On the one hand, there is a clear partisan divide in the responses to even factual economic questions. Take a CBS News/YouGov poll from April that found Democrats were not only 41 percentage points more likely than Republicans to say the state of the national economy was good overall, but also 29 points more likely than Republicans to correctly say that the number of jobs in the United States has increased over the past year.
There are similar fault lines on concrete questions about people’s behavior and experiences. In a recent CNN poll, Republicans were 22 points more likely than Democrats to say they would reduce their driving due to economic conditions, and 25 points more likely to report delays in receiving purchases. In another poll this spring, Republicans were more than twice as likely as Democrats to say they changed their summer vacation plans because of gas prices, 41% to 16%, eclipsing cleavages along other demographic lines. And in a Monmouth University poll released late last year, the share of Republicans saying they find it at least fairly easy to pay grocery bills fell 38 points from 2019. while the share of Democrats saying the same rose 7 points.
“There has always been a sort of partisan divide over the question of making ends meet depending on who controls the White House,” Monmouth returning officer Patrick Murray wrote. « But the huge shift in this poll, led mostly by Republicans, raises the question of whether we are measuring the primacy of partisan identity more than an accurate self-assessment of economic conditions at home. »
The same goes for broader economic perceptions. Polls have found that Democrats’ and Republicans’ relative impressions of the strength of the economy reverse soon after the 2020 election. And these increasingly visible partisan dynamics pose new challenges to attempts to gauge objective opinions. Americans.
“Unfortunately, the magnitude of the partisan divide in expectations has completely overpowered rational assessments of ongoing economic trends,” wrote Richard Curtin, who heads the closely watched University of Michigan consumer sentiment surveys, which poll Americans on the economy since the 1940s. In a January note, he noted that the partisan gaps observed on consumer expectations under the Biden and Trump administrations were about twice as large as they had been under the presidencies of Obama and George W. Bush. “This situation is likely to encourage poor decisions by consumers and policymakers. While there have always been partisan differences in favored policies, the overwhelming size and persistence of the partisan divide has generated considerable economic uncertainty. »
But while the results are obviously influenced by partisanship, the precise form this influence takes is less clear. In one reading, the divides are largely an example of partisan cheerleading — with respondents using polls as a convenient opportunity to tout their partisan allegiances, rather than a faithful account of what they believe to be the truth. On another reading, they are a sign that partisan thinking, fueled by polarization and reinforced by partisan divides in news consumption, has completely overtaken how many Americans perceive their own reality.
This ambiguity can make it difficult to determine how much people mean what they say to pollsters – a challenge that has key implications for understanding public opinion not just on the economy, but also on everything from vaccine refusal sensitivity to political misinformation. , such as the false belief that the 2020 election was stolen.
People don’t always speak literally, and they don’t always answer surveys literally either. For some Americans with strong political views, answering a poll can present an easy opportunity to verbally demonstrate their loyalty, an effect known by terms such as “partisan response” or “expressive response.” « Just as people like to support their favorite sports team and claim that their team’s players are superior, even when they’re not, » the authors of a 2013 article explained, « polls give citizens the ‘opportunity to encourage their partisan team’.
When the impulse for hyperbole clashes with partisan loyalties, it can produce results that stretch credulity on far more than economic valuations. The same painful ten in practice.
It may also complicate efforts to gauge Americans’ willingness to embrace political violence. It is a topic that has generated some disagreement among political scientists, who have argued over the extent to which a theoretical willingness to endorse a form of political violence translates into support for specific acts of violence or a willingness to personally commit such acts. And partisan envy can even drive people to deny clearly visible evidence. In a 2017 study, when Americans saw side-by-side footage of the crowds on the National Mall during Trump and Obama’s inaugurations, 15% of Trump voters said the sparser photography of the event Trump was showing a larger group of people.
« The debate over whether it’s partisan cheerleading or real partisan reasoning – where people really, really believe those things – comes down to what’s really in people’s hearts and minds. It’s ultimately kind of a really hard thing to know or get a sense of, » said Brian Schaffner, a political scientist at Tufts University who co-authored the study with YouGov’s Samantha Luks. “But the fact that as soon as the presidency moves from one party to another, all of a sudden people assess their economic situation very differently suggests that these are not necessarily sincere beliefs. … Some of this, at least, is driven by people who only use the poll to express their support for their party.”
The tendency of some survey respondents to respond expressively is not new and is not always driven strictly by partisanship. Immediately after the terrorist attacks of September 11, 2001, for example, the proportion of Americans who told Gallup that the economy was good or excellent rose from 32% to 46% in what may have been a brief period of reluctance. patriotic to criticize the United States in any way.
But in an era marked by strong partisanship, pollsters and academics have made various efforts to separate such an expressive response from genuine belief. An experiment, conducted in 2008, found that offering small financial incentives for giving correct answers to partisan factual questions (for example, how jobs numbers or the federal deficit had changed under the Bush administration ) was enough to narrow the gap between Democratic and Republican responses. The authors cited the result as an indication that « some of the partisan bias in responses to fact-finding surveys – perhaps a very large part of it – is emotional and insincere. »
A more recent effort used a technique known as the list experiment to try to gauge the share of Republicans who truly believed the 2020 election was stolen from Donald Trump. In this format, survey respondents are asked to read a list of statements and then – rather than recording their responses to each individually – count the total number with which they agree. This, at least in theory, deprives them of « the reward of saying something to a pollster that expresses a feeling rather than a belief ». The proportion of Republicans who said they believed Biden legitimately won was just as low among those who completed the slate experiment as among those who answered a more standard question, which the authors said indicated that « the Republicans report an honest belief that Biden’s election was illegitimate. »
Asking follow-up questions can help interviewers refine their understanding of the strength of respondents’ beliefs. In 2011, a Washington Post poll found that one-tenth of Americans subscribed to the false theory that Obama was born outside the United States. However, when pressed, most admitted it was just their « suspicion », rather than something they believed there was « hard evidence » for. More recently, a CNN poll found that 61% of Americans who falsely argue that Biden did not legitimately win the 2020 election say they believe there is strong evidence for their claims, with 39% saying that this is just a suspicion.
Data on people’s actual behavior can sometimes also validate polls. A useful case study for this approach has been the US Covid-19 vaccination campaign, for which there are both a plethora of surveys on how Americans are vaccinated and corresponding statistics from the Centers for Disease Control and Prevention. While neither data source is perfect, when the vaccine was first rolled out in the United States, the CDC numbers generally reflected overall vaccination rates as reported in polls — and also confirmed that relatively low among Trump voters as measured by the polls was real.
This range of evidence suggests that there is no single answer about the extent to which expressive response determines survey results. « One of the most consistent things I’ve found in my research is that none of these phenomena are all or nothing. They’re all context-specific, » said Matt Graham, a political scientist at George University. Washington, who looked at different methods. to measure the effect of the expressive response.
Nor is there necessarily a clear dividing line between a sincere belief and a sufficiently determined willingness to act as if that belief were true.
On a practical level, the difference does not always matter so much. In this year’s midterm elections, for example, a voter who honestly believes that unemployment is on the rise, and a voter sufficiently motivated by partisan animosity to act as if it is, are likely to cast the same vote.
But it’s a helpful reminder that public opinion is complicated — and the way people express their opinions about politics isn’t always entirely literal.
