Can we trust polls in 2020?

Can we trust polls in 2020?
© Greg Nash

President TrumpDonald John TrumpTrump opens new line of impeachment attack for Democrats Bloomberg to spend 0M on anti-Trump ads in battleground states New witness claims first-hand account of Trump's push for Ukraine probes MORE, the national polls tell us, is at an historically low level of sustained disapproval. Most polls indicate that Joe BidenJoe BidenBloomberg to spend 0M on anti-Trump ads in battleground states New witness claims first-hand account of Trump's push for Ukraine probes Obama cautions 2020 hopefuls against going too far left MORE currently leads his closest Democratic rivals in the campaign by a comfortable margin. The polls also show that virtually every Democratic candidate in the presidential field ties or defeats Trump. Even Marianne Williamson beats Trump, according to a poll of people who live with Marianne Williamson.

Unbelievable. I mean, literally unbelievable, at this point anyway. Have we, Democrats and Republicans alike, forgotten how polls have jilted us in the past? All of those unkept promises? The teasing and luring us into an embrace of certainty and security? Polls have broken our hearts before.

Why would we go back now? If you analyze the right trends in the right polls over an extended period of time, you can sense certain forecasts. Not the forecast of a hurricane hitting Alabama, which would be dead wrong, but some useful patterns to get a picture of where voters stand.

ADVERTISEMENT

I invited Cornell University professors Peter Enns and Jonathon Schuldt, who are affiliated with the Roper Center for Public Opinion Research, to join a conference call that was open to the public, to share what they are seeing. The Roper Center is the leading educational facility in the field of public opinion and has the largest archive of survey data in the world. It was hosted by the nonpartisan Institute of Politics and Global Affairs at Cornell University. In standard polling parlance, here are the toplines.

Given the state of polls in 2016, and following the ultimately incorrect forecasts of a win by Hillary Clinton, can we trust polls going into 2020?

Most surveys in 2016 actually got the popular vote spot on. It was the Electoral College that was not accurately forecast. We often interpret polls to forecast, but forecasting is not always the intent of polling. Schuldt said that he “puts less stock in forecasting models and more stock in places that aggregate high quality polls.” He mentioned FiveThirtyEight run by Nate Silver and the Real Clear Politics average of polls. But the point is that “a single poll is a single poll, and it is helpful to understand the averages and trends to truly get a sense of where the public sentiment is.”

Is the Trump base the biggest ever, or more like an appendage?

Schuldt explained that while it is true that the approval rating of Trump is very high among “strong Republicans, there is an important nuance to consider. If you break Republicans into different groups, depending on how strongly they affiliate with their party, you see rather large variability in their approval.” Trump performs very well among strong Republicans, but his support is more tepid with voters who identify as Republicans or “leaners,” or Republicans who say that they “somewhat approve” or “approve” of his job performance, so the entirety of his support is not quite as intense as popular narratives suggest. Enns and Schuldt estimate that his core base is 15 percent to 20 percent of the national electorate.

ADVERTISEMENT

Do tweets with racial overtones work in favor of the president?

Not with the voters he ultimately needs for reelection. A percentage of the electorate has attitudes that lean in the direction of racial resentment, but they are outnumbered by voters across all parties who lean strongly in the other direction. The question here is whether racially charged invective will create enough of a backlash with those currently undecided voters.

In a competitive Democratic primary with numerous candidates, are Democrats looking for ideological affinity or pragmatic electability?

The aggregate data makes it pretty clear that for most Democrats, 2020 is about defeating Trump and is not a choice between party policy proposals or ideological affinity. The bottom line is that if you decide to flirt with polls going into 2020, make sure that you have an honest and long term relationship. In politics as in life, it is best to save yourself the heartbreak.

Steve IsraelSteven (Steve) J. IsraelThe bizarre circle of Donald Trump, Rudy Giuliani and Mike Bloomberg The Tea Party has died of hypocrisy Specter of Nixon impeachment looming over Republican Party MORE represented New York in Congress for 16 years and served as the chairman of the Democratic Congressional Campaign Committee from 2011 to 2015. He is now the director of the Institute of Politics and Global Affairs at Cornell University. You can find him on Twitter @RepSteveIsrael.