Category Archives: Press Releases

Accuracy of the Final 2010 Polls

8th May 2010

The table below compares the final estimates of the outcome of the General Election made by companies that are members of the British Polling Council (BPC) with the actual result across Great Britain as a whole.

While not proving as accurate as the 2005 polls, which were the most accurate predictions ever made of the outcome of a British general election, the polls nevertheless told the main story of the 2010 election — that the Conservatives had established a clear lead. All but one of the nine pollsters came within 2% of the Conservative share, and five were within 1%.

The tendency at past elections for polls to overestimate Labour came to an abrupt end, with every pollster underestimating the Labour share of the vote, though all but one were within 3%. However, every pollster overestimated the Liberal Democrat share of the vote.

Con Lab LibDem Other Average Error
% % % % %
Angus Reid 36 24 29 11 3.25
Com Res 37 28 28 7 2.25
Harris 35 29 27 10 1.5
ICM 36 28 26 10 1.25
Ipsos MORI 36 29 27 8 1.75
Opinium 35 27 26 12 2.25
Populus 37 28 27 8 1.75
TNS BMRB 33 27 29 11 3.25
YouGov 35 28 28 9 2.25
Actual Result 37 30 24 10

NOTE. The table includes the final poll conducted by each company where that poll was conducted either wholly or partly on or after Monday 3rd May. Average error is the average of the difference between the poll result and the actual result across all four estimates.

Further information:-

  • Nick Moon (GfK NOP) 020 7890 9830
  • Andrew Cooper (Populus) 020 7253 9465

Polls Commit To Speedy Disclosure In 2010 Election

2nd April 2010

Britain’s opinion pollsters announce today that they have committed themselves to releasing speedily the technical details of all voting intention polls published during the course of the general election campaign. In the event of a May 6th election, members of the British Polling Council, who between them comprise nearly all the companies who undertake opinion polls in the UK, have agreed they will place the key details of all polls published on or after Thursday 8th April on their web sites within at most 18 hours of their initial publication. These details will include:

The precise wording of the questions asked

The sample size, how and when interviews were conducted, and how the data have been weighted to ensure they are representative.

Computer tables showing the breakdown of the results for key demographic groups.

John Curtice, President of the British Polling Council, said, ‘Opinion polls often play an important role in shaping the mood and rhythm of an election campaign, sometimes controversially so. It is therefore important that they should be as open and transparent as possible about how their results have been obtained. This commitment will ensure that anyone who is interested in or has concerns about a particular poll will be able to ascertain quickly and easily for themselves just exactly how it was conducted.’

Notes to Editors.

  1. The British Polling Council (BPC) was founded in 2004 with the primary object of promoting the disclosure of the technical details of the results of published opinion polls so that consumers can be make an informed judgement as to the reliability and validity of their results. Further details can be found at www.britishpollingcouncil.org
  2. Under its rules of disclosure, BPC members are usually required to make the technical details of polls available within 2 working days of initial publication. This new 18 hour commitment, which will operate during the 4 weeks immediately prior to polling day, will ensure these details are available more quickly at a time when polls are the subject of particular media interest and commentary.
  3. The membership of the Council currently comprises:
    • Angus Reid
    • Cello MRUK
    • ComRes
    • Dods Polling
    • GFK-NOP
    • Harris Interactive
    • ICM
    • Ipsos MORI
    • Marketing Means
    • Opinion Research Business
    • Opinium
    • Populus
    • TNS System Three
    • YouGov
  4. For further details contact John Curtice on 07710-348 755, Andrew Cooper on 07500-858626, or Nick Moon on 07770-564 664.

ComRes and BBC Watchdog Programme

2nd December 2008

The British Polling Council (BPC) has received a complaint concerning a research project by ComRes for the BBC’s Watchdog programme.

In all over 24,000 people responded to an invitation by the BBC Watchdog programme to complete an online questionnaire. In its report ComRes confirmed that it is a member of the BPC and prominently set out the specific BPC requirements for information that, under the objects and rules of the BPC, must accompany the publication of polls and surveys.

It should be understood that the BPC is concerned only with polls and surveys that set out to measure the opinions of representative samples, for example the views of all adults, or all voters. Therefore, the publication of results must include a statement specifying the universe (all adults / voters etc) effectively represented.

In the view of the management committee of the BPC, the project conducted by BBC Watchdog and ComRes cannot be considered to be a poll or survey as defined by the BPC simply because the results are a simple addition of the responses given by people who decided to log on and answer the questions. Demographic questions were not included in the online questionnaire. Therefore it is impossible to know whether the results obtained are representative of all adults or not, and impossible to weight the data to reflect the demographic profile of all adults.

ComRes understand and accept that the project should not have been described as a survey within a meaning that would qualify it to fall within the BPC rules, and have undertaken not to make any such claims in future. On this basis the BPC has decided to take no further action.

Further information:-

New President for British Polling Council

7th October 2008

The British Polling Council is very pleased to announce that Professor John Curtice has agreed to become the new President of the organisation.

John takes over the role from John Barter who has been President since the formation of the British Polling Council in 2004. John wishes to retire from the role and give the new incumbent ample time to assume the responsibilities of President before the next election. All members of the BPC would like to thank John for his stewardship of the BPC over the past few years and wish him well in his retirement.

John Curtice is Professor of Politics and Director of the Social Statistics Laboratory at the University of Strathclyde. He was co-director of the British Election Study, the principal academic survey based study of voting behaviour, from 1983 to 1997, and has been a co-editor of NatCen’s British Social Attitudes series since 1994. He has written and commentated widely in the media and in academic publications on polls and polling, and worked at some stage in collaboration with all the current major polling companies. He was also a member of the Market Research Society’s enquiry into the conduct of opinion polls after the 1992 general election.

The code of the British Polling Council has been effective in ensuring that consumers of poll information have sufficient information available on which to judge the validity of results. John will be working together with the officers of BPC and with the full co-operation of all members to ensure that standards of disclosure are maintained and our rules enhanced wherever necessary to ensure full compliance by all members.

Further information:-

Complaint against Ipsos MORI regarding Publication of Results for a survey for Transport for London

21st May 2007

A meeting of the members of the British Polling Council (BPC) was held on Friday 16th May. Discussion concerned an Ipsos MORI survey for Transport for London, aspects of which were released to the press by the Mayor of London, Ken Livingstone on 18th December.

At the time of publication by Mr Livingstone, Mr Henry de Zoete sought further specific information from Ipsos MORI. When this was not forthcoming, Mr De Zoete complained to the BPC that Ipsos MORI was in breach of its rules.

The BPC appointed a Disclosure Committee to examine the complaint made by Mr De Zoete. This Committee concluded that the findings of the survey did fall under BPC rules following release by the Mayor of London, and that Ipsos MORI did not act in conformity to these rules when it did not make available full details of the survey when it was requested to do so. This conclusion was accepted in full by the Management Committee of the British Polling Council.

Ipsos MORI has accepted the findings of the British Polling Council and has apologised for not making the information available when requested. As Ipsos MORI has now made available computer tables from its survey, the full meeting of the BPC decided to take no further action against the company.

In order to prevent a recurrence of this problem, Ipsos MORI and other members of the BPC will carefully review the contracts entered into with clients to ensure that these contracts do not conflict with their obligations under the Objects and Rules of the BPC.

Members re-affirmed that the rules of the BPC cover all polls and surveys they conduct that are published and where there is a legitimate public interest in the full findings being made publicly available. Members also agreed that the rules of BPC do not conflict with the obligations of members to other associations to which they may belong, including the Market Research Society and ESOMAR.

Members of the British Polling Council will be working to clarify certain aspects of the rules to ensure that all members are in future clear as to their responsibilities under the code.

Further information:-

Accuracy Of The Final 2005 Polls

6th May 2005

The table below shows the final polls and the estimates made of the outcome of the General Election by members of the British Polling Council (BPC).

Collectively, these polls are the most accurate predictions ever made of the outcome of any British General Election. As can be seen from the table no estimate was, on average, more than 1.5% adrift from the final outcome, and every individual estimate was within 2% of the result for each party.

  ICM Ipsos MORI NOP Populus YouGov Result
  % % % % % %
Labour 38 38 36 38 37 36
Conservatives 32 33 33 32 32 33
Liberal Democrats 22 23 23 21 24 23
Other Parties 8 6 9 9 7 8
Average Error 1% 1% 0.25% 1.5% 1%  

NOTE. Polls conducted wholly or partly since Monday 2nd have been included in the table. The accuracy of each estimate is shown by the use of average error, being the average of the percentage differences between these four estimates and the final result.

Contacts:-

  • Nick Sparrow (ICM) 020-7436 3114
  • Andrew Cooper (Populus) 020-7253 9465

Election / Referendum Campaign Polls — Release Of Data On Early Voters

21st December 2004

The Electoral Commission has made recommendations to the British Polling Council (BPC) in its report on the European Elections. These recommendations concern election and referendum campaign polls that may include interviews with people who have voted at the time a poll is taken (by post or other remote method). In respect of such polls members of the BPC have agreed to the following restrictions on the publication of poll data.

BPC members conducting polls during an election or referendum campaign seek to ensure their samples are representative of all voters. These campaign polls may therefore contain responses from people who have voted by post as well as those who have yet to vote. Other forms of voting may be introduced that allow people to register their votes before polling day. BPC members will publish voting figures for the whole sample (including people who have voted as well as those who have not done so) in accordance with the BPC rules of disclosure. However (as a specific exception to these rules) BPC members will NOT place in the public domain or cause to be published ANY data that relates specifically to the sub sample of people who have already voted at the time a poll is taken including their number within the total sample.

Such data, if it is collected, may only be published after the polling stations have closed.

The BPC has had initial discussions with the Electoral Commission and will seek an early opportunity to discuss in more detail with the Electoral Commission what further guidelines should be adopted by polling companies for polls conducted in postal only elections or in any future election when a clear majority of voters are expected to cast their votes before polling day either remotely or by post.

The Electoral Commission website

British Polling Council Launched

15th November 2004

The British Polling Council (BPC) is officially launched on Monday 15th November.

Following the publication of any poll, BPC members have agreed to fully disclose all relevant data about their polls. Members will publish details of the questions that have been asked, describe fully the way the data has been analysed and give full access to all relevant computer tables. By fully disclosing all relevant data members of the BPC aim to encourage the highest standards in public opinion polling.

The Founder members of BPC are Comres, ICM, Ipsos MORI, NOP, ORB, TNS/System 3 and YouGov.

Since the formation of the new council was first announced three other companies have applied to join and the BPC will welcome new members committed to full disclosure of public opinion research.

Further information can be obtained from the Secretary, Nick Sparrow at ICM Research, or by exploring www.britishpollingcouncil.org

On November 15th the British Polling Council (BPC) will be launched

1st November 2004

A number of Britain’s leading pollsters have today announced plans to launch the British Polling Council.

The objective of the Council is to ensure standards of disclosure which will give consumers of survey results that enter the public domain an adequate basis for judging the reliability and validity of the results. Through full disclosure the Council aims to encourage the highest professional standards in public opinion polling and to advance the understanding, among politicians, the media and general public, of how polls are conducted and how to interpret poll results. The BPC will also provide interested parties with advice on best practice in the conduct and reporting of polls.

The following companies are applying to be founder members of the new organisation:

  • Comres
  • ICM
  • Ipsos MORI
  • NOP
  • ORB
  • Populus
  • TNS System 3
  • YouGov

John Barter, ex Chairman of NOP and a past Chairman of the Market Research Society has agreed to become President of the BPC. He will refer any questions raised under the rules of disclosure to an investigating committee comprising three people drawn from the Sub Committee on Disclosure. The following have agreed to serve on this sub-committee in a personal capacity.

  • Simon Atkinson (Ipsos MORI)
  • David Butler (Fellow of Nuffield College Oxford)
  • David Cowling (BBC)
  • David McKie (ex Deputy Editor of The Guardian)
  • Nick Moon (NOP)
  • Adam Phillips (A past Chairman of the Market Research Society and Chairman of ESOMAR’s Professional Standards Committee)
  • Colin Rallings (Professor at the Local Government Chronicle Election Centre at University of Plymouth)
  • Peter Riddell (The Times)
  • Peter Kellner (YouGov)

The BPC invites applications for membership from organisations that are engaged in public opinion polling using sampling methods and weighting procedures designed to accurately represent the views of all people within designated target groups (such as all adults, or voters etc).

The BPC has been modelled on the successful National Council for Published Polls in the USA and BPC members are grateful for the advice received from the NCPP in starting the BPC.

Commenting on the birth of the new organisation John Barter said.

“The organisations that are seeking membership of the BPC recognise the need for uniform standards of disclosure about how polls are conducted so that consumers of poll findings have an adequate basis for judging the reliability of the findings and all the proposed founding members enthusiastically support this new Council. Once the Council is established readers of poll findings will have full access to information on how polls have been conducted, what questions were asked and how the data collected has been computed to produce the published results”

The formation of the British Polling Council follows the publication of an Early Day Motion signed by more than 80 MPs from all the main parties which expressed regret at “the decline of self-regulation of public opinion polling companies in the United Kingdom” concern that “there are no sufficient checks on the integrity of polling or polling organisations” and expressed “concern at the proliferation of non-scientific/empirical polling, in particular the use of techniques designed to secure the results favoured by those who commission the polls, and lack of transparency in the methodology employed”.

The BPC shares these concerns. For these reasons its membership will be restricted to organisations that:

  1. set out to measure the opinions of representative samples scientifically, and
  2. uphold the principle of transparency.

In agreeing to disclose full information about how samples are drawn, how raw data are weighted, and the full wording of questions and answers, the BPC seeks to demonstrate its commitment to the highest standards and greater public understanding of the methods used to conduct representative surveys.

The BPC cannot extinguish the danger that non-members might produce unscientific, biased or opaque polls; it does commit its members to standards of disclosure that are intended to satisfy the legitimate concerns of the general public.