ComRes Clients
Home  >>  Poll Digest  >>  Public Poll Methodology

Public Poll Methodology

ComRes is a member of the British Polling Council (BPC) and abides by its rules.

We agree with the BPC’s principal objective which is to “ensure standards of disclosure designed to provide consumers of survey results that enter the public domain with an adequate basis for judging the reliability and validity of the results”.

The information below is supplied to meet this objective.

Telephone Methodology

Interview method - telephone
Population effectively sampled - all adults aged 18+

Sampling Method

Within each government office region a random sample of telephone numbers is drawn from the entire BT database of domestic telephone numbers. Each number so selected has its last digit randomised to provide a sample including both listed and unlisted numbers.

Data weighting

Data are weighted to the profile of all adults aged 18+ (including non telephone owning households). The weighting is undertaken by sex, age, social class, household tenure, work status, number of cars in the household and whether or not respondent has taken a foreign holiday in the last 3 years. Targets for the weighting are derived from the National Readership survey, a random probability survey comprising 34,000 random face-to-face interviews conducted annually.

ComRes also applies two further weights to its political polls. First, we weight polls according to the reported likelihood that respondents will vote. That likelihood is obtained by offering respondents a scale from 1 to 10 where 1=certain not to vote and 10=absolutely certain to vote. If respondents answer 4 or below we assume they are unlikely to vote and no effort is made to ascertain their voting intentions. Those who answer between 5 and 10 are asked which party they would vote for and their answers are weighted from 0.5 to 1.0 in accordance with their reported likelihood of voting.

Secondly, to ensure the sample is politically representative, we weight by how respondents recall having voted at the last general election. This weighting is based in part on comparing the distribution of this recall vote (for those who declare for whom they voted) with the actual result of the 2010 General Election and in part with the average past vote obtained in our previous twelve polls. Three-quarters of the weight is based on the former comparison and one-quarter on the latter. This procedure is designed to correct for any sampling bias and sampling error (respectively), while acknowledging the existence of response error in the reporting of past vote.

If respondents decline to name a party in response to the voting intention question, they are asked how they would probably vote if it were a legal requirement to do so and allocated this party. All who continue to remain undecided or refuse to say for whom they would vote are then allocated a party, according to the party with which they most closely identify. In both cases, the data are weighted by reported likelihood of voting.

The BPC website is at www.britishpollingcouncil.org

Online Methodology

Interview method - online
Population effectively sampled - all adults aged 18+

September 2010 sees ComRes’s first published online voting intention poll. ComRes is running both telephone and online methodologies concurrently and is the only polling company to publish results using both methods. The details of the poll methodology is summarised below.. As with our telephone polls, the data is weighted to be politically representative and propensity to vote is also taken into account to model the voting sample.

Sampling Method

ComRes interviews a representative sample of GB adults online. Respondents are selected using two methods. Firstly, respondents are selected from an online panel or more than 250,000 GB adults. Secondly, about 70% of the sample is selected through Random Online Sampling. This methodology randomly selects people through website invitations or pop-ups and profiles them using a series of demographic data questions.
Within each government office quotas are set for the demographic profile of the sample in each region.

Data Weighting

Data are weighted to the profile of all adults aged 18+ (including households without internet). The weighting is undertaken by sex, age, social class, household tenure, work status, number of cars in the household and whether or not respondent has taken a foreign holiday in the last 3 years. Targets for the weighting are derived from the National Readership survey, a random probability survey comprising 34,000 random face-to-face interviews conducted annually.

Similarly to the telephone poll, ComRes also applies two further weights to its political polls. First, we weight polls according to the reported likelihood that respondents will vote. That likelihood is obtained by offering respondents a scale from 1 to 10 where 1=certain not to vote and 10=absolutely certain to vote. If respondents answer 4 or below we assume they are unlikely to vote and no effort is made to ascertain their voting intentions. Those who answer between 5 and 10 are asked which party they would vote for and their answers are weighted from 0.5 to 1.0 in accordance with their reported likelihood of voting.

Secondly, to ensure the sample is politically representative, we weight by how respondents recall having voted at the last general election. This weighting is based in part on comparing the distribution of this recall vote (for those who declare for whom they voted) with the actual result of the 2010General Election and in part with the past vote recall . This procedure is designed to correct for any sampling bias and sampling error, while acknowledging the existence of response error in the reporting of past vote.

If respondents decline to name a party in response to the voting intention question, they are asked how they would probably vote if it were a legal requirement to do so and allocated this party. All who continue to remain undecided or refuse to say for whom they would vote are then allocated a party, according to the party with which they most closely identify. In both cases, the data are weighted by reported likelihood of voting.

Make an Enquiry

 
 
 

 

OR CALL US:

London: +44 (0) 20 7871 8660
Brussels: +32 (0) 2 234 63 82

Comres
in the News

Comres
Blog