Scottish Independence Referendum: The final polls

Ipsos is delighted to be the joint closest of the final polls on the Scottish independence referendum. Our final poll for The Evening Standard put No on 53% and Yes on 47%, compared with a final result of 55% for No and 45% for Yes. This is well within the margin of error for polls of this sample size.

This still slightly overestimated the Yes vote, and we will be exploring the patterns and explanations behind this in the coming days. The key reason is likely to be differential turnout between supporters of the two camps, which is very difficult to predict and control for – see an outline of some of the issues.

With a final turnout approaching 85% there is no doubt that the debate engaged the nation, and we’re pleased to have helped inform the discussion on how public sentiment in Scotland (and the rest of the UK) developed. We would like to thank all the teams that were involved across Ipsos for their excellent work and dedication.

Technical note:

• For the full data from our polls on the Scottish Independence Referendum please visit our microsite
• Technical details (for final poll as referenced above)
  • This presents the topline results from Scotland
  • Results are based on a survey of 991 respondents (adults aged 16+) conducted by telephone
  • Fieldwork dates: 16 September to 17 September 2014
  • Data are weighted by: age, sex and working status using census data; tenure using Scottish Household Survey data; and public-private sector employment using Scottish Government Quarterly Public Sector Employment series data
  • Where results do not sum to 100%, this may be due to computer rounding, multiple responses, or the exclusion of “don’t know” categories
  • Results are based on all respondents registered to vote (980) unless otherwise stated