Border poll research questions involve cross-section of the population
Polling was carried out by Belfast-based polling and market research company LucidTalk.
The project involved interviewing a random sample of 1,089 Northern Ireland residents, aged over 18, by telephone (approximately 90%), and face-to-face (10%).
The sample of 1,089 was carefully selected to be demographically representative of the residents within the targeted geographic area, ie Northern Ireland.
Polling took place from September 11-24 . The project used 20 poll questions agreed with the Belfast Telegraph and other poll project partners.
All questions were agreed with the project partners, and also to British Polling Council (BPC) and professional market research standards, to ensure neutrality and balance. Results presented are weighted to match the Northern Ireland demographics of gender, age, religion, socio-economic group, region (within Northern Ireland), population density and phone status.
This demographic analysis ensured the final results represented an accurate view of current Northern Ireland opinion.
All reported margins of sampling error will include the computed design effects for weighting.
All data results produced are accurate to a margin of error of +/-2.9%, at 95% confidence.
Polling for the border question was carried out approximately 50% before and during the Scottish referendum voting day on September 18, and 50% after this date, to ensure a balanced sample.
LucidTalk is a member of all recognised professional Polling and Market Research organisations, including the UK Market Research Society, the British Polling Council (BPC) and the European Society of Market Research organisations. All polling, research, sampling, methodologies used, market research projects and results and reports production have been carried out to the professional standards laid down by the BPC.