The 2016 presidential election has proven frustrating for many Republicans. The GOP nominee, a businessman with no political resume, has alienated key demographics, insulted prominent members of the press, and divided leading voices in the party. But despite all this, a recent average of national polls suggests he has closed the gap with Democratic nominee Hillary Clinton.
One new CNN poll has Trump ahead by 1 point in a head-to-head match-up, while the RealClearPolitics’ (RCP) average of recent polls has Clinton up by a modest 3.3 points — a lead far smaller than the nearly seven-point lead she had in late August. And while a recent Washington Post-SurveyMonkey poll shows Hillary maintaining a comfortable lead, two poll experts cautioned a measure of skepticism about that polls’ methodology.
“Mark Blumenthal and Jon Cohen are two of the best pollsters in the business,” RCP Senior Elections Analyst Sean Trende told The Stream about the Post‘s online poll. “They do not put out a junk product. That said, it is a new methodology, and even the best methodology can’t fix sampling error.”
Robert Morris University (RMU) professor Philip Harold, Ph.D., was cautious about the poll results. “They surveyed a lot of people, and included all 50 states, but that does not mean it is an accurate poll,” he explained in an e-mail. “There is really no way in this poll to get the opinions of people who are less willing to share their views, which opens it up to significant non-response bias.”
According to Harold, the poll “assumes, in other words, that there is no significant difference between the populations of supporters for the candidates, and their willingness to share their opinion. That is a significant assumption, and one which is belied by the recent history of polling for Trump support, which finds his supporters more reticent to share their opinion (e.g. throughout the primary season he consistently gained 2% more support than the polls predicted).”
“The medium of an online poll may also very well favor Clinton supporters — people who are in front of computers all day and who take online surveys are not the downscale, working class voters who are enthusiastic about Trump,” Harold said. “The Texas results of this poll show that something might be off — it shows a dead heat between Trump and Clinton, which no one really thinks is true.”
Do Third-Party Candidates Matter?
One way of measuring the race is how Trump and Clinton match up when Libertarian Party nominee Gary Johnson and Green Party nominee Jill Stein are included in the polls. RCP’s average in a four-way race found that Clinton’s lead slips to 2.4 points over Trump, indicating that the secondary candidates are taking more from Clinton’s support than Trump’s.
It’s “hard to say” whether Johnson and Stein will factor into November’s elections, said Trende. “Third parties tend to fade down the stretch, except when they don’t (e.g. [Ross] Perot [in 1992]). Given the sky-high unfavorable for both candidates, I wouldn’t be at all surprised if things didn’t fade this year.”
Harold, however, said, “I tend to think that as people’s minds are focused in the months, days and weeks leading up to election day, Stein and Johnson’s numbers will drop. They will be factors and will pick up votes, definitely. But more of a factor will be who votes.”
“Clinton’s challenge is motivating voters who traditionally have had lower turnout, but who voted in much greater numbers the past couple elections for Obama — that is difficult to do,” Harold continued. “Trump’s task is to bring those who have not voted to the polls — that is also very difficult to do.”
“Getting supporters to the polls is what keeps the Clinton campaign up at night,” he said. “They have lots of money and technology, but they do not have an inspiring candidate. The key state in the electoral college is Pennsylvania. Trump needs to win there, and he should win there, given the kind of campaign he’s running. The nice 6-to-9-point gap there Clinton has opened up keeps the Trump campaign up at night.”
In 2012, many conservatives assumed that pollsters were overstating President Barack Obama’s excitement among voters. As Hot Air’s Ed Morrissey told The Stream, “The reason we were doing this so-called unskewing of the polls, which turned out to be a huge mistake, was that we were convinced the electoral model had changed from 2008 to 2010.”
“We were wrong in our assumptions because we were making assumptions,” said Morrissey. Obama ended up beating GOP nominee Mitt Romney by the same margin as the polls predicted. But in its recent poll showing Trump with a three-point lead over Clinton, The Los Angeles Times indicated that what conservatives got wrong in 2012 could end up being correct in 2016.
Harold also says pollster accuracy is a legitimate question in 2016. “Trump won the Republican primaries by bringing in new voters to the ballot box. That makes it very, very difficult for pollsters, who are relying on the statistics from previous elections, to filter out those who do not plan to vote and to weigh their samples.”
According to Trende, however, “Different pollsters have different methodologies. So while the averages are good, there is always a chance that some particular pollster ‘figured things out.’ It might be the LA Times. But it might also be SurveyMonkey (which has HRC +6), so over time you’ll be *closest* to the correct result using averages.”
And what if the popular vote in November comes down to the wire? “If you look closely, in [the RCP] averages, in 538’s averages, and in the SurveyMonkey polls, electoral vote #270 actually falls a touch on the Trump side of the popular vote. In a tied popular vote race, Trump probably has an edge.” (For more from the author of “Polling Average Shows Trump-Clinton Race Tightening” please click HERE)