Friday Polls - What Are We Asking?

Asymmetric partisanship in poll responses

This week, future guest of the pod (kidding, but I am working on it) Philip Bump of The Washington Post reported on some recent Pew Research polling on people's attitudes towards the two leading Presidential candidates. Once again, the research showed that >90% of Trump supporters think he's great, 10/10, no notes. Biden supporters are also about 90% committed to voting for him in 2024.

Voters seem to be in general agreement that the stakes of this election are high - nearly 70% say it really matters who wins the election, and this is about the same regardless of whether they support Trump or Biden.

But there is an asymmetry here. Trump supporters are much more likely to profess extremely positive views about their candidate, or otherwise back his play. For example:

• 82% of Trump supporters said he was a good or great president; 57% of Biden supporters described him that way.
• 89% of Trump supporters say Biden is a terrible or poor president; 10% of Biden supporters say the same.
• 49% of Republicans think he did nothing wrong when asked about the criminal allegations against him for trying to overturn the 2020 election and another 21% think he did something wrong but not illegal; 78% of Democrats think he broke the law.
• 67% of Trump supporters are extremely or very confident he respects the country's democratic values; 74% of Biden supporters are extremely or very confident he does.
• And when it comes to the loser in the electoral college conceding his loss, there's another alarming asymmetry: 44% of Trump supporters say it's very important for the losing candidate to concede, while 77% of Biden supporters say the same thing.

What are we to make of this? On the one hand, it is alarming that Republicans and Trump supporters seem to be willing to ignore or brush aside Trump's actions after the 2020 election, especially when other polling indicates similar anti-democratic tendencies. Last fall and winter polls showed about 60% of Republican voters believed "the big lie" that the 2020 election was fraudulent. In an October 2023 poll from PRRI, 48% of Republicans said they agreed with this statement: "Because things have gotten so far off track in this country, we need a leader who is willing to break some rules if that's what it takes to set things right." Only 29% of Democrats agreed with that statement, which is alarming enough.

But then PRRI kicked it up a notch:

A third of Republicans agreed that "true American patriots may have to resort to violence in order to save our country." Only 13% of Democrats said the same – although it should be noted that both groups had increased in their support for this statement since 2021.

Bump characterized the Pew data as "warning signs are blinking" about anti-democratic impulses among one party's base of supporters.

And there is plenty of reason to worry that this is true. However, at the same time, it raises a question for me about the value of even asking these questions in our highly partisan times. If the way to be a member of a political party now is to be "ride or die" for that party, then it may be the case that these questions are no longer being answered in good faith.

As Bump put it,

There’s a level of abstraction here that’s important to recognize. Asking if a candidate you support respects democracy is a bit like asking if he is a good dude; you’re going to be inclined to say yes. Overlay the willingness of Republicans and Trump supporters to dismiss the idea that he did anything wrong after the 2020 election and you get equivalence with Biden — who has shown no similar inclination to set aside democratic determinations.

It seems like there are, in effect, two kinds of polls being conducted. One is among those who identify as Democrat/lean Democrat or as Biden supporters. They can disentangle their views about Biden from their willingness to support him for president in November and tell you what they think about his various qualities and behaviors, even if those thoughts aren't exactly glowing.

The other poll is among those who identify as Republican/lean Republican or as Trump supporters. They are unwilling or unable to pull those views apart – they intend to vote for him, so he was an excellent president, did nothing wrong, supports our democratic values, was the victim of electoral fraud, and doesn't need to concede if he loses again.

I'm not sure there's a ton of value in continuing to ask Trump supporters about the various qualities of the man – they either genuinely believe that he's the best or are engaging in expressive responding, but either way are going to say he's the best on almost every metric.

As I compare the Pew results to the PRRI survey last fall, I wonder if it isn't more productive to ask these more abstract questions about "true patriots" and "a strong leader who breaks some rules" to understand the tilt toward authoritarianism rather than grounding any question in Trump's specific behavior – while the partisan effects are alarming, they're not as dramatic as when we're talking about Trump specifically.

One of the trickier aspects of designing any survey is dealing with different subsegments and routing them through different sets of questions – it just makes writing the instrument more of a hassle, and then it makes comparing answers nearly impossible. There is value in seeing the two polls compared; but if we want to understand what's going on with Trump voters, especially to understand their tolerance for authoritarianism, political violence, anti-democratic behavior, or general lawlessness, we may have to ask Trump supporters an entirely different set of questions.

Programming note!

We were off this week for the regular show – I had to have a minor medical procedure. But we will be back on Monday with a shorter show in which I'll walk you through a brief history of studying people as data/polling. And then we have some excellent guests coming up and will be back on our regular schedule the week after. If you have any questions for our guests – or requests for topics or guests – please drop me a note. I'd love to hear from you.