Click on the title of the poll to see the original results as published by the polling company. Please help us update our records and let us know if any of these links don't work, or if you find any other relevant opinion polls.
| Very good idea | 18% |
| Good idea | 32% |
| Bad idea | 26% |
| Very bad idea | 23% |
| Don't know | 1% |
| Very good idea | 13% |
| Good idea | 29% |
| Bad idea | 28% |
| Very bad idea | 28% |
| Don't know | 2% |
ICM Research interviewed a random sample of 1007 adults aged 18+ by telephone between 8th-9th November 2006. Interviews were conducted across the country and the results have been weighted to the profile of all adults.
| Very good idea | 12% |
| Good idea | 34% |
| Bad idea | 29% |
| Very bad idea | 22% |
| Don't know | 3% |
| Very good idea | 10% |
| Good idea | 31% |
| Bad idea | 33% |
| Very bad idea | 23% |
| Don't know | 3% |
ICM interviewed a random sample of 1001 adults aged 18+, by telephone between 21-23rd July 2006. Interviews were conducted across the country and the results have been weighted to the profile of all adults. This is the first poll of this series to show a clear majority in opposition to the ID cards scheme.
| Yes. I am now in favour of them | 3% |
| No. I am STILL in favour of them | 15% |
| Yes. I am now against them | 14% |
| No. I am STILL against them | 68% |
Conducted with 600 silicon.com readers 7-13th March 2006 this is poll is not representative of the total population. However, it is interesting to assess the shift in opinion from people who say they were always against them (68%) and respondents who appear to have decided more recently that they're against them (14%). This is particularly interesting compared to a similar poll carried out in August 2004 (1,500 respondents) where only 40% of respondents were opposed to id cards, compared to 82% now.
| I am in favour | 52% |
| I am opposed | 37% |
| Don't know | 11% |
Conducted 21st – 22nd February over the internet with a nationally representative sample of 2019 people. Compared to June 2005 when the Daily Telegraph last conducted a poll on id cards, a higher number are in favour (52% vs 45%) and fewer people are opposed (37% vs 42%). The results are not, however, directly comparable as the question has changed slightly. In addition, the context has changed now that the ID card Bill has been passed. Daily Telegraph commentary.
| Very good idea | 16% |
| Good idea | 36% |
| Bad idea | 26% |
| Very bad idea | 20% |
| Don't know | 2% |
ICM interviewed a random sample of 1002 adults aged 18+, by telephone between 17-19th February 2006. Interviews were conducted across the country and the results have been weighted to the profile of all adults. Results are broadly stable compared to the previous poll in November 2005 with the same question.
| Against | 33.5% |
| Undecided | 13.9% |
| In favour | 52.6% |
1,122 students surveyed using existing online panel. Study conducted by the OU around privacy attitudes and acceptance of id cards among students in Nov/Dec 05. The study demonstrates that the elements of compulsion and a centralised database (currently favoured by the government) are most negatively viewed compared to other proposed schemes.
| Very good idea | 14% |
| Good idea | 36% |
| Bad idea | 24% |
| Very bad idea | 23% |
| Don't know | 2% |
ICM interviewed a random sample of 1013 adults aged 18+, by telephone between 18th and 20th November 2005. Interviews were conducted across the country and the results have been weighted to the profile of all adults. The results show an increase in people seeing id cards as a bad idea compared to the same question in a previous poll carried out in June 2005.
| Introducing identity cards | 61% |
Populus interviewed a random sample of 1,506 adults aged 18+ by telephone between July 22nd - 24th 2005. Interviews were conducted across the country and the results have been weighted to the profile of all adults. The bulk of the questions related to voting intentions and attitudes, and a large number of questions relating to attitudes around the London Bombings, police powers and muslims.
| Yes - should be | 53% |
| No - should not be | 43% |
| Don't know | 4% |
1,005 telephone interviews conducted 15-17th July 2005. This was the last question on a survey which focussed on voting intentions and attitudes towards terrorism, muslims and the London bombings.
| Agree | 61% |
| Disagree | 34% |
| Don't know | 5% |
1,005 telephone interviews carried out 8th – 10th July 2005.
| Support | 65% |
| Oppose | 31% |
| Don't know | 4% |
Conducted 8th and 9th July 2005, immediately after the July 7th London bombings, as part of ICMs telephone omnibus. Sample of 533 adults is not considered robust and leading question implies link between terrorism and identity cards at a time when people are looking for ways to combat terrorism.
| In favour | 50% |
| Opposed | 38% |
| Don't know | 12% |
YouGov questioned 1854 adults aged 18+ throughout Britain online on the 8th July 2005. Carried out immediately after the London Bombings on July 7th 2005, the poll shows a small shift in favour of id cards. However, more than half the people questioned decided they didn't believe id cards would help prevent terrorism.
| Yes they would | 33% |
| No they wouldn't | 56% |
| Don't know | 11% |
| In favour | 45% |
| Opposed | 42% |
| Don't know | 13% |
Conducted online 28th to 30th June 2005 with a nationally representative sample of 3717 adults. The shift in results from September 2003 (in favour, 78%) is significant and represents a shift in public opinion as more details of the proposed plans for id cards become clear.
| Very good idea | 14% |
| Good idea | 41% |
| Bad idea | 25% |
| Very bad idea | 18% |
Conducted between 10th and 12th June 2005 as part of the ICM telephone omnibus with a random sample of 1010 adults. Equates to 55% in favour of id cards and 43% opposed. The question is a costed assessment of opinion, whereas other polls typically assess opinion uncosted first. It also assumes respondent knowledge of the current cost of passports and requires them to calculate the additional cost of the id card.
| Yes | 49% |
| No | 47% |
Populus interviewed a random sample of 711 adults aged 18+ by telephone between April 18th & 19th 2005. Interviews were conducted across the country and the results have been weighted to the profile of all adults. The poll is really assessing fears around terrorism, rather than opinions around id cards.
| Easier to commit | 7% |
| Harder to commit | 49% |
| No difference | 41% |
| Don't know | 3% |
Conducted by telephone 25-28th February 2005, with a nationally representative sample of 963 adults, this poll discusses id cards in the context of electoral fraud and new voting methods. As such it is fairly leading, as this question and those preceding it, make the assumption that lack of identification is a cause of electoral fraud. This poll is really assessing fears around electoral fraud, rather than opinions around id cards.
| Very good idea | 29% |
| Good idea | 52% |
| Bad idea | 12% |
| Very bad idea | 5% |
| Don't know | 2% |
Carried out as part of ICMs telephone omnibus 1st and 2nd December 2004 with a random sample of 1022 adults, this poll shows 81% of respondents in favour of identity cards. As an organisation, Reform is against increases in public spending and government intervention, and this is highlighted in their emphasis on the compulsory nature of the proposed system. Interestingly, the poll assesses costed opinion too, and demonstrates that at a suggested cost of at least £35 per person, those in favour fall to 68%.
| Agree strongly | 38% |
| Tend to agree | 34% |
| Tend to disagree | 14% |
| Disgree strongly | 10% |
| Don't know | 5% |
The Joseph Rowntree Reform Trust Ltd commissioned this opinion poll from ICM Research as part of its 1991-2004 State of the Nation series of polls investigating public attitudes to government and constitutional reform in the UK. The poll was designed by Professor Helen Margetts, Mark Ross and Professor Stuart Weir in consultation with Nick Sparrow of ICM. 2,373 adults were interviewed face to face on the street between 26 May and 4 June 2004.
| Strongly support | 40% |
| Support | 21% |
| Strongly oppose | 12% |
YouGov questioned a representative sample of 2,003 electors throughout the United Kingdom between May 11 & May 13.The Poll was commissioned partly in response to a MORI survey published a month earlier by IT supplier Detica, and highlights a number of concerns over proposed financial penalties. BBC analysis.
| Strongly in favour | 50% |
| Moderately in favour | 30% |
| Neither in favour or opposed | 8% |
| Moderately opposed | 5% |
| Strongly opposed | 6% |
| Don't know | 1% |
Detica is an IT consultancy specialising in the delivery of intelligence systems, and has been offering consulting advice to the government for 30 years.
This poll was conducted by telephone 18-23 March 2004 with a nationally representative sample of 1000 adults and shows 80% in favour of a national identity card scheme. However, it is worth noting that only 27% of respondents claim to know a fair amount or a great deal about the scheme, while the majority of respondents only heard about it, or only know a little about it. The poll then goes on to describe the scheme, before assessing opinion. In this way the poll forms opinion while also measuring it.
| In favour | 78% |
| Opposed | 15% |
| Don't know | 7% |
Conducted online with a nationally representative sample of 2312 adults 2-4th September 2003. This poll shows 78% of respondents in favour of identity cards, which is significantly higher than the June 2005 poll carried out with the same question and methodology which had only 45% in favour. YouGov considers this change to be entirely down to a shift in public opinion over time.
| Support | 85% |
| Oppose | 11% |
| Don't know | 4% |
Conducted 21st September 2001 shortly after 9/11, this is a telephone sample of only 513 adults. The question on identity cards comes after a series of questions about the terrorist attacks and the government response. Although the question itself does not suggest a direct link between id cards and terrorism, the context is still somewhat leading and these results should be used with caution.
| Support | 75% |
| Oppose | 17% |
| No opinion | 8% |
Conducted 19-22 May 1995, MORI interviewed a nationally representative sample of 996 adults in home, face-to-face for the BBCs On the Record program.
Opinion polls offer a snap shot of public opinion at a specific point in time. To be credible, they need to cover the right sample, ask the right questions and get the figures right!
To interpret the results correctly, we need to understand who commissioned the work, who carried out the research using what methodology and sample, what else was happening during the fieldwork dates and how biased the questions might be.
Who commissioned the research will influence what questions are asked and how they are asked, with a potentially significant impact on the results. Every organisation commissioning a poll has an agenda, and a question they want to answer. Even in polls which have been well-designed and carried out by a reputable polling house, a bias is likely to be inbuilt, whether deliberate or not! It is therefore important to understand what question is being asked, and why.
A valid opinion poll needs to be carried out by a reputable polling house. Most pollsters are members of professional organisations which aim to maintain standards.
These include people like Association of Professional Opinion Polling Organisations and the British Polling Council. Member include MORI, NOP, Gallup, Harris, ICM, and internet pollster YouGov.
Face to face, telephone and internet polls will all generate different results, and it is difficult to compare them to each other even if the questions are identical. Issues facing each methodology include the reliance upon a professional fieldforce for face-to-face interviewing, and the fact that respondents tend to give fewer negative responses when face-to-face. Telephone interviewing is beginning to face complications due to the increasing use of mobile phones, which makes it harder to obtain a representative sample, while internet polling also struggles to provide a representative sample, although this is overcome as ever more households connect to the internet.
Telephone phone-ins where respondents phone into TV or radio shows or respond to ads in the paper, are not true opinion polls. This is because respondents are self-selected and do not offer a representative sample of the population. Results from phone-ins should be used with caution.
Reputable polling houses consider 1000 respondents to be the minimum sample size for robustness.
In addition to sample size, the sample needs to be demographically representative of the total UK population, or representative of a specific subsection that the poll is aiming to get opinions from, eg ethnic minorities or a disabled group.
In addition to being a demographic match (ie representative of gender, age, ethnicity, religion, region, class, lifestage), a robust poll should also be attitudinally representative. For example, it is possible that certain types of people are less likely to want to take part in polls, particularly on controversial issues or when they feel they might hold a minority view.
Most people don't hold deep-seated opinions on every single issue, particularly new issues or those which they feel don't concern them. Because opinion polls are a snapshot in time, this means that the results can be swayed by topical news or events which took place at the same time.
For example, asking questions about terrorism directly after well-publicised terror attacks will generate different responses than when asked during a more peaceful period.
Questionnaires need to be short, simple and unambiguous, They should avoid jargon and shorthand, as well as uncommon or sophisticated words. Importantly, they should avoid introducing bias into the responses.
Bias can be introduced in a number of ways. For example, an openly leading question or a question that is structured in such a way that a particular response appears obvious. “do you agree...” will frequently preface a leading question, as would any particularly positive, negative or emotive terminology,
Other ways of introducing bias include a certain presumption about the subject being researched. For example, asking respondents how many books they've read in the past year, presupposes that all the respondents read books. It would be more appropriate to ask whether they read, first.
Pre-coded responses also tend to introduce bias. The options available will rarely cover all the possible responses, and often reflect the researcher's opinion of what all the possible answers and opinions might be. Respondents often find it frustrating to be unable to express views in their own words, and even where the option exits for additional open answers, respondents struggle to find their own responses, once their thinking has been led by pre-coding.
Lists of pre-coded responses also tend to introduce bias in other ways, with several studies showing that the location of different responses within a list having an impact on the outcome. Put simply, if what is considered the most popular response is put at the beginning of a pre-coded list, it is likely to become so.
Even a well conducted poll can be misrepresented! To be accurately interpreted, analysis of the results of a poll needs to take everything into account – who commissioned it, carried it out, on what sample, during what time period, using which methodology and for what purpose?
On top of this, it is important to remember that polls and questionnaires are much better at telling us what rather than why. It is very difficult to uncover respondents' deep-rooted opinions, feelings and motivations through a set questionnaire – more in-depth qualitative research is needed for that, particularly for sensitive, emotive or embarrassing topics.
Respondents also struggle to explore or give opinions on unfamiliar concepts. For example if they are being asked about something new which is difficult to understand, or which doesn't exist yet, their opinions tend to be less well-formed as it is difficult to imagine how it might impact their lives. This also means that they are more likely to be influenced by media reports around the topic, or even by questionnaire design.
Finally, sometimes, what respondents don't say is just as important as what they do say. Non-response rates tend to be very important around intimate or sensitive topics. Despite the best efforts of sample design, non-respondents may be different to respondents in some significant (usually but not always attitudinal) way. Unfortunately it is very difficult to get accurate demographic or attitudinal information on non-respondents, and therefore any impact is often surmised, rather than proven. One notable example does exist, however. Political opinion polls covering the 1992 election in the UK failed to accurately predict the election result. According to an MRS inquiry, this was due to a number of reasons including a late swing in votes, and sampling inadequacies. Significantly, the inquiry also found that the poll results were biased because Conservative voters were less likely to reveal their loyalties due to an unwillingness to answer certain questions and possibly, due to a refusal to take part in polls at all.
Support for identity cards has fallen from a high of 85% after 9/11 to 45% in June 2005, and around 52% in February 2006 once the ID Cards Bill was passed. Not all the polls are comparable, and some are less robust than others. However, the organisations MORI and YouGov both agree that these changes represent shifting public opinion against id cards, and are not merely the result of differing methodologies and samples.
The polls carried out in September 2003 and June 2005 by YouGov for the Daily Telegraph are the clearest indicator that public opinion is turning against id cards. Both polls have the same question wording, same sample design and similar preceding questions. However, in September 2003 74% of respondents supported the scheme, whereas by June 2005 this has dropped to 45%. Interestingly, there were 7% ‘don't know' responses in 2003, and 13% in 2005, indicating an increasing number of people who feel they don't have enough information to form an opinion.
In the June 2005 poll, 84% of respondents are worried about the potential disruption and inconvenience of implementation, while 66% believe that the government should not be spending £6 billion on the scheme (this increases to 81% against when the estimated costs rise to £10-19 billion).
Even in September 2003, where support for id cards was high, respondents were unsure of the supposed benefits, with 82% agreeing that it would cut down on benefit fraud, but 66% believing that the cards could be forged. Respondents were also clear about where the limits of the card should be, with 86% stating it should be free of charge; 72% stating they did not want data-sharing between government departments; and 65% stating it should only establish identity and not carry biometric or home address details.
NO2ID/ICM polls are also broadly comparable with the same question asked in June 2005, November 2005, and February 2006. The results show a decline in support between June and November 05. In June, 55% supported ID cards compared to 50% in November. In addition, the number of people feeling strongly about the issue and agreeing that id cards are “a very bad idea”, increased from 18% in June to 23% in November.
As the true details of the proposed identity card scheme and national identity register have become clearer, public support has fallen.
The NO2ID Campaign
Box 412
19-21 Crawford Street
London W1H 1PJ
enquiries@no2id.net
Tel: 07005 800 651
Press: click here
