On Latinos, Misinformation and Uncertainty: New Polling Insights

A collaboration between Equis Research and Equis Institute

Equis regularly monitors a variety of narratives that proliferate in digital spaces. We do this to understand what kind of information Latinos interact with online, including disinformation, misinformation and hyper-partisan content, all in both English and Spanish. (Our latest narrative landscape report is available here.)

But beyond tracking this content, Equis wanted to understand the pervasiveness of false information circulating in Latino information spaces, and the extent to which false narratives — whether spread with intent to deceive or not — take hold.

To that end, we surveyed 2,400 Latino adults in the United States earlier this year. Our takeaways and recommendations are below.

For the purposes of this study, we focused on misinformation: false or inaccurate information spread without proven intent to deceive or where intent cannot be ascertained. We chose narratives that fact-checkers would agree are explicitly false. Each respondent was presented with eight different narratives, randomly selected from four buckets:

• COVID-19 (e.g., “kids can’t get sick from COVID-19”)
• Right-wing narratives (e.g., “Trump won the 2020 election”)
• Left-wing narratives (e.g., “Trump faked COVID-19”), and
• Non-partisan narratives (e.g., “the moon landing was faked”).

We also used four non-partisan statements that are conspiratorial-sounding but true (e.g., “the US National Security Agency (NSA) has authority to spy on Americans”), and an additional placebo statement crafted for the purposes of this study.

For each narrative, we asked respondents whether they had heard the statement, and how confident they felt the statement was true or false, even if they hadn’t heard it before.

Our biggest takeaways from this exercise:

• The landscape of misinformation is characterized both by a high level of familiarity with the narratives we tested and also a high level of uncertainty about whether these narratives are true or false;
• In fact, we walked away believing that uncertainty is a much more important focus of attention than belief;
• As such, we also saw the potential strategic value, counter to some conventional guidance, of airing out and debunking certain narratives;
• And we concluded that “susceptibility” is not the right framework for understanding (or countering) disinformation, misinformation or propaganda.

There was a high-level of familiarity with the narratives we tested.

Half of the narratives we tested were familiar to more than 50% of Latino respondents, and some — like the Big Lie (the false narrative that claims the 2020 election was stolen from Donald Trump via fraud), and vaccine misinformation — reached upwards of 70%.

At the same time, there is a great deal of uncertainty about whether these narratives are true or false — as we’ll come back to momentarily.

The world is not quite upside down yet. Generally speaking, we found that false statements tended to penetrate — achieve both familiarity and belief — less than true ones. In fact, our placebos were the least likely to penetrate.

The crisis, as always, is on the margins, where dangerous narratives can do big damage among even small slices of the population. False COVID-19 narratives, on the whole, penetrated less than others (though we cannot ignore that 12% of Latino adults had heard and believe the extreme claim that “the COVID-19 vaccine is more dangerous than the COVID-19 virus itself.”)

That said, the greatest crisis is one of trust. The following figure gives us a closer look at the levels of uncertainty among respondents presented with false narratives.

The people we’re calling uncertain are those who said they weren’t sure if the narrative was true or false, or who only leaned slightly in one direction or the other — they thought the narrative could be true/false but they weren’t willing to go so far as to say they were “pretty sure” or “very sure.”

For more than half of the narratives, the “uncertain” are the majority group, and in some cases the supermajority. Even where they are not the majority, they remain a healthy minority, even for such seemingly “out-there” narratives such as the “vaccine microchip” or the “flat-Earth” conspiracies.

In our view, this widespread uncertainty represents both a threat and an opportunity for those who want to see an informed populace. When people are flooded with questionable news, it threatens to undermine trust in even true statements from credible messengers. But uncertainty also signals a healthy skepticism and an openness to persuasion.

In our analysis, we identified a “high priority” group: those who are highly exposed to misinformation and retain a high level of uncertainty (21% of the sample). These are people who inhabit the spaces where misinformation circulates but who maintain some skepticism about what they are seeing and reading. As such, we think they are particularly in need of (and open to) factual information from trusted sources in order to sort out fact from falsehood.

In terms of demographics, we find that this high-priority group is younger, disproportionately female, more Spanish dominant, less college educated, and lower on social trust than the rest of the sample.

In addition, when we look at media consumption among the priority group, we do not find them over-indexing on fringe media sources. Instead, they are most likely to be getting their news and information on mainstream online platforms — especially Instagram, YouTube, Google search, and TikTok.

Who believes the false narratives? Our poll upends the popular and often-damaging stereotypes of “susceptibility.” It is not the case that Latinos who are less educated or less politically informed are the most credulous. Instead, it is politically engaged Latinos — also the most likely to be college-educated and affluent — who have higher levels of both exposure to misinformation and belief in false narratives.

The results suggest that interest in and engagement with politics, congruent with the findings of our earlier research with Harmony Labs, positions some people to encounter more misinformation, and in turn they are more likely to have confidence in their own conclusions about the veracity of that information, even if their conclusions are incorrect.

This is borne out in the kinds of narratives that tend to be believed: partisan narratives have greater penetration than either lies about COVID-19 or less-political conspiracy theories.

One conclusion: hard partisans are quite willing to believe falsehoods about the opposing party. This is true of partisans on both the right and left: a significant number of highly-engaged Democrats were willing to believe anything said about Donald Trump; at the same time, engaged Republicans in our poll were on the whole more likely to believe propaganda about Democrats than Democrats were to believe right-wing narratives.

Meanwhile, low-engagement Latinos respond (quite reasonably) to the misinformation they encounter with a greater level of uncertainty. They don’t know what to believe or where to turn for accurate information and that, we think, is just as dangerous (if not more so) than the smaller group of people who come to believe the misinformation they encounter.

It cannot be said loud enough: those who are unsure of what to believe are not being fooled, they are reacting skeptically to new, strange or contradictory information presented as fact, as they should.

Among those Latinos who aren’t familiar with the narratives we tested, uncertainty is highest about claims of a partisan nature.

The poll also found that the most widespread false narratives are often the most likely to be rejected: among those exposed, uncertainty declines as the narrative becomes more widespread. This could be the case, we suspect, because these narratives have been publicly vetted — fact-checked multiple times and debunked by multiple outlets.

Conclusions & Recommendations

Countering disinformation, misinformation, and propaganda requires a multi-stakeholder approach that elevates the importance of engaging Latino communities in the United States. While there has been an increase in awareness about Spanish-language disinformation and misinformation — including about the narratives circulating online and the insufficient steps platforms have taken to moderate and reduce the spread of this content before it goes viral — more can be done to understand how far false and misleading narratives are spreading in Latino spaces online, how these impact communities across state lines, and how trusted messengers can play a part in reducing uncertainty and changing perceptions.

Platform Accountability: Ahead of this year’s midterm elections, social media platforms must ramp up efforts to reduce the spread of Spanish-language disinformation and misinformation. They can do so by expanding their Spanish-language resources and at the bare minimum, take the same actions on Spanish content as they do for English content. Social media platforms can improve machine learning models to better detect inauthentic coordination and networks spreading Spanish-language disinformation, hire more culturally fluent Spanish-language moderators, make sure they have effective and efficient integrity systems in place for non-English languages, have more Spanish-language fact-checking partners, and increase their transparency on content moderation practices.

Fact-Checking and Filling Information Voids: Media outlets, fact-checkers, and civil society, for their part, should explore new and creative avenues to disseminate fact-checked, accurate content using social media platforms most used by Latinos in the United States, including YouTube and WhatsApp, something Latin American civil society and media have implemented in various countries. Building a robust and year-round digital media infrastructure for Latinx voters, including in Spanish, is also crucial. Too many Latinos feel left out of mainstream American society and politics, and Spanish-language information voids are filled with right-wing disinformation and problematic narratives, which has resulted in little sense-making media and information for new Latino audiences. The media and civil society have an opportunity to engage the Latino community by investing in creating organic content and online distribution channels, building trust with Latinos, and using trusted messengers to deliver accurate information to the Latino community.

Correcting Stereotypes: Finally, as policymakers, community advocates and counter-disinformation experts, we must be careful about how we talk about “susceptibility” to disinformation and misinformation. Latinos in the United States are not a monolith and neither are they gullible. In the counter-disinformation space, what Latinos are is underserved.

For political and civic groups who are concerned about the effects of propaganda on Latino voters: you would be well-served to avoid language in your communications that seems to characterize as dupes those whom you are trying to bring to your cause. The focus of your attention should be on out-competing the purveyors of false or misleading narratives: aim to persuade your target audiences by proactively meeting them where they are, both in terms of what they care about and where they consume media.

For more up-to-date findings on misinformation and disinformation, please follow the Digital Democracy Institute of the Americas (DDIA), a non-profit, non-partisan organization that connects insights and actors from across the Americas to shape a more participatory, inclusive, and resilient digital democracy for Latinos.