President Trump promises to be a key factor in November’s midterm elections — both because Democrats want to vote against him and Republicans want to vote to support him.
SurveyMonkey asked registered voters this question in late August, and teamed up with researchers at the University of Michigan and Georgetown University to dissect the responses, identifying the most common words voters used in explaining their support.
Across the nearly 5,000 reasons voters described, the findings relating to Trump are particularly striking. Trump was mentioned by about 1 in 7 voters, 14 percent. But there was a large gap depending on which party voters supported, with more than twice as many voters who support Democrats for Congress mentioning Trump as those who support Republicans, 21 percent vs. 9 percent.
Democratic voters' greater emphasis on Trump contrasts with results from a closed-ended question asking about Trump’s influence on voting in a similarly timed Washington Post-ABC News poll. That survey found 59 percent of voters overall saying it was “extremely” or “very” important to share a candidates' opinion on Trump, including similar shares of voters who support Democratic and Republican candidates (62 percent and 61 percent, respectively).
The SurveyMonkey poll also found Democrats and Republicans were much more likely to cite the opposite political party in explaining their support for Congress. Voters who support Republicans for Congress were 20 percentage points more likely to mention the Democratic Party in explaining their vote than the Republican Party (33 percent vs. 13 percent). Likewise, voters supporting Democrats were far more likely to mention the Republican Party than their own party (39 percent vs. 18 percent).
This suggests that, while Democrats see Trump as something to be voted against, Republicans also have their own boogeyman: The Democratic Party. And in politics, fear is often the most powerful motivator.
Even in many of their responses mentioning Trump, Republican voters cited the danger Democrats posed to him, rather than simply saying they love Trump.
“From my perspective, Democrats are in lockstep in the effort to remove Donald Trump from office by any means for reasons even they can’t clearly define,” said one Republican response that mirrored others.
Added another more bluntly: “Because liberals are nuts [and] Trump has improved the economy.”
The results are consistent with “negative partisanship,” the idea that with declining popularity of political parties, dislike or even antipathy for the opposite party is the biggest unifying force.
That kind of thinking was prominent in many of the open-ended responses by Democratic-leaning voters.
"I am against almost everything Trump represents” said a 46 year-old from Illinois who supports Democrats for Congress.
But even among Republican-leaning voters, desires to support Trump’s agenda were mixed with criticism of Democrats. “Because Democrats are obsessed with hating Donald Trump and forgotten that there is a country to run, plus, Donald Trump and the R’s have gotten this economy back on stable ground and growing,” one respondent said.
The SurveyMonkey poll showed that even while most people don’t volunteer Trump in describing their vote, voters are paying an enormous amount of attention to him. Over two-thirds, 68 percent, of registered voters said they had heard “a lot” about Trump in the previous week, and another 19 percent had heard “a little,” with Democrats and Republicans reporting at similar levels.
Democratic voters talk about winning the House, while Republicans talk about ideology and the economy
While voters’ explanations of their support for Congress focused most on Trump and the parties themselves, those supporting Democrats and Republicans differed sharply in emphasizing or ignoring various topics.
Democratic-leaning voters were much more likely than Republican-leaning voters to mention the word “House” (7 percent vs. 2 percent) often referring to their desire for Democrats to take control of the House of Representatives. More Democratic voters also mentioned “check,” usually describing their desire for a check against Trump’s power as president (2 percent vs. almost no Republicans surveyed).
Meanwhile, Republican-leaning voters were more likely to mention political ideologies, with 9 percent of Republican-leaning voters mentioning “conservative” and 3 percent mentioning “socialism.” Few Democratic-leaning voters mentioned either term (2 percent of Democratic-leaning voters mentioned “liberal,” 3 percent of Republican-leaning voters did).
Republican-leaning voters were also somewhat more likely to mention two major campaign issues. The economy was mentioned by 4 percent of Republican supporters and 1 percent of Democrats, while Republican-leaning voters were also slightly more apt to mention immigration (2 percent vs. 1 percent). These differences are small and probably insignificant, but the relative silence among Democrats on these issues signals that they are not rallying cries for their supporters.
There’s little question that Trump is playing a major role in voters’ attitudes about the 2018 election. The latest Washington Post-ABC News poll found more than 8 in 10 voters who approve of Trump support Republicans for Congress, while more than 8 in 10 of Trump disapprovers support Democrats. Among independents, Trump approval is a strong indicator of which way voters are swinging.
But despite Trump’s omnipresence in the news, most voters don’t think of their own congressional vote as being, first and foremost, about Trump.
The poll was based on interviews with 7,997 registered voters nationwide Aug. 22-28, and included 4,997 responses to the question asking why voters supported a party’s candidate for Congress. The poll was conducted by SurveyMonkey among a sample of the millions of people who take an unrelated user-generated survey on the firm’s platform, a non-probability sampling method of all U.S. registered voters. Analysis of open-ended survey responses used unweighted data. See here for more on SurveyMonkey’s methodology.
The poll and analysis above are part of a collaboration between SurveyMonkey, The Washington Post and researchers at the University of Michigan and Georgetown University. Data collection and analysis were managed by Mark Blumenthal and Sarah Cho of SurveyMonkey, with design and analysis by Josh Pasek, Stuart Soroka and Michael Traugott from the University of Michigan and Jonathan Ladd of Georgetown.
If the election for the U.S. House of Representatives were held today, would you vote ... ? Options: Definitely for the Democrat, Probably for the Democrat, Leaning Democrat, Leaning Republican, Probably for the Republican, Definitely for the Republican, Would not vote.