than both ICM and YouGov weight their polls politically it this was a problem it would be something that the pollsters should be able to address using political weighting – if YouGov’s sampling produced too few Lib Dems they would weight they upwards, if ICM’s produced too many they would weight them down (though in actual fact ICM tends to weight the proportion of part Lib Dem voters upwards, sometimes quite heavily). That bring us to…

Weighting targets. This is potentially where most of the difference lies. I suspect that when pollsters weight their polls politically ICM and Populus are weighting their sample to have a slightly higher proportion of past Lib Dem voters than YouGov are. Unfortunately it is impossible to directly compare the proportions used because ICM and Populus weight using recalled past vote and YouGov weight using party identification.

The data used for political weighting. This is the most subtle difference, and the one that I suspect is behind a fair amount of the difference. Phone polls do their political weighting based on data that is collected now. They then have to take account of false recall and forgetfulness when drawing up weighting targets. In contrast YouGov weight their polls using the data they collected back in May 2005 when it was fresh in respondents minds who they had actually voted for that day. The people who voted Lib Dem in 2005 but who don’t recall or say they voted for a different party if you ask them now are, perhaps not surprisingly, also far less likely to say they would vote Lib Dem in a general election tomorrow. In other words, the past Lib Dem voters that phone pollsters find are the more committed Liberal Democrats. “Flakier” Liberal Democrats who are more likely to switch to other parties are more likely to have forgotten they voted Lib Dem in the first place. Those identified as Lib Dems in YouGov’s samples probably contain more of those “flaky” Lib Dems than those identified as Lib Dems in phone samples.

What is the ultimate reason for the difference? I don’t know. It looks as though we can rule out likelihood to vote weighting and we can see that the “spiral of silence adjustment” is only a very small factor. The mode of questioning may be having some effect – people who say “Liberal Democrat” to a live interviewer might be willing to admit to a computer screen that they will actually vote for a smaller fringe party. In my opinion the difference is more likely to be somewhere in the weighting, and here is it almost impossible to draw any conclusions – ICM and YouGov weight using a different measure, based on data collected at different times and, as with all political weighting, chosing the targets they weight to is as much an art as a science.

Who’s right? It is impossible to say. If the reason had turned out to be something to do with likelihood to vote then you could have a rationial discussion about to what extent and how turnout should be factored in, ditto the way don’t knows are dealt with. It really is very difficult to make informed judgements about weighting. I am not one who believes that is any accurate way of telling how well parties are doing apart from polls: people vote differently in general elections to local elections; even if they were a decent guide, we must be approaching a record period without a Parliamentary by-election; local by-elections don’t even seem to be a good guide to local elections these days, let alone anything else! The only reliable way to tell which poll is producing more accurate results will be to wait until the next general election and see what the actual results are. Sadly, they doesn’t help you much in deciding who is right now.

I’ll give you two warnings. Firstly, it isn’t always the poll that is different from the rest that is wrong. In 1992 Harris was showing a Tory lead when everyone else had Labour ahead. People dismissed Harris as being wrong, the rest is history. Secondly, it is very tempting to believe the poll you want to believe – to see fault in every detail of the methodology of the poll who you really hope isn’t right and the obvious superiority of the poll you hope is true. In few if any cases do I think that people are deliberately talking up polls that favour their own party. I just think that somewhere deep in our subconscious we all tend to will ourselves into finding arguments in favour of the methodology that produces the results we like more convincing