Polls used to be seen as the gold standard for assessing politicians, elections and voter concerns. In recent years, polling’s reputation has been tarnished. In the 2017 U.K. election, final polls underestimated the Labour vote and overestimated support for the U.K. Independence Party. Almost every poll in the 2016 U.S. election missed support for Republican Donald Trump, who won the presidency. In 2016, pollsters failed to predict the clear victory of the “leave” camp in the U.K. referendum on whether to stay in the European Union and the rejection of the Colombian peace deal with rebels. In 2015, polls were wrong on outcomes in Israel, the U.K. and Greece. The bungles have undermined the industry’s claim to scientific rigor. Can poll crafters devise a better formula that delivers more accurate results in this no-time-to-spare mobile era?
Ahead of the 2018 U.S. midterm elections in November, many people fear that the polls can’t be trusted. While post mortems of the 2016 election noted that national polls correctly predicted that Hillary Clinton would win the total U.S. popular vote, polls at the state level were badly off and underestimated Trump support. Because the U.S. president is ultimately chosen by the Electoral College, which is guided by state results, almost no polls predicted the Trump victory. Pollsters certainly face a range of constraints. In the U.S., a majority of people now live in homes without a landline phone. So to reach a representative group, firms have increased calls to mobile phones, which are now three-quarters of some samples. To do this, pollsters have to dial numbers by hand (U.S. law bans cell phone autodialing) and make more calls, since mobile users tend to screen out unknown callers and fewer will sit through 20 minutes of questions. This isn’t cheap — mobile-phone surveys can cost nearly twice as much — or easy. Pew Research’s response rate on its 1997 polls was 36 percent; it was just 9 percent in 2016.
