‘Unrepresentative samples’ were to blame for pollsters’ failure to predict last year’s general election result, an inquiry has found.
By using methods that ‘systematically over-represented Labour supporters and under-represented Conservative supporters’, polling organisations left the nation expecting a hung parliament rather than the Tory victory that emerged.
The report by the British Polling Council and Market Research Society concluded ‘statistical adjustment procedures applied to the raw data’ could not undo the damage caused by these flaws.
The BPC has called for a raft of measures outlined in the report to be implemented immediately to improve the accuracy of future polls.
They include greater transparency about how polls have been weighted and to ‘specify what changes, if any, have been made since a company’s previous published poll’.
It also plans to develop industry-wide approaches to gauging confidence in a poll’s estimate of a party’s share of the vote and to calculating the statistical significance of the change in a party’s estimated vote share since a previous poll.
Professor John Curtice, president of the British Polling Council, said: ‘The inquiry has undertaken what was an important but demanding task in a timely and professional fashion.
‘I am confident that all those with an interest in understanding the difficulties that beset the polls in 2015 will find its report an illuminating and profitable read. The council now wishes to ensure that its work is put to best use so that the transparency and accuracy of opinion polls is enhanced in future.’
