A talk by Jonathan Mellon (Research Fellow, Nuffield College, University of Oxford).
The 2015 UK General Election saw an unprecedented interest in election forecasting models. Despite the interest, these models performed extremely poorly largely due to their reliance on pre-election polling. Pre-election polls for the 2015 UK General Election missed the final result by a considerable margin: underestimating the Conservative Party and overestimating Labour. But why did the polls miss so badly? We analyse evidence for five theories of why the polls missed using British Election Study (BES) data. We find limited evidence for systematic vote intention misreporting, late swing, systematically different preferences among “don’t knows” or differential turnout of parties’ supporters. By comparing the BES face-to-face probability sample and BES Internet panel, we show that the online survey’s polling error is primarily caused by under-sampling non-voters, then weighting respondents to represent the general population. Consequently, demographic groups with a low probability of voting are over-weighted within the voter subsample. Finally, we show that this mechanism is likely partially responsible for the polls overestimating the Liberal Democrats in 2010, showing that this is a longstanding problem.
Jonathan Mellon's guest lecture is part of the seminar 'Election Forecasting' but is open to all Hertie School students.