As we approach a year on since the 2017 General Election, BMG has published its report on the performance of its final General Election poll including sources of error and some detail on how we may change our approach at future elections.
Our report concludes that:
- There were two sources of the error; turnout adjustments and the final sample
- We should avoid a compressed fieldwork period
- Final polls should account for where parties are, and are not, standing
- Conduct larger samples for General Election polls
- New detail about the uncertainty of General Election polls
- A flexible, cross-election approach is required, accounting for electoral registration
The full report can be found here.
For comment, or if you would like to find out more about how BMG will be updating its polling methodology, subscribe to our newsletter, or get in touch by email or phone.
0121 333 6006
Dr. Michael Turner – Research Director & Head of Polling – BMG Research