Polling Observatory #1: Estimating support for the parties (with some trepidation…)

DipticBy The Polling Observatory (Robert FordWill JenningsMark Pickup and Christopher Wlezien). The homepage of The Polling Observatory can be found here, and you can read more posts by The Polling Observatory here.


This post is part of a long-standing series (dating to before the 2010 election) that reports on the state of the parties as measured by vote intention polls. By pooling together all the available polling evidence we can reduce the impact of the random variation each individual survey inevitably produces. Most of the short term advances and setbacks in party polling fortunes are nothing more than random noise; the underlying trends – in which we are interested and which best assess the parties’ standings – are relatively stable and little influenced by day-to-day events. Further details of the method we use to build our estimates can be found here.

It is now six months since the television headlines rolled at 10am on May 7th, with the exit poll dropping the bombshell that the polls had got it badly wrong. The election forecasters fared little better, including us: even though our vote model had predicted a Conservative lead of 2-3 points, our seat prediction was nowhere close to the majority achieved by David Cameron. It is with a little trepidation then that the Polling Observatory team returns to provide its assessment on the state of public opinion in late 2015.

As regular readers will know, we pool all the information that we have from current polling to estimate the underlying trend in public opinion, controlling for random noise in the polls. Our method controls for systematic differences between pollsters – the propensity for some pollsters to produce estimates that are higher/lower on average for a particular party than other pollsters. While we can estimate how one pollster systematically differs from another, we have no way of assessing which is closer to the truth.

One possibility with this method is to use the result of the last election to ‘anchor’ our estimates of bias in the polls against the last election result. This treats the election result as if it was produced by a pollster with no systematic error. We can then estimate the systematic difference of each pollster with this hypothetical perfect pollster. With this method, for example, if pollster X produces results which are systematically 2 percentage points higher for the Conservatives than what would be produced by this perfect pollster, we would interpret a poll indicating 40% support for the Conservatives from such a pollster as 38% support for the Conservatives. This approach can be useful where there are recurring historical patterns (such as the tendency of the polls to overestimate the Labour vote and underestimate the Conservative vote), and might allow us to control for systematic bias in the polls.

We have chosen, for now, to anchor our estimates on the average pollster. This means the results presented here are those of a hypothetical pollster that, on average, falls in the middle of the pack.[1] We have chosen to use such a middle pollster rather than anchor on the election result because we believe that the inaccuracies/biases revealed in the polls in May will be different from those which may occur in this election cycle.[2] All of the pollsters have been undertaking reviews of their methods following the big polling miss in May, and it is unlikely that the biases in polling will be unaffected by the changes they are gradually introducing. Because of this, we offer our estimates of party support with an important caveat: while our method accounts for the uncertainty due to random fluctuation in the polls and for differences between polling houses, we cannot be sure that there is no systematic bias in the average polling house (i.e., the industry as a whole could be wrong once again). It may be that the polls are collectively right or wrong. It may also be that a pollster producing figures higher or lower than the average is more accurately reflecting the state of support for the parties than their competitors. Our estimates cannot adjudicate on whether figures on the high or the low side for a party better reflect the underlying preference of the electorate. The only test is on Election Day. Fortunately, none of this prevents us from identifying and reporting on the underlying trends over time.

In terms of the overall story, there has been little apparent change in vote preferences since the election in May. This despite the triumphant budget announced by George Osborne, the surprise ascension of Jeremy Corbyn to leader of the Labour Party (and the onslaught on him and his team from outside and inside the party), and the tax credits row that has quickly taken the shine off the government’s honeymoon period. Unlike the last election, there has been no sudden flight of voters from one party to another, as occurred with the collapse of Liberal Democrat support in the first six months after the Coalition government was formed.

Our estimates suggest that Conservative support has slipped slightly since the heady days of May and June, from around 40% to closer to 37% at the start of November. Despite Labour being divided and in some disarray over its direction, it has made slight gains from around 30% to 32%. This upward drift in the polls largely occurred before election of Jeremy Corbyn as Labour leader, so cannot be attributed to a Corbyn effect. Whether these gains will persist as the election nears and PM Corbyn becomes a possibility, is of course open to debate. At present, though, there is no sign of Mr Corbyn’s election having any impact on his party’s overall support. UKIP support has remained steady at around 13%, and the party shows no signs of going away – even with its own internal conflicts following Nigel Farage’s “unresignation” in the summer. Lagging somewhat behind, the Liberal Democrats continue to flat-line at just under 7%. One of the patterns of the last parliament was the stubborn immovability of Liberal Democrat support. New party leader Tim Farron has much work on his hands to win back voters, and so far there are no green shoots for the party in our estimates. Finally, speaking of the Greens, their support appears to have been squeezed since Labour election Jeremy Corbyn – perhaps because voters attracted by their distinct left wing platform now feel more at home in the Labour party. It has fallen around 1.5 points since the summer. Our estimates for all the parties suggest that the electorate is still to make up its mind on both the new government and the fragmented and much changed opposition. But there are some big events on the horizon, in particular the EU referendum, which may yet provide a shock to move political support in one direction or the other.

UK 01-11-15 anchor on average (1)

One of the reasons why the polling miss back in May came as such a shock was that by election eve there was broad consensus among the pollsters about the level of support for the parties (though of course we noted house effects earlier in the campaign). However, in the period since May the polling has been characterised by much more variation in the standing of the parties. This is revealed in the figure above. The size of the confidence intervals for our estimates in the period since the election (an average of 2.3 points) are more than twice those for the 2010-15 election cycle or for the month just before the start of the short campaign (each an average of 1.1 points). This indicates a much higher level of uncertainty about the state of public opinion today. Part of this could be due to a lower volume of polling since May, or more variation in polling methodologies as pollsters take different approaches in response to May’s polling miss. The greater uncertainty may also reflect the much lower frequency of polling since the election – election watchers used to multiple daily polls have now to accept a more meagre diet of one or two polls a week. The greater uncertainty may, however, also reflect something more fundamental: genuine uncertainty, and hence greater volatility, in the minds of the electorate. Voters are faced with an unexpected Conservative majority government and an unfamiliar and polarising opposition leader attracting widely varying reactions in the media and within his own party. In such circumstances many may be genuinely unsure as to their preferences. Only time will tell whether this uncertainty lasts until the next general election. For now, it provides an important reminder of the need to take single poll results with a degree of caution.

 

Robert FordWill JenningsMark Pickup and Christopher Wlezien

 

[1] The average difference between this middle pollster and those pollsters that produce estimates that are systematically higher for a given party is the same as the average difference between this middle pollster and those pollsters that produce estimates that are systematically lower for that same party.

[2] We came to a similar conclusion during the last election cycle when it became apparent that our method of anchoring on the election result was excessively reducing the estimated level of support for the Liberal Democrats.

 

Polling Observatory Latest #GE2015 Forecast: the Conservatives make slight gains, but the likeliest result is deadlock

DipticBy The Polling Observatory (Robert FordWill JenningsMark Pickup and Christopher Wlezien). The homepage of The Polling Observatory can be found here, and you can read more posts by The Polling Observatory here.


As we enter the closing stretch of the campaign, substantial uncertainty remains about the final outcome. Taking out the random noise, the polls are still showing a close race ahead of May 7th. Some have pointed to differences between telephone and internet pollsters, with the former having shown a steady, if slight, Conservative lead all year. Our method allows us to control for systematic differences between polling houses and variation in the ‘poll of polls’ that is due to changes in the mix of pollsters in the field at a given point in time.

The latest Polling Observatory forecast covers all polls completed up until April 30th, and shows support for the two main parties is still in the balance – with Labour on 33.1% and the Conservatives on 34.2% — though the confidence intervals are such that we cannot say for certain that the Conservative lead is greater than zero.

Our vote forecast points to a higher level of support for the Conservatives than two weeks ago, up 1.4 points at 35.0%, with Labour on 32.6%, up 0.1 points. This reflects the squeeze that the “big two” have put on other parties in the final weeks of the campaign. The Conservative lead now stands at 2.4%, but with considerable uncertainty remaining in our forecast.

Forecast 01-05-15

This slight shift in the balance of polling is reflected in our latest seat estimates. The Conservatives’ median estimate rises by six seats, Labour falls by six seats, and the Liberal Democrats fall by four.  This puts the median Conservative seat lead at just two. However, as the confidence intervals attached to our estimates reveal, this projected lead is highly uncertain, a veritable coin-flip, with a 53 per cent chance that the Conservatives will have more seats than Labour. A majority for either is at present very unlikely, e.g., the likelihood of a Conservative majority is tiny (less than 0.2%). Our estimates further reflect the gains made by the SNP in recent polling in Scotland, with the nationalists now forecast to win 54 out of 59 seats north of the border.

Table 1: Seat estimates, with confidence intervals and change on April 15th

Party March 1st estimate April 1st estimate April 15th estimate April 30th estimate
Conservative 265 271 268 274 (+6)

(251,305)

Labour 285 276 278 272 (-6)

(244, 295)

Liberal Democrat 24 27 28 24 (-4)

(18, 29)

UKIP 3 3 3 2 (-1)

(1, 4)

SNP 49 49 49 54 (+5)

(46, 58)

Others 6 6 6 6

(4, 8)

Northern Ireland (not forecast) 18 18 18 18

The Conservatives’ paths to a governing coalition are even more winding than their slight lead in votes and seats. They cannot reach a majority with the backing of the Liberal Democrats (combined 298 seats, 15 short of a majority) or with both the Liberal Democrats and the Northern Irish DUP (combined 306 seats, assuming the DUP once again win 8 seats), or even by adding UKIP to that two party combination (308 seats total). It would be very hard, with this seat outcome, for the Conservatives to sustain a government without some form of acquiescence from the SNP. Things are rather more promising for Labour.  While they cannot reach a majority with the help of the Liberal Democrats (combined 300 seats), they can with SNP.  Whether that happens remains to be seen, of course.

Table 2: Most plausible governing combinations, based on March and April seat forecasts

Party March 1st estimate April 1st estimate April 15th estimate April 30th estimate
Conservatives + Lib Dems + DUP 298 307 305 306
Conservatives + Lib Dems + DUP + UKIP 301 310 308 308
Labour + SNP 334 325 327 326
Labour + Lib Dem 309 303 306 300
Labour + SDLP + Plaid Cymru + Green + Lib Dem 316 310 313 307
Labour + Lib Dem + SNP 358 352 355 354
Labour + SDLP + Plaid Cymru + Green + Lib Dem + SNP 365 359 362 361

Our projected numbers suggest that while the ballots may all have been counted by May 8th, the shape of the new government may be up in the air for some time after.

Update: we have mow updated our forecast with all polls up to the end of Tuesday 5th May, giving the final Polling Observatory forecast for this parliament:

Conservatives 34.5% (32.6, 36.4)

Labour 32.4% (29.7, 35.2)

Liberal Democrats 8.7% (6.9, 10.6)

In terms of seats, this translates into:

Labour 273 (246, 295)

Conservatives 271  (248, 299)

Liberal Democrats 24 (19, 28)

Scottish National Party 55 (49, 59)

Ukip 2 (1, 4)

Other 6 (4, 9)

Robert FordWill JenningsMark Pickup and Christopher Wlezien

Polling Observatory analysis cited in OfCom’s statement on party election broadcasts

DipticBy The Polling Observatory (Robert FordWill JenningsMark Pickup and Christopher Wlezien). The homepage of The Polling Observatory can be found here, and you can read more posts by The Polling Observatory here.


Regular readers of the blog might be interested to know that our Polling Observatory analysis of support for the parties (a joint venture between the Universities of Southampton and Manchester, Simon Fraser University and the University of Texas at Austin) featured today in OfCom’s statement on party election broadcasts. You can read the full OfCom report, ‘Review of Ofcom list of major political parties for elections taking place on 7 May 2015’, here.

Robert FordWill JenningsMark Pickup and Christopher Wlezien

Polling Observatory #45: Parties neck-and-neck as we approach the formal campaign

DipticBy The Polling Observatory (Robert FordWill JenningsMark Pickup and Christopher Wlezien). The homepage of The Polling Observatory can be found here, and you can read more posts by The Polling Observatory here.


This is the forty-fifth in a series of posts that report on the state of the parties as measured by opinion polls. By pooling together all the available polling evidence we can reduce the impact of the random variation each individual survey inevitably produces. Most of the short term advances and setbacks in party polling fortunes are nothing more than random noise; the underlying trends – in which we are interested and which best assess the parties’ standings – are relatively stable and little influenced by day-to-day events. If there can ever be a definitive assessment of the parties’ standings, this is it. Further details of the method we use to build our estimates of public opinion can be found here.

UK 01-03-15 anchor on average

As we approach the electoral business-end of this parliament, the party machines are gearing up for the campaign proper – with a lot of nervous watching of the polls for signs of a sea change in opinion. Are the public steady in their voting intentions, or are they starting to move as polling day nears? Our latest report tracks the polls up to March 1st, so just before the eruption of ‘debate-gate’ arguments about the Prime Minister’s reluctance to appear in televised debates, and before the most recent run of polls that look potentially more favourable for the Conservatives, though as ever we reserve judgement until the underlying trend is clear.

Our evidence suggests that the parties remain neck-and-neck, with Labour support at 32.2% the same as last month, and the same as our estimate at the start of December too. The most recent erosion in Labour’s lead, which has now all but disappeared, has not been due to a decline in Labour support, as was true for most of last year, but instead due to  a slight improvement in Conservative fortunes. The Tories have gained 0.3 points this month, rising to 31.5%, one of their best performances in the past year or two. However, despite recent gains, we still have them in the 30 to 32% rut that they have been in since the infamous “omnishambles” budget in 2012. If the more recent uptick in support seen in March’s polling to date is sustained, the Conservatives might break out of the rut just as the election finish line approaches. This remains to be seen.

UKIP have fallen back again, down 0.7 points to 14.8%, their lowest share since September. While not quite at the peak they achieved in the autumn, possibly due to reduced attention over the past month, Farage’s radical right outsiders have become part of the political landscape and continue to attract enough support to threaten to wreak havoc in May. The Liberal Democrats have made a slight gain, rising 0.4 points to 8.4%, but still look destined for severe electoral pain. One much-caveated forecast based on data from the British Election Study suggesting they could win just one seat. Their fate in Parliament continues to hang on the ability of their MPs to generate ‘personal votes’ sufficient to swim against the tide of national unpopularity.

Finally, the Green surge appears to have levelled off for now. We have the party up  just 0.1 point past this month, to 6.4%. This is a new record, and puts support for the party just below the Lib Dems, but  the questions remain how ‘soft’ much of the Green vote is as Election Day approaches, and how effective the Greens will be in channelling rising national popularity into successful local campaigns.

Based on these current estimates of support for the parties, our vote forecast points towards a result in May that is too close to call. We put both Labour and the Tories on 33.7% but with a wide range of uncertainty.  Our forecast indicates that Labour support could fall within the 30.1% to 37.2% range and Conservative support in a slightly narrower range — between 31.9% and 34.4%. In short, history suggests there is still time for either Labour or the Conservatives to pull ahead, though neither has a clear advantage right now. The Liberal Democrat forecast continues to edge downwards as the long hoped for recovery in support continues to elude them. It now stands at 8.8%, which would be just over a third of the vote that they won in May 2010.

Vote Forecast 01-03-15

As the four party politics of 2014 has given rise to five or six party politics, and as UKIP and the SNP appear to have become part of the electoral landscape that has proved resilient, the Polling Observatory team has produced its first set of forecasts at the constituency level. On Friday, 13 March, 2015, we will release these seat level forecasts, along with an explanation of how our model of constituency simulations works, and  estimates of each party’s chances in every seat.

Robert FordWill JenningsMark Pickup and Christopher Wlezien

Polling Observatory #44: Race continues to narrow with less than 100 days left

DipticBy The Polling Observatory (Robert FordWill JenningsMark Pickup and Christopher Wlezien). The homepage of The Polling Observatory can be found here, and you can read more posts by The Polling Observatory here.


This is the forty-fourth in a series of posts that report on the state of the parties as measured by opinion polls. By pooling together all the available polling evidence we can reduce the impact of the random variation each individual survey inevitably produces. Most of the short term advances and setbacks in party polling fortunes are nothing more than random noise; the underlying trends – in which we are interested and which best assess the parties’ standings – are relatively stable and little influenced by day-to-day events. If there can ever be a definitive assessment of the parties’ standings, this is it. Further details of the method we use to build our estimates of public opinion can be found here.

UK 01-02-15 anchor on average

Politics has returned with a vengeance after a brief festive cease-fire. All the parties have moved into campaign mode with just a few months left until election day, and news schedules are now daily filled with briefings, policy announcements, and the wearying trench warfare of spin, with hyperbolic claims and counter-claims hurled back and forth. What have the voters made of it all? Our last report tracked polling up to the beginning of December, so the latest estimates from the Polling Observatory cover two months: the December lull as well as the January campaigning.

So far, there is little evidence of any decisive impact on the overall balance of power. Labour support has been stable for the past two months – our February 1st estimate of 32.2% is exactly the same as our estimate for December 1st. There has been a great deal written in the media about an alleged slump in Labour fortunes, but this story seems to be behind the curve of polling – Labour support did experience a substantial drop in the early autumn, but has been stable since. Given that much of the autumn fall in Labour support seems to be concentrated in Scotland, where the recent wave of constituency polls by Lord Ashcroft has confirmed a massive swing from Labour to the SNP, it is possible that overall support for Ed Miliband’s party has not declined at all in England and Wales in the past year.

Conservative support has picked up a bit over the past two months, but most of the gain merely recovered the ground lost in November. We now have them at 31.2%, one point behind Labour, and still stuck in the 30 to 32% band where they been marooned for almost three years. The Conservatives will hope that their positive economic message, and David Cameron’s sizeable ratings advantage over Ed Miliband, will start to deliver polling gains as the election approaches, but as yet we see little sign of this.

UKIP received less attention over the winter than they enjoyed during their vintage autumn, crowned by two by-election victories. This may explain the slight dip in their poll ratings, down 0.7 points to 15.5%. Both of the main parties will hope that Farage’s insurgents will be squeezed in a more sustained way as election day approaches, but there is no evidence of this yet – 15.5% remains above the highest ratings the party received before 2014.

The Liberal Democrats slid once again over the past two months – down 0.5% to 8.0%, a record low on our revised methodology. The party’s famed constituency campaign organisations become ever more vital to its prospects one election day as its national poll numbers continue to flatline.

The other big political story of the past two months has been the “Green surge”, with support for the environmentalists soaring, particularly among disaffected younger voters, and pushing the Liberal Democrats into fifth in some polls. This month we have for the first time added estimates for the Greens. Our systematic inspection of the polling evidence does not support the narrative of a “surge” concentrated in the past few months, which seems to be the result of selective analysis of the most favourable polls. Instead, we find that support for the Greens has been steadily increasing for about a year, and – at 6.3% – is now more than double the level recorded at the beginning of 2014. As so often in this turbulent election cycle, the true impact of the Greens’ rise is hard to gauge at present – while there is a sustained and genuine shift towards them, it is concentrated among the segment of the electorate (under 25s) that is least likely to vote, and also most likely to be adversely affected by new voter registration rules. In addition, the Greens’ organisation is relatively weak, and so there remain doubts about whether the party has the capacity to mobilise and turn out its new support base. Despite the conventional wisdom that the Greens are hurting Labour, it is striking that the rise in Green support over the past few months has much more closely mirrored the (continued) decline in Liberal Democrat support.

With the four party politics of 2014 now giving way to five or six party politics, it is becoming ever more important to consider May 2015 at the constituency level. We have been working hard on developing our constituency level prediction model over the past couple of months, and we will very shortly unveil our seat level forecasts for the election. These will then be updated regularly along with the polling estimates.

Robert FordWill JenningsMark Pickup and Christopher Wlezien

Polling Observatory #43: Stability returns with race close to dead heat

DipticBy The Polling Observatory (Robert FordWill JenningsMark Pickup and Christopher Wlezien). The homepage of The Polling Observatory can be found here, and you can read more posts by The Polling Observatory here.


This is the forty-third in a series of posts that report on the state of the parties as measured by opinion polls. By pooling together all the available polling evidence we can reduce the impact of the random variation each individual survey inevitably produces. Most of the short term advances and setbacks in party polling fortunes are nothing more than random noise; the underlying trends – in which we are interested and which best assess the parties’ standings – are relatively stable and little influenced by day-to-day events. If there can ever be a definitive assessment of the parties’ standings, this is it. Further details of the method we use to build our estimates of public opinion can be found here.

UK 01-12-14 anchor on average

After the storm, the calm. Last month we found one of the largest shifts in opinion we have recorded since 2010, as Labour support plunged nearly three points in a few weeks. This month things have been more settled. Labour have recovered slightly, rising 0.6 points to 32.2%. The Conservatives fell back a little in November, down 0.7 points to 30.0%. As a result, Labour’s lead, which had fallen to less than one percentage point at the end of October, has recovered to a still anaemic 2.2% as the Christmas break approaches. Labour will take solace from the fact that their autumn slump has halted, and that the Conservatives’ three year failure to recruit new support has continued for yet another month. The party seems never to have recovered from the damage done to its reputation by the omnishambles budget in the spring of 2012. However, a lead of two points remains awfully precarious, and as we saw in October all that Cameron’s party need is one strong month to pull ahead in the polls. Both parties will go into their Christmas break with reasons to hope, and plenty to worry about.

UKIP had another strong month in November, with Mark Reckless, their second defection from the Conservatives, comfortably elected in UKIP colours in a seat without a demographic pro-UKIP lean. The sustained upward trend in UKIP support continues for another month, as Farage’s insurgents rise to 16.2%, a new record, up one point on last month. The pollsters have now arrived at a clearer consensus on UKIP support, reflected in the narrower “confidence interval” in our estimate, shown by the dashed lines. Farage and his colleagues will certainly be among the nation’s most confident politicians going into Christmas break. A year ago, many doubted that the party could convert their rising support into Westminster seats. No longer. Now the questions under heated discussion at political Christmas parties will be: “how many seats? Where? From whom?” The party can take great pride in its achievements to date, but longer term challenges remain. Even if it were to win 10 seats, the top end of most expectations, that would see 15% of the vote converted to less than 2% of the elected parliamentary intake.

The national polling provides little Christmas cheer for the Lib Dems. We have them at 8.5% this month, the same as last month. Their struggles to hold off the challenge for fourth place from the Greens continue, though as yet we do not have an estimate of Green support. The main source of solace for Clegg’s party comes from the Ashcroft constituency polling, which shows many Lib Dem incumbents in a much stronger position than national polling suggests, although Clegg himself seems to be struggling to hold off a Labour challenge in his Sheffield Hallam seat.

This month we can also bring you an update on our national polling forecast figure. We didn’t publish a forecast last month, so the changes reported are on the figures from two months ago. Labour’s decline in the polls over that period also is reflected in our forecast, though we do anticipate some recovery from the current level. We forecast a share of 33.4% for Labour next May, representing a fall of 2.8 points over the past two months. The Conservatives’ forecast share has not risen, however – we have them winning 33.8%, up just 0.1% from October’s forecast. The Liberal Democrats are also expected to recover somewhat based on historical trends in the polls. Our current forecast is for 9.2%, up 0.5% on the previous estimate. As before, we do not make a direct forecast of UKIP support as our forecasting method is based on historical polling trends, and there is not sufficient data to apply this method to UKIP support. Forecast 01-12-14

The current polling and the forecast both point to a near dead-heat between the top two parties. Yet neither may be particularly reliable anymore as an indicator of how the next Parliament will look. British politics has never been more fragmented, and that fragmentation means geography and constituency context  could be decisive. Surging support for the SNP, UKIP and the Greens is impossible to understand without focussing on the constituency battles where these parties will look to convert votes into seats, while the fate of the Liberal Democrats will turn on whether their legendary local campaigning skills can still deliver in a Siberian climate for the national party. We will shortly unveil our seat-level forecasting model, which attempts to capture some of this variation from seat to seat and produce a more accurate assessment of how the fragmented national political competition will play out in the hundreds of local contests which will decide next year’s outcome. In the meantime, we will continue to keep our usual close watch on the polls, as the closest political contest in a generation enters its final stages.

Robert FordWill JenningsMark Pickup and Christopher Wlezien

Polling Observatory #42: Sharp drop in Labour support adds further confusion to the most chaotic election in living memory

DipticBy The Polling Observatory (Robert FordWill JenningsMark Pickup and Christopher Wlezien). The homepage of The Polling Observatory can be found here, and you can read more posts by The Polling Observatory here.


This is the forty-second in a series of posts that report on the state of the parties as measured by opinion polls. By pooling together all the available polling evidence we can reduce the impact of the random variation each individual survey inevitably produces. Most of the short term advances and setbacks in party polling fortunes are nothing more than random noise; the underlying trends – in which we are interested and which best assess the parties’ standings – are relatively stable and little influenced by day-to-day events. If there can ever be a definitive assessment of the parties’ standings, this is it. Further details of the method we use to build our estimates of public opinion can be found here.

UK 01-11-14 anchor on average

In last month’s Polling Observatory we noted remarkable stability in the polls despite a hugely eventful political month. This month we find the opposite pattern. A relatively subdued political month has been accompanied by one of the largest shifts in opinion we have observed since the beginning of this parliament. Labour’s vote share, at 31.6%, is down 2.8 points in just a month, erasing nearly all of the fragile lead over the Conservatives that the party have been clinging to over the past six months. This plunge in support is among  the largest shifts in opinion we have recorded since 2010, and implies that about one in twelve Labour voters has drifted away from the party in the past few weeks. A significant part of this drop may be the result of the seismic shift in opinion north of the border, where support for the Scottish National Party has surged dramatically, threatening Labour’s long hegemony in Scottish Westminster polls and votes. The Conservatives, by contrast, have recovered some of the ground they lost last month, rising 0.6 points to 30.7%. The top two parties are now within just a single percentage point of each other, pointing to a tightening race with just six months to go to the general election.

Meanwhile UKIP have consolidated their large gain last month, and are stable on 15.2%. However, as the widening dashed lines around our latest UKIP estimate indicate, there is an unusually high degree of uncertainty about UKIP support at the moment. This reflects the substantial spread in UKIP support in the polls. Some pollsters are showing the party at 20% or higher, and indeed in the aftermath of the Clacton by-election one Survation poll reported UKIP as high as 25% (and another ComRes poll put them on 24% around the same period). In contrast, other pollsters have them stable in the mid-teens, for example with both Populus and YouGov often finding UKIP support in the 13% to 15% range.

As we have discussed previously, our method makes it possible to estimate the ‘house effect’ for each polling company for each party, relative to the vote intention figures we expect from the average pollster. That is, it tells us simply whether the reported vote intention for a given pollster is above or below the industry average. This does not indicate ‘accuracy’, since there is no election to benchmark the accuracy of the polls against. It could be, in fact, that pollsters at one end of the extreme or the other are giving a more accurate picture of voters’ intentions. In the table below we report the ‘bias’ towards or against each of the parties for all current polling companies. From this, it is quickly apparent that the largest range of house effects are found in the estimation of UKIP support, with Survation’s figures 4.3 points higher than the average pollster, followed by Opinium at 2.8 points higher and Lord Ashcroft at 2.1 points higher. In contrast, ICM put UKIP 2.6 points lower, ComRes (telephone) 2.5 points lower and Ipsos-MORI 1.8 points lower. As when we reported on this previously, the uncertainty seems to be associated with the method a pollster employs to field a survey. All the companies who poll by telephone (except Lord Ashcroft’s weekly poll) tend to give lower scores to UKIP. By contrast, three of the five companies which poll using internet panels give higher than average estimates for UKIP. The diversity of estimates indicates the continued uncertainty about the extent to which UKIP is reshaping the political landscape at the present time, where the lack of a clear precedent means that pollsters have little previous information to use to calibrate their estimates.

With the top two parties effectively tied, and the pollsters divided about the performance of the surging insurgents who may decide their fates, the outcome of the 2015 election has never been less certain.

House Mode Adjustment Prompt Con Lab Lib Dem UKIP
ICM Telephone Past vote, likelihood to vote UKIP prompted if ‘other’ 1.3 -0.8 2.7 -2.6
Ipsos-MORI Telephone Likelihood (certain) to vote Unprompted 0.5 0.3 0.5 -1.8
Lord Ashcroft Telephone Likelihood to vote, past vote (2010) UKIP prompted if ‘other’ -0.9 -0.6 -1.0 2.1
ComRes (1) Telephone Past vote, squeeze, party identification UKIP prompted if ‘other’ 0.3 -0.1 0.1 -2.5
ComRes (2) Internet Past vote, squeeze, party identification UKIP prompted if ‘other’ 0.3 -0.7 -1.0 1.8
YouGov Internet Newspaper readership, party identification (2010) UKIP prompted if ‘other’ 1.9 2.1 -1.3 -0.4
Opinium Internet Likelihood to vote UKIP prompted if ‘other’ -0.8 -0.8 -2.2 2.8
Survation Internet Likelihood to vote, past vote (2010) UKIP prompted -1.7 -1.4 -0.3 4.3
Populus Internet Likelihood to vote, party identification (2010) UKIP prompted if ‘other’ 2.4 1.7 0.1 -2.1

Another polling sub-plot which emerged this past month has been the emergence of the Greens as yet another potent force in the fragmenting political landscape. A number of polls have put their support at 6% or 7%, a massive increase on their 2010 showing of less than 1% (though this is artificially deflated as the party stood candidates in less than 50% of constituencies), and close to or even above the struggling Liberal Democrats. We do not currently estimate support for the Greens, but will investigate adding them to our model if the current surge in support is sustained. Currently, we have the Lib Dems at 8.5%, up 0.3 points on last month. Although falling behind the Greens is symbolically bad for the party, and provides seasoned poll watchers with an exciting new story, the substantive impact of this new twist is likely to be limited. The Lib Dems’ fate still depends on how well they can hold on to votes in their traditional constituency strongholds.

Despite the sharp fall in their support this month, Labour still hold important structural advantages thanks to the biases in the electoral system, which mean their votes translate more effectively into seats. Lord Ashcroft has been doing the psephological community a huge service by systematically polling the individual marginal seats which will decide the result next year. Using this method, he has already identified enough prospective Labour gains to put the party ahead on seats next year, and he still has many more strong prospects to poll. However, as Lord Ashcroft’s himself wisely reminds us, polls are a snapshot not a prediction, and Labour’s leads in many key marginals look awfully fragile. The opposition remains, slightly, in front for one more month. But there are six more to go, and if just one of these looks like October, the contest will be thrown wide open.

Robert FordWill JenningsMark Pickup and Christopher Wlezien