President Trump departs the White House in February for a trip to his private Mar-a-Lago resort in Florida. (AP)

It is well known that Democrats and Republicans have different views of the world. From current affairs to specific facts, Americans’ opinions are sharply delineated by their political party.

But, since President Trump’s election, Americans also have become more polarized about something different: their views of the future.

Over the past three years, I have been asking Americans about the probability of bad events. Using a University of Chicago research grant, I fielded four surveys between September 2015 and January 2018. In each survey, respondents used a 100-point scale to estimate the likelihood of:

  • an economic recession in the coming year
  • a major terrorist attack on U.S. soil in the coming year
  • a catastrophic disease outbreak (like Ebola or Zika) in the next year
  • war with China or Russia in the next five years.

On the scale, a zero indicates the event has no chance of happening, and a score of 100 indicates it is certain to take place. Respondents indicated their estimates along this range.

On the whole, Americans are fairly pessimistic about the future.  In the January survey, on average Americans said that there was a 48 percent chance of a recession and a 50 percent chance of a major terrorist attack in the coming year. Americans also said that there was a 41 percent chance of a major war with China or Russia and a 43 percent chance of a disease outbreak.

These estimates have been fairly constant over time. Although Americans have become slightly more pessimistic about disease and war, they also have become less pessimistic about a recession and terrorist attacks, at least since March 2016.

But there have been striking divergences based on partisanship, especially since the 2016 election. When Barack Obama was president, it was Republicans who were slightly more pessimistic about the future. Here is the estimate chance of a recession, for example:

After the 2016 election, Republicans became infused with greater optimism. In March 2016, Republicans estimated that the economy had a 55 percent chance of entering a recession. By December, this dropped to 38 percent.

The same polarization has happened in the estimate of wars and terrorist attacks.

Before the 2016 election, Republicans were more likely to predict wars or terrorist attacks than either Democrats or independents. But after the election, Democrats and independents became more pessimistic about wars, disease and recessions. For example, in March 2016, Democrats estimated only a 29 percent chance of war with China or Russia; by December 2016, this jumped to 44 percent. Meanwhile, Republicans became less pessimistic about terrorist attacks: their estimate dropped from 65 percent in March 2016 to 49 percent in January.

The same partisan polarization affected even views of a major disease outbreak — as Republicans became less worried about this after Trump’s election.

What is most striking about these results is not that Republicans are separated from Democrats but that they are separated from independents, too. The only exception is with fears of terrorist attacks. When Obama was president, Republicans were much more pessimistic about attacks than the rest of the nation. All Trump’s election did was bring them back to the national average.

In short, Trump’s election has divided Americans in their views of the future. Trump’s presidency may be making Republicans feel more optimistic, but the rest of the country is feeling worse.

Eric Oliver is professor of political science at the University of Chicago and co-author of the forthcoming book “Enchanted America: How Intuition and Reason Divide Our Politics” (University of Chicago Press).

Methodological note: The surveys were conducted in September 2015 (1,275 respondents), March 2016 (1,147 respondents), December 2016 (1,347 respondents), and January 2018 (1,018 respondents). The respondents were provided from Qualtrics, Survey Sampling International, and Lucid.  The results were weighted by age, education and race from the 2014 and 2016 Current Population Survey to approximate a nationally representative sample.