UK Election Inquiry Blames Unrepresentative Sample

Draft findings from an inquiry steered by the British Polling Council (BPC) and MRS into why UK pollsters overestimated Labour and underestimated Conservative support in last year's UK general election, suggest that unrepresentative samples led to an inability to reach enough Conservative voters.

Of the final eleven polls conducted, ten suggested that Labour and the Tories were neck-and-neck, which would have resulted in a hung parliament. The inquiry was set up shortly after the results were announced, to look into the possible causes of 'apparent bias', and to make recommendations for future polling. Under the chairmanship of Professor Patrick Sturgis (pictured) of the University of Southampton, the inquiry will submit its final report in March. Preliminary findings show that unrepresentative samples were the main cause of polling inaccuracy, with Labour voters over-represented and Conservative voters under-represented.

According to an article in today's Guardian, Sturgis' conclusions were informed by three recent analyses of YouGov, British Election Study, and British Social Attitudes data. Those unrepresented in the polls included the over 70s, who are more likely to vote Tory, but were not accounted for in YouGov's Internet panels; and young non-voters - generally the under 30s, who generally lean to the left of the political spectrum.

MRS CEO Jane Frost thanked Sturgis and his panel for the speed of their response and commitment to 'producing robust and insightful conclusions'. Commenting on the findings, John Curtice, President of the BPC, said: 'Today's unveiling will provide the polling companies and everyone else with an interest in its work to hear and respond to the inquiry's initial conclusions. It hopes that this process will prove helpful to the work of the inquiry and ensure that its final conclusions are based on the strongest possible body of evidence'.

Web sites: www.mrs.org.uk and www.britishpollingcouncil.org .