|
The total number of people completing the online surveys was 98. Of these 68 were from the UK and 30 from any of the other 26 states of the EU (we have no analysis or data on the citizenship or location of these respondents). Two separate surveys were compiled for each group with slightly different sets of questions: the UK group having 51 compulsory questions and the EU group 44. Overall, given the number of questions, the respondents showed a high rate of completion with 54 UK respondents and 25 EU respondents reaching the end of the survey.
For our analysis and presentation of the statistics only the results of those respondents who finished the survey were used. That is to say that a total of 19 partially completed surveys were discarded. This validation process had only a minor effect on the results shown on this web site. These submissions were discarded as they may have skewed the results, where related questions later in the survey were not answered, so could not be directly compared with the same set of respondents.
As far as we can tell there is no evidence of respondents filing the survey multiple times (ie there were few duplicate IP addresses, and where there were duplicates this was identified as a corporate gateway address and the respondents had given different answers).
Respondents to the survey were selected only by their willingness to complete the questions – there was no attempt made to find an accurate cross-section of society. Invitations to take the survey were sent out to friends, family and colleagues, as well as being posted to fairly large distributed mailing lists and in the forum areas of web sites. Some of the forum postings were made to newspaper sites, including English language areas of news sites published within the EU but outside the UK. In some cases sites or discussion fora that appeared to support ‘UK independence’ or promote euroscepticism were selected as invite platforms, in order to gain a wide selection of views.
No incentive was offered to respondents in the invitation, other than the offer that if they submitted their email address they would be notified when this research material was published. The central reason for not offering an incentive was so as not to encourage frivolous or repeated submission of the survey. Some texts on the subject of incentives in surveys were consulted, with the overall impression that there was no increase in participation or accuracy when using an incentive.
| 
