优蜜传媒

PRINCETON, N.J. -- Hillary Clinton, the front-runner for the Democratic presidential nomination, has offered specific positions on a variety of issues while campaigning. But when 优蜜传媒recently asked Americans to say what they recall reading or hearing about her, one word -- "email" -- drowned out everything else.

These data were collected as part of a late August 优蜜传媒project asking Americans to say what they had read or heard recently about 22 major Republican and Democratic candidates. Each person interviewed was asked about no more than four candidates with whom they were familiar. Clinton's favorable rating among all Americans during the Aug. 24-30 field period tilted slightly negative, with 40% viewing her favorably and 50% unfavorably.

The accompanying word map uses font size to indicate the relative frequency in which specific words (excluding articles, pronouns and neutral or nonspecific words) appeared in the full list of responses for Clinton.

In the verbatim responses from about 750 U.S. adults familiar enough with Clinton to offer an opinion of her, the word "email" came up 329 times, phrased variously as "email," "emails," "email scandal," "email scandals," "that email thing," "email stuff" and "private emails." Relatedly, there were 83 mentions of "server." All of these refer to the controversy involving Clinton's use of private email servers to conduct government business while she was secretary of state.

By contrast, there were few mentions of the substantive themes Clinton has talked about on the campaign trail. For example, "economy" appeared on the list only four times, the same number as for "the middle class." "Gun control" appeared seven times, with even fewer mentions of "college" and "capital gains tax." Even the catchall descriptions "policy" or "policies" were mentioned just nine times.

Republicans' responses were overwhelmingly negative toward Clinton, and Republicans were especially likely to refer to the email issue in negative terms. Republicans also tended to mention "Benghazi," referring to questions about how Clinton handled security for the U.S. diplomatic mission in Libya after the U.S. ambassador and three other Americans were killed in a terrorist attack in 2012.

Even when the responses of the 279 Democrats and Democratic leaners surveyed are isolated in a word map, "email" trumps everything else.

These responses don't necessarily mean Americans are completely unaware of Clinton's policy proposals, where she stands on the issues or her credentials. And the finding that so many Democrats, most of whom (77%) viewed Clinton favorably at the time of the survey, mentioned "emails" is not necessarily a negative, but rather a reflection of news media coverage, for better or for worse. But the frequent mentions of the email controversy certainly suggest that Clinton's attempts to get other messages out to the public, which might either change her image or excite those who already like her, were largely drowned out by the media coverage of the email matter at the time of the survey.

For Donald Trump, Immigration Overshadows Reporter Controversies

While Clinton has been defined by a controversy she first tried to minimize and more recently has tried to quell by apologizing for, Donald Trump is primarily associated with his core campaign issue: immigration.

In verbatim responses from 718 U.S. adults familiar enough with Trump to rate him, the word "immigration" appears 99 times, making it the most frequently used substantive word or term in his word list. This is followed by the phrase "wants to," used 48 times, only four of which are followed by "be president." The rest relate to actions Trump wants to take as president, such as "wants to close the borders," or "wants to make America great again." These came at a time when Trump's national favorable rating was similar to Clinton's: 38% favorable and 51% unfavorable.

Twenty-nine people used "policy" or "policies" in response to what they have heard about Trump -- a relatively modest number, but more than was the case for Clinton.

Other often-mentioned words are "businessman," "business," "immigrants," "illegal" or "illegals," "border" or "borders" and Trump's "money" -- all referring to issues or themes Trump himself is trying to promote.

On the negative side, there are numerous references -- often critical ones -- to Trump's various "comments" or "statements" in the news, particularly his comments about Mexico and women. While it is unclear if Trump relished or wanted to downplay his quarrel with Fox News anchor Megyn Kelly, a number of people mentioned her by name, with others mentioning Fox News generally or "that female reporter." Similarly, the responses include mentions of TV reporter Jorge Ramos or his Spanish network Univision -- referring to Trump's throwing Ramos out of a press conference last month. However, immigration still largely overshadowed these matters.

Then there are the adjectives those interviewed use to describe Trump. Although his supporters consider him "refreshing," "honest," "smart" and "good," others call him "arrogant," "racist," "an idiot," "rude" and a "big mouth." His brash, take-no-prisoners style has clearly filtered through to a portion of the public, and evidently cuts both ways.

Trump's word map based on Republicans, including independents who lean Republican, is highly similar to that based on all Americans, but with less criticism of his personal style and more statements of approval -- for example, that he is "good," and "like him." This is consistent with his broadly positive image among Republicans, with 65% viewing him favorably at the time these verbatim responses were collected.

Bottom Line

Much of what has been called the narrative of a presidential campaign is interpreted through the filter of the media. Reporters, opinion column writers, editors, bloggers and social media mavens all compete to summarize what they think the stream of information about the candidates is indicating.

What's missing is a solid, empirical sense of what information about the candidates is actually getting through to the average American. These results provide an unfiltered glimpse into what information about the candidates is resonating most with the public.

The data suggest that Clinton has not been able to control the messaging about her and her candidacy, given that the email controversy is the information about her that has been most likely to filter through to the average American.

For Trump, on the other hand, a core component of his campaign -- his stance on immigration -- is registering with Americans, along with various reactions to his abrasive yet highly reaction-generating style. When Trump is on stage with 10 other GOP candidates in Wednesday night's debate, the way his behavior comes across to viewers will factor into whether he is able to sustain or build on his current momentum.

In some ways, these findings may be counterintuitive. The policy-oriented Clinton was defined most heavily at the time of this survey by controversy stemming from past conduct, while the political novice Trump was better known for his policy positions, namely his focus on immigration.

This analysis has focused on the two candidates with the highest overall name identification at this point in time; future reports will look at the images of other candidates.

Survey Methods

Results for this 优蜜传媒poll are based on telephone interviews conducted Aug. 24-30, 2015, on the 优蜜传媒U.S. Daily survey, with a random sample of adults, aged 18 and older, living in all 50 U.S. states and the District of Columbia. Hillary Clinton's results are based on 1,095 national adults asked to give their opinion of her as favorable or unfavorable, and 754 with an opinion who were asked to say what they had recently read or heard about her. Donald Trump's results are based on 1,058 national adults asked to give their opinion of him as favorable or unfavorable, and 718 with an opinion who were asked to say what they had recently read or heard about him.

Each sample of national adults includes a minimum quota of 50% cellphone respondents and 50% landline respondents, with additional minimum quotas by time zone within region. Landline and cellular telephone numbers are selected using random-digit-dial methods.

Learn more about how the works.