This Poll Might Not Say What You Think It Says

The debate over "Law & Order" (not the television show) is more complicated than it seems.

Hello, and welcome to Margin of Error,* my newsletter about polling and the way it is covered. If you like this newsletter, please subscribe and spread the word.

My goal is to publish about once a week and provide a bit of a different and stepped-back look at the world of polling than you’re used to, cutting against The Narrative that you often see from Twitter and from punditry.

People’s brains are largely still broken from the anomaly that was the 2016 election, when President Donald Trump upended a lot of what we thought we knew about the polls. In a historically stable race that has so far been the 2020 general election, pundits—and news stories—obsess over whether, and how, Trump could win, even as polling averages have him trailing between 7 and 8 points.

The latest issue that has gained increasing prominence as Trump’s potential elixir is that of “law and order.” This was the focus of a recent poll from Monmouth University, as well as the central theme of a story on swing-state polling from the New York Times.

Here is the press release from Monmouth, titled “Law & Order Seen As Major Problem.” Combined with this quote from the pollster, one implied takeaway is that Democratic nominee Joe Biden should be concerned, even though it’s not really apparent why:

“It’s not clear whether Trump’s law and order message has moved the needle at all because we don’t have trends on this question. But there is some potential for softening Latino support for Biden, for example, given the racial differences in opinion among non-Republicans.”

This message belies what is more interesting. Of course, different people can have different interpretations. And, to that note, a disclaimer: I am not a pollster. But there’s some underlying data that gets less attention in this poll that seems like a flashing warning sign for the incumbent president:

• Only 24% of Americans say the president’s handling of the protests have made things better, while 61% think he has made things worse. (This includes a not-great 46-30 split among even Republicans.)

• Overall, 45% of Americans think Biden would have better handled this situation. Just 28% think he would have been worse in handling it.

• The kicker: 52% of Americans are confident that Biden can “maintain law and order” if he’s elected president, compared with 48% that say the same about Trump.

A new episode

In other words, there’s a pretty clear narrative in this poll. But it’s not the one they think it is.

I want to go back to the words I emphasized in the quote above: “Trump’s law and order message.” The media has ceded this framing to the president, seemingly for no other reason than him tweeting the exact phrase “LAW & ORDER” somewhat constantly over the past few months.

That theme of the media ceding the framing has carried through the past five years—everything from Trump’s nicknames to the organizations still doing “Trump says” headlines. There still hasn’t been much introspection as to whether the framing holds up under the surface.

Here’s the New York Times story from this weekend that I mentioned earlier. It focuses on polls it conducted with Siena College in the battleground states of Minnesota, Wisconsin, New Hampshire, and Nevada. Its dek, which is news speak for the sentence under the headline that serves to further contextualize the story, starts this way:

“President Trump has leveled scathing law-and-order attacks on Joseph Biden for weeks.”

This framing might seem innocuous. But it goes a long way toward reinforcing perceptions that the president himself is (baselessly) trying to foster. He has, for instance, seemed to encourage more violence in many situations. This has been a hallmark, from when he encouraged his supporters up protesters at his rally to famously tweeting after the killing of George Floyd that “when the looting starts, the shooting starts.”

If you dive into the polls themselves, you’ll find some contrasting themes. The following analysis focuses on Minnesota and Wisconsin, where the pollsters completed a deeper dive around crime and public safety than in the other two states. They’re also the two states of the four that have had direct connections to the issue, and they’re aptly the two states where “law and order” is seen by poll respondents as an equally important issue as the pandemic.

The poll asked several pertinent questions, but four stood out: Whether respondents thought a bigger problem was racism in the criminal justice system or riots in some US cities; which candidate they better trusted to handle the protests and violent crime; and whether or not they believe Trump has encouraged violence.

The results are illuminating:

The most interesting is the question on the issue noted earlier: whether Trump has “encouraged or discouraged” violence in America. This is not even discussed in the Times story. What is discussed is how some voters are concerned that Biden has not done enough to denounce the riots.

Indeed, those numbers should be something of a warning sign for the Biden campaign. It’s at least partially a reason why the candidate gave a recent speech denouncing rioting and violence, why he’s released an ad featuring those words in his speech, and why he has started to mention the issue more often in general.

But it’s notable that you won’t see the same level of concern trolling around the president’s problem with “encouraging violence,” especially after a weekend when he seemed to endorse extrajudicial killings of suspects.

I’m not here to cherry pick or to line edit stories, but it’s worthwhile to point out a the underlying trends that align with the overall narrative of the Times piece. The article portrays the “law and order” topic as a strength for Trump when, quite clearly, the results are at best mixed and at worst show a clear liability.

Thanks for reading Margin of Error. If you have any tips, comments, or insights about polling, email me at bplogiurato@gmail.com, or find me on Twitter @BrettLoGiurato.

If you liked what you read, share it with a friend and make sure to subscribe.

* Hat tip to Myles Udland for the title of this blog. Myles has a great newsletter, I’m Late to This, to which you should subscribe.