Political Polling Did Polling Fail Us in the

Political Polling • Did Polling Fail Us in the 2016 Presidential Election? • Which Polls are Good? Which are “Biased? ” • Some recent Five. Thirty. Eight. Com columns

Nate Silver, ex-NY Times, now Five. Thirty. Eight. Com

as of end of May 2018

Why were campaign journalists so confident of Clinton's chances? [Silver] • Real shortcomings in how American politics are covered, including pervasive groupthink among media elites, • An unhealthy obsession with the insider’s view of politics, • A lack of analytical rigor, • A failure to appreciate uncertainty, • A sluggishness to self-correct when new evidence contradicts pre-existing beliefs, and • Misinterpretation and misreporting of the polls

Misinterpretation and misreporting of the polls – Throughout the campaign, the polls had hallmarks of high uncertainty, indicating a volatile election with large numbers of undecided voters. – And at several key moments they’d also shown a close race. In the week leading up to Election Day, Clinton was only barely ahead in the states she’d need to secure 270 electoral votes – Because the polls all showed Clinton favored, it probably reinforced traditional reporters’ confidence in Clinton’s prospects – Paid too much attention to the national popular vote – Focused on Clinton’s potential gains with Hispanic voters but less on indications of a decline in African-American turnout – Paid more attention to Democratic turnout strategies than to Republican ones – Failed to notice how late polls shifted towards Trump both before and after the Comey letter to Congress on Oct. 28 th

American Association of Public Opinion Research study: Possible Reasons Why the Polls Missed • Late-Breaking Vote for Trump • Upper Midwest • Many polls did not weight for education (overrepresentation of college grads in their sample) • More post-election reports of Trump voting than preelection [shy voters? Or change of mind? ] • Change in turnout distribution between D’s and R’s • Ballot order effects in Michigan, Wisconsin, & Florida • Publicity of “ 90% likelihood of Clinton win” tilted turnout • FBI Letter: mixed evidence… Clinton slide pre-dated it • Republicans “coming home” • Clinton campaign late effort elsewhere than Mich/Wisc • Lukewarm Dem’s stayed home

Republican Senate candidates’ Polls and Results

Polling in Close Elections Comes Down to Method, Weighting, Database, and Modeling Turnout

(Republican pollster) (Democratic pollster)