Conventional wisdom about the 2020 presidential election says that Democrats are laser-focused on removing President Trump from office — so much so that most are willing to give up the policies they would prefer in exchange for victory. Apparently any Democrat who can win will do.

It’s easy to see why this might be true. Trump’s overall approval ratings are not great, but Democrats disapprove of his performance at astonishingly high rates. A recent Washington Post-ABC poll found that 95 percent of self-identified Democrats disapproved of the president.

The desire to take out Trump may have prompted Democratic voters to seek out a candidate who is — above all else — electable. According to journalists covering the presidential race, electability has been “the most important factor in the 2020 Democratic presidential primary.” As two journalists wrote in the Atlantic last fall, pundits have “made ‘electability’ into a hyper-focus of 2020 coverage … rendering it a hackneyed cable-news buzzword that launched a thousand takes.”

Surveys show that Democratic voters are emphasizing electability

Evidence for Democrats’ fixation with electability comes largely from surveys of voters. In a recent YouGov national survey of likely Democratic primary voters, respondents were asked to choose whether it was “more important” to select a nominee who “agrees with your position on most issues” or “can win the general election in November.” Sixty-five percent prioritized electability, and only 35 percent chose policy agreement.

This mix, in which one-third of voters pick issue purity and two-thirds favor electability, has been seen in other surveys, too. Exit polls from the Democratic primary in New Hampshire show that only 33 percent of voters said they prefer a nominee who agrees with them on major issues, while 63 percent prioritize a nominee who can defeat Trump.

If electability is the overriding motivation for Democratic voters in 2020, it makes sense that total support for all the moderate candidates has consistently added up to more than the percentages behind the most progressive candidates, Sens. Bernie Sanders (I-Vt.) and Elizabeth Warren (Mass.).

Reasons to doubt the electability fixation

But is electability really all Democrats care about? Let’s examine the evidence. A HuffPost survey conducted last month found that only 22 percent of Democratic primary voters planned to vote for a candidate who is not their “favorite” but has a better shot of winning. “Strategic voting” of this sort always happens to some degree in a multi-candidate primary race and is not unique to 2020.

Standard survey questions may be misleading about how widespread the desire for electability is this year. They give Democratic voters a forced choice: caring only about issues or about electability, not any other options. Based on what political scientists know about primary voting, this either-or choice is likely to mask a more complicated and wider set of considerations.

To explore this possibility, my colleagues at the University of Wisconsin at Madison’s Elections Research Center and I presented Democratic primary voters with a longer menu of reasons for their choice of candidates. In statewide surveys of 3,600 adults across the battleground states of Michigan, Pennsylvania and Wisconsin, we asked respondents whether they planned to vote in their states’ primaries.

For those who said yes, we asked which candidate they preferred and why they selected that person. Compared with the forced dichotomy used in so many other surveys, our survey provided a list of six options for the “main reason” they preferred the candidate. Their choices were:

  1. They are the most qualified to be president.
  2. They have the best chance of beating Donald Trump.
  3. I know more about them than any of the other candidates.
  4. I agree with them on most issues I care about.
  5. I like their political style.
  6. They are a different kind of candidate.

The last category included other reasons respondents might offer.

If electability is really the driving force in 2020, voters should feel free to ignore the other options and say that defeating Trump is their top consideration. They did not.

As you can see in the figure below, the largest group of voters — 37 percent — chose agreement on the issues as their reason for voting for their candidate. That’s similar to the one-third share seen in other surveys.

But the remaining two-thirds choose other options. Only 22 percent of Democratic primary voters say electability mattered most. Another 20 percent said qualifications for office are most important. Smaller numbers of voters said they were drawn to a candidate’s style or wanted to support a “different kind of politician.” These results indicate that many respondents in previous surveys artificially picked electability as a top motivation simply because it was the only alternative to issue voting.

As you might expect, voters who like Sanders and Warren, known for their progressive policies, were most likely to say that what attracted them was agreement on the issues. Former vice president Joe Biden’s supporters say they picked him because he is most qualified. Backers of former New York mayor Mike Bloomberg emphasize electability. We ran our survey before former South Bend, Ind., mayor Pete Buttigieg and Sen. Amy Klobuchar (Minn.) dropped out of the race. Likely voters who supported them said they did so not primarily because of electability but because they agreed on the issues.

Of course, primary voters make their choices for a wide variety of reasons. Many probably cannot easily separate one reason from the others.

Voters’ reasons for picking a candidate will surely fluctuate, as will the relative importance of being able to defeat Trump. But anyone who comments on the race to win the Democratic presidential nomination may wish to be cautious in declaring that electability is the only thing primary voters care about, leading them to push their policy beliefs aside for the sake of victory.

Barry Burden (@bcburden) is a professor of political science and director of the Elections Research Center at the University of Wisconsin at Madison.