With British Election Days Away, Reason to Question Polls

If the election result is only a few percentage points different from the polls, it could change the whole outcome.

View of the Houses of Parliament in London, England, December 21, 2011
View of the Houses of Parliament in London, England, December 21, 2011 (Ben Sutherland)

With surveys showing Britain’s Conservative and Labour Parties neck and neck for Thursday’s election, both are hoping to prove the pollsters wrong. There is reason to believe they might.

Prime Minister David Cameron’s party is hoping for a replay of the 1992 election when the polls predicted a narrow win for the opposition Labour Party but John Mayor actually won a comfortable majority in the House of Commons with 42 percent support.

Labour may want to repeat a more recent experience. In 2010, the polls gave it between 24 and 29 percent support, with most predicting 28 percent. In fact, it got 30 percent support although it still lost 91 seats.

May 2015, the New Statesman‘s election website, reports that in all but one of the twenty most recent elections, pollsters overestimated Labour’s support. The exception was 2010 when they overestimated the Liberal Democrats’ popularity instead.

This should be encouraging to the Conservatives who are urging voters to stick with a successful government rather than risk an unstable Labour alliance with the Scottish National Party. But with less than a week to go before the election, they had probably expected to be up by now. The Conservatives have delivered on their key promises to reduce the deficit and revitalize the British economy. Labour, by contrast, has lurched to the left under Ed Miliband’s leadership when voters have traditionally punished the socialist party when it veered too far from the center.

One thing that’s working against the Conservatives is the support for the Euroskeptic United Kingdom Independence Party. Polling at around 14 percent nationwide, Nigel Farage’s party is unlikely to win more than a handful of seats in Westminster. But it could split the right-wing vote in many constituencies, allowing Labour to take seats that would otherwise have gone to the Conservatives.

Under Britain’s first-past-the-post voting system, whichever party wins a plurality of the votes in a given constituency wins.

This also benefits the Scottish National Party. Although less than half of Scottish voters supports the separatists, they could nevertheless win almost all of Scotland’s 59 seats in Parliament.

The Liberal Democrat leader, Nick Clegg, is expected to benefit from tactical voting. The Guardian reports that half of Conservative voters in his Sheffield Hallam constituency are likely to support him in order to stop Labour taking over his seat.

There is no such polling data available for all of Britain’s 650 constituencies, however, making it difficult to predict the outcome.

May2015 takes into account the dozens of constituency-level polls the Conservative peer Michael Ashcroft has conducted but extrapolates the rest of its predictions from national polls. It gives the Conservatives 273 seats, 34 less than they have now, and Labour 268, up by ten. Other forecasts similarly show the Conservatives ahead but falling short of a majority. But if the actual election result is only a few percentage points different from the polls, that could affect the outcome in dozens of constituencies.

Leave a reply