[ad_1]
In the late 20th century and early 2000s, missing some of those voters hadn’t been a big deal, as white graduates and non-white graduates had voted alike. In 2016, those voters switched to Mr. Trump, and the polls had failed to capture him.
The AAPOR report offers an optimistic conclusion. The national polls had been nearly correct, overestimating Clinton’s share of the vote by just one percentage point. It was well within the range of historical survey errors. And there were obvious steps the industry could take to improve itself in the future, by including more working class voters or by weighting those who responded the most strongly.
Perhaps most importantly, according to the voting association, the 2016 experience did not suggest a systematic problem in which polls favored a party. Some years, like 2012, the polls slightly underestimated Democrats ‘share, and other years, like 2016, they slightly underestimated Republicans’ share. The report states that the direction of these failures was “largely random.”
The midterm elections of the following year, 2018, initially seemed to support this conclusion. Polls rightly suggested Democrats would win in the House, while Republicans would retain the Senate. State polls have averaged about four percentage points, which was historically normal.
The underlying details, however, contained some cause for concern. While polls in some liberal states, such as California and Massachusetts, had underestimated Democratic vote share in 2018, polls in several swing and conservative states, including Florida, Georgia, Michigan, and Pennsylvania , again underestimated the Republicans’ share.
For the second time since Mr. Trump entered politics, polls had failed to reach enough Republican voters in the swaying states that decide modern presidential elections. A third election – his reelection campaign – was looming in 2020, and it was one that millions of Americans, his supporters and critics, would follow with passion.
[ad_2]
Source link