Why were the national media and the forecasts of the polls wrong for the 2016 presidential election?
The media missed the strength of Donald Trump’s widespread support. Also, more consideration should have been given to the anger of the electorate and its economic insecurities.
Trump carried the South except for Virginia and most of the Midwest including the majority of 15 battleground states. Many polls did not include enough non-college educated white voters, independents, and people who had not voted in previous elections. Candidate Trump reminded the country a number of times that people at his rallies said they had never voted but that they were not going to vote and for him. These people can be difficult to reach for a poll because many of them cannot afford landlines and may not be available for a survey on cell phones.
On the Friday before the election, I thought that the American people could wake up to the election of Donald Trump. The reason for my thinking was that many people engage in what is called the “social desirability effect,” in which people selected for a poll would not say that they would vote for Trump. Also, since the polls pointed to a Clinton win for many weeks, a lesser sense of urgency by Democrats to mobilize their followers may have developed.
Polling is not a fully scientific enterprise, and at best is comparable to an educated guess of an outcome.