A voting sign stands outside the Rock Dam Rod and Gun Club, which served as the polling place for 96 voters during the midterm election in Foster Township, Wis. (Reuters)

With the midterms over, many political observers are already eyeing the 2020 elections. As they did in this round, progressive forces hope to capitalize on the 52 percent public dissatisfaction with President Trump. They can’t look to Republicans, who continue to support the president with a healthy 88 percent approval rating. Any gains are likely to come from motivating Democrats and independents to show up and vote for Democratic candidates.

This round, some in the Democratic Party chose to emphasize economic inequality. For example, the party’s “Better Deal” stressed jobs, reducing prescription-drug costs and fighting the concentration of economic power, in an attempt to recover its former union and working-class base. But many independent political groups such as Indivisible and MoveOn instead focused their campaigns on warning about the Trump administration’s violations of long-standing democratic norms.

Recent research by Samara Klar and Yanna Krupnikov suggests that independents often shy away from identifying as partisans because they’re embarrassed by the current mood of extreme partisanship. But what counts as partisan: a focus on pocketbook issues and economic fairness or a focus on protecting American democracy from Donald Trump?

Describing Trump as a threat to democracy may be seen as nonpartisan: Some Republicans believe this to be true. For this reason, we expected that independents would be more receptive to messaging by Indivisible and MoveOn that Trump is a threat to democracy.

Here’s how we did our research

To address this question, we conducted two largely identical experiments. The first used data from 607 participants recruited through Amazon’s Mechanical Turk in March. To repeat our analysis using a more representative sample, we ran the same experiment on a sample of 601 participants collected in partnership with the online survey-sampling firm Lucid in late October. All participants in both samples identified either as Democrats or as independents; independents who leaned Democratic were classified as Democrats in the analyses.

We randomly assigned participants to read one of two passages. Half read a passage emphasizing the economic threat potentially posed by the Trump administration’s policies. This message — which echoed themes from the Democrats’ Better Deal message — emphasized that “wealth is concentrated in fewer and fewer hands, and the Trump administration is increasingly staffed with investment bankers and other fat cats who have no idea how the average American lives,” and argued that “Trump and his administration continue to bleed this economy dry with economic policies that will make the rich get richer, while the middle class disappears and the poor get poorer.”

The other half read a passage warning of the Trump administration’s existential threat to American democracy. This passage suggested that “Donald Trump is a threat to American democracy” and that “the Trump administration’s indifference to ethics, the rule of law, and the right to voice one’s concerns at the ballot box and in protests endanger the American way of life.”

Participants were then asked how likely they were to donate to a campaign, work for a campaign, vote in the primary election and vote in the general election. The Mechanical Turk experiment asked these questions about the 2018 and 2020 elections; results were identical for both the 2018 and 2020 questions, so we focus solely on the former. The Lucid experiment (conducted in late October) asked these questions about the 2020 presidential election.

As it turns out, the type of political activism didn’t make a difference; the results were always the same. We controlled for other standard variables known to predict political engagement and compared the responses of Democrats and independents. Here’s what we found.

Democrats were highly motivated to get involved no matter what. Independents were a different story.

No matter which message they read, the results suggest that Democrats were highly likely to become politically involved, and to vote. Independents, however, were far more likely to say they planned to become politically involved and to vote after they read about the Trump administration threatening democracy and the rule of law. On a scale ranging from 0 to 100, independents who read about the threat to democracy were 10 points higher (in the Mechanical Turk study) and 11 points higher (in the Lucid study) to say they’d become politically active than those who received the economic threat. In fact, independents who read about Trump’s threat to democratic institutions were roughly as likely as Democrats to say they would take some political action.

What this means for Trump-era elections

Our analysis has its limits. For example, because we focus solely on Democrats and independents, our findings don’t tell us anything about what might motivate Republicans’ political activity. Further, our experiment asked only about what respondents intend to do — which isn’t the same as measuring what they actually will do. What someone says about his or her prospective political activity two years from now isn’t necessarily the most reliable predictor of future behavior.

Still, our experiments offer some clues about how Democrats can motivate citizens to mobilize and vote their way. They needn’t worry about their base; Democrats in our study appeared highly motivated no matter what appeal they read. But independents were more mobilized by messages about threats to democracy than about economic inequality.

Why worry about independents? Well, they decided the 2018 midterms. They may have something to say in 2020, too. Those who’ve argued that the Democrats should return to “populist” economic appeals rather than on fighting for American institutions may have it wrong.

Christopher Sebastian Parker is a professor of political science at the University of Washington who specializes in American politics, in particular race and politics and reactionary political movements.

Rafael Aguilera is a PhD student in social psychology at the University of Minnesota whose interests center on political psychology, intergroup relations and attitudes toward cultural diversity.