A recent Knight/Gallup survey found that “Four in five Americans are concerned—either very (48%) or somewhat (33%)—that misinformation on social media will sway the outcome of the 2020 presidential election.” That’s an astonishing degree of agreement among the polarized U.S. public.
It’s likely the artifact of the outsized media and research attention that misinformation on social media has received since the 2016 U.S. presidential election. But fears of misinformation swaying the 2020 election are overstated—there is little evidence to suggest that misinformation has had large direct effects on voter’s attitudes or behaviors. Instead, researchers and journalists could have been examining the more important things contributing to political outcomes and polarization, such as race, ethnicity, gender, and religion—in short, social identities as they map onto partisan identities.
While a large number of research studies since 2016 have demonstrated the stunning prevalence of mis- and disinformation on social media, there remains little in the way of evidence of their effectiveness to justify the widespread concern of the American public. Despite hundreds of conferences and millions of dollars in funding, empirical evidence for the effects of misinformation on the choices voters make is lacking in comparison to the well-documented effects of race as a structuring fact of political life—which has received much less research attention despite the growing prominence of movements for racial justice.
Nevertheless, much of the research community focused on social media and democracy post-2016 has zeroed in on misinformation and disinformation—especially by strategic actors both foreign and domestic—and the role of “facts” in the public sphere. This aligns with our greatest hope for democracy: That informed citizens will make informed choices at the polls that align with their policy preferences and understandings of major public issues.
As mountains of empirical evidence have shown, however, politics does not work like that. Most people generally have little interest in, and know little about, politics, and instead rely on things such as partisanship to make voting easier. And political scientists have shown that contemporary partisan identities are aligned with social identities in increasingly stark ways along the boundaries of race and class, geographic location, and religion.
On one level, this is democratically empowering. It’s not misinformation or targeted advertising that directly drives the vast majority of citizens’ political attitudes and voting patterns. Instead, people see themselves as members of social groups and then in turn choose the parties that best represent those groups. This is part of what makes misinformation less effective than you might think: Instead of being swayed by a particular narrative (false or otherwise), people generally discern their own political identities, and those of others, to make choices at the polls.
On another level it is a great cause for concern. The two parties are generally sorted now along racial and ethnic lines: The Republican Party is a predominately white political party and the Democratic Party is multiethnic. Republicans are more religious, rural, and wealthy; Democrats more secular, urban, and poor. Mapped onto political divisions, all of these social divisions make it increasingly difficult for many of us to see the other side as good-faith democratic citizens.
Politicians and especially the president have sought to stoke these divisions for electoral gain. Central to Trump’s victory in 2016 were his direct appeals to whites holding negative attitudes about those of other races and ethnicities. Crucially, Trump made white racial identity not only salient, he gave many of these whites a very clear political choice on the ballot. The upshot is the further hardening of white racial attitudes within the Republican Party, as Democratic whites have moved toward positions that are more in-line with racial justice. Indeed, one outcome of identity sorting is that a social movement for racial justice is now clearly embraced within the Democratic Party, which opens the door for institutionalized policy-making.
But it is a lot easier to talk about people being duped into voting for Trump by Russian Facebook posts, for instance, than about us being different types of people with potentially different values and ends for polities. It is also easier than talking about the interplay between social identity, racism, sexism, and the status of various social groups—the actual drivers of politics.
This is not to say that misinformation is something nobody should worry about. It’s that the factual element shouldn’t be the focus. Journalists and voters should pay more attention to the motivations, content, and drivers of mis- and disinformation. Research on those examines how people might share information, regardless of its factual basis, to forge collective identity with others, how actors strategically utilize racial divisions to further disinformation, and how far-right groups manipulate the media to spread radicalizing conspiracy theories, in turn driving misguided searches for “the truth.” Research in this vein helps explain why and how certain groups in this country seek to deepen political and social divides, at times using misinformation as a tool.
This research on how mis- and disinformation entrenches existing divides is deeply important—but there is no current evidence suggesting direct effects on the scale of changing the outcome of a national election. The recent Knight/Gallup survey results suggest that fears about misinformation have become the focus of public attention, when what we should be grappling with and addressing are deeper divides that cut to the core of the nation. This includes the fact that throughout American history, whites have long engaged in “post-truth” politics to secure economic, political, cultural, and social power.
Future Tense is a partnership of Slate, New America, and Arizona State University that examines emerging technologies, public policy, and society.