Governance poll: Pre-COVID-19 America had growing confidence in the economy and rising sexism

Presidential campaigns offer a unique opportunity to observe how individuals respond to political events and messages.

The GW Politics Poll contained two separate cross-sectional studies that help us evaluate how voters’ opinions changed during the course of the 2020 Democratic presidential primary. The first poll was fielded between September 26 to 30, 2019, and the second between February 3 to14, 2020. In each survey, a stratified sample of 1,200 individuals were interviewed.

More than four months of the political campaign separate the two surveys. During this time, candidates spent hundreds of millions of dollars on advertising (including during the Super Bowl), made media appearances, and campaigned heavily in-person. Voters might have tuned into four or five nationally televised Democratic primary debates over the course of the four months between surveys. Importantly, President Trump remained an active campaigner through his rallies, advertising expenditures, and near-constant Tweeting. And not to be overlooked was the impeachment, trial, and acquittal of President Trump, which also occurred during this time window.

So did all of these events impact voters, and if so, to what extent?

Improving perceptions of the economy

In the GW Politics Poll, voters were asked if they thought the economy was getting better, worse, or staying the same. Despite evidence that perceptions of economic performance are strongly colored by partisanship (see here, here, here, and here), voters of all stripes perceived the economy as getting better.

In September, 66.3 of Republicans thought the economy was getting better, and that number jumped to 84.3 percent in February. Fewer Republicans over time thought the economy was staying the same or getting worse. For Democrats, only 7.3 percent thought the economy was getting better in September, compared to 9.3 percent in February. But the biggest changes for Democrats were among people who thought the economy was getting worse (which declined 55.0 to 45.5 percent) and those who said it was staying the same, which increased from 37.7 to 50.2 percent. Over time, Independents were more likely to say that the economy was getting better or staying the same, and fewer said it was getting worse.

The unemployment rate during the time of September 2019 through January 2020 remained very low and stable, between 3.5 and 3.6 percent. During the fourth quarter of 2019, the economy was healthy and stable—GDP growth was at 2.1 percent. . As the campaign progressed, more and more people perceived its health, regardless of their party affiliation.

This was a significant disadvantage for candidates championing “revolution” (in the case of Bernie Sanders) or “big structural change” (in the case of Elizabeth Warren) to address income and wealth inequalities. These were hard cases to make when more and more people saw the economy as improving.

It seems that the focus of the campaign further strengthened Republicans’ views of the health of the economy. During this time, the Trump campaign was very active in trumpeting economic growth and the unemployment figures — especially unemployment among blacks and Latinos. The greatest effect was to crystalize positive opinions of Trump’s stewardship of the economy among his most loyal supporters.

Signs of rising sexism

In mid-January of 2020, a news story broke in which Bernie Sanders was reported to have told Elizabeth Warren in a private meeting that he did not think that woman could win the presidency. The story featured prominently in the Democratic debate the following day. Despite Hillary Clinton winning the Democratic nomination in 2016 and outpacing Donald Trump in the popular vote by nearly three million votes, the story brought to the surface concerns about lingering sexist attitudes in the electorate.

One might think that a campaign featuring nationally prominent Senators Elizabeth Warren, Amy Klobuchar, Kirsten Gillibrand, and Kamala Harris, would serve to reduce overt sexism in the polity, but in fact, the reverse was true.

The GW politics poll sought to gauge sexist attitudes. It asked respondents the degree to which they agree or disagree with the following statements:

• Many women are actually seeking special favors, such as hiring policies that favor them over men, under the guise of asking for “equality.”
• Women are too easily offended.
• Women seek to gain power by getting control over men.
• When women lose to men in a fair competition, they typically complain about being discriminated against.

The answers to these questions were then calculated into an index ranging from zero (least sexist) to one (most sexist). Over time, the index of sexism increased from an average of .451 in September to .509 in February.^[1] But this increase is sexism overall is due almost completely to the change for Democrats, whose sexism index score increased from .278 to .309.^[2] Sexism among Independents increased slightly over time, and sexism among Republicans decreased slightly (although not significantly) from September to February.^[3]

Interaction with a presidential campaign featuring prominent women candidates did not ameliorate sexist attitudes, instead, it fed them. It is possible that the elevation of women candidates into the harsh spotlight of the presidential nomination contest, and the resulting scrutiny of women candidates’ personal characteristics, activated latent sexist predispositions. As research on women candidates has shown, the type of personal characteristics that are on display in a competitive campaign environment do not often match the sexist stereotypes that voters have about women and femininity more generally.

Additionally, many pointed to the risk of nominating a woman to stand up to President Trump in the general election. As this became an increasing concern, whether it was actually vocalized by Bernie Sanders to Elizabeth Warren or not, the issue of electability was an important one for Democratic voters.

Taken together, these two analyses show us that when the campaign ramps up, the effects on voters may not be as anticipated. In one case, the campaign helped to crystallize opinion about the health of the economy. In another case, the spotlight on female candidates hurt their chances among the very voters to whom they were attempting to appeal. Future iterations of the GW politics poll in 2020 will help us to tease out other impacts of the campaign as it evolves over the course of the year.

^[1] Increase statistically significant, (F(1, 2393) = 21.928, p < .001).

^[2] Increase statistically significant, (F(1, 759) = 6.027, p < .05).

^[3] Neither difference is statistically significant.