Moving the Needle on Uncertainty: The Role of Trusted Messengers in Countering Disinformation

This test was conducted as a partnership between Equis Research and Equis Labs

Much has been said about the role of “trusted messengers” in countering false and misleading claims and narratives, but less is known about which messengers are actually most effective at moving the needle on countering disinformation and misinformation online.

To shed light on the efficacy of three messenger personas in shaping reactions to false narratives, Equis conducted a panel test of 5,516 Latino adults in the United States in partnership with Civis Analytics. The test built on takeaways from Equis’s Misinformation Poll that found high levels of uncertainty among Latinos regarding misinformation narratives online (respondents were not so much likely to believe false narratives as be unable to say confidently whether they were true or false.) This messenger test thus sought to reduce uncertainty and move people toward certainty that false narratives are false.

For this test, Equis produced six videos, each featuring the same messenger — Camilo — as a Journalist, as an Activist, or as himself with no added title or credentials, in both English and Spanish.

The false claim around which we created the counter-messaging in the videos was: “Kamala Harris refused to be sworn in on the Bible.” We chose this false narrative — with which Latinos in our Equis Misinformation Poll weren’t especially familiar and, regardless, weren’t on the whole willing to dismiss as false — to best gauge if we could move those that were unsure toward certainty. The six videos featured the actor communicating the same explainer message written as an informative, “did you know” statement. The message was framed as follows:

In defining the most successful messenger, Equis considered the messenger’s effectiveness in achieving three outcomes:

• Reducing uncertainty about the veracity of the original false claim
• Increasing belief that the original false claim was false
• Lowering belief that the original false claim was true

The Key Takeaways

Messenger Effects

1) All three of the messenger personas Equis tested significantly lowered uncertainty and increased certainty (or belief) that the original false narrative was false.

All three messengers — the Journalist, the Activist, and Camilo as himself — were effective, at a 95% confidence interval, in significantly reducing uncertainty about the original false claim.

Panel respondents were randomly assigned to watch one of the six videos or a control video, regardless of language preference. All were asked a set of questions before and after watching one of the videos to determine whether they had been exposed to the Kamala Harris false narrative prior to taking the test and whether they believed the false narrative before and after.

The Journalist in English, the Activist in English, and Camilo in English reduced uncertainty by 25% or more across the board. All three messengers in English also increased the margin of those who went from uncertain to “extremely” or “pretty sure” the false narrative was false by over 22%.

2) The English-speaking Journalist was the most successful messenger.

This messenger achieved all three of the criteria for success: he reduced uncertainty, most effectively moved people from “uncertain in what to believe” to being “extremely sure” or “pretty sure” the false narrative was false, and most capably reduced belief — by 4% — that the original false narrative was true.

Breakdowns of percent changes in uncertainty and belief are below.

Note: This chart depicts total percent changes in uncertainty and belief.

Note: This Sankey chart shows the messenger’s effectiveness at moving people from where they were before watching the videos (on the left) to where they were after watching the videos (on the right).

3) For Spanish-dominants, the Journalist in English and the Journalist in Spanish were more or less equally effective messengers.

The Journalist messengers in both English and Spanish reduced uncertainty by between 22% and 23%, and moved the “uncertain” toward believing the narrative was false at nearly equal rates of 22%.

4) Democrats were more trusting, more amicable toward, and more moved by the English-language Journalist than Republicans. But despite reporting lower levels of trust and likeability for the top messenger, Republicans were still moved by the English-language Journalist.

The English-language Journalist reduced uncertainty and increased certainty the original false claim was false with respondents of all political affiliations, yet this messenger was most effective with Democrats.

The English-language Journalist moved Democrat-identifying respondents down on uncertainty and up on belief at far higher rates than Republican-identifying respondents.

Around 66 percent of Democrats found the English-language Journalist likable, whereas only 40% of Republicans did. Notably, when it came to the question of trustworthiness, around 60% of Democrats found the messenger trustworthy, against only 23% of Republicans.

Despite the sizable difference between rates of trust and likability reported by Democrats and Republicans, the key takeaway here is that the English-language Journalist was still effective at moving all sides down on the “uncertainty” scale and up on the “certain it was false” scale.

Uncertainty at the Baseline

In conducting this test, Equis measured baseline uncertainty by language subgroups and age. Baseline uncertainty is the extent to which respondents were unsure about the veracity of the false narrative before watching our videos.

5) There were minimal to no differences in levels of uncertainty (coming into the test) for Latinos of different language preferences.

The percentage of panel respondents who came into this test uncertain was consistent regardless of languages spoken at home. Contrary to popular assumptions, it was not the case that the Spanish-dominant respondents were more or less uncertain or believing about the veracity of the original false claim.

6) Young people were much more uncertain about the veracity of the original false narrative. Older Latinos were better at saying the false narrative was false coming into the test. All were moved after watching our videos.

The group reporting the most uncertainty about the veracity of the original false narrative was Latinos between the ages of 18–34. Around 80% of respondents in this age bracket reported being unsure whether the narrative was true or false, while 9.9% reported being “extremely” or “pretty sure” it was false and 9.7% reported being “extremely” or “pretty sure” it was true.

It was the eldest group of respondents who were the most likely to report being certain that the original false narrative was false. Around 55% of respondents aged 65 or older reported being unsure whether the false narrative was false, whereas over 35% reported being “extremely” or “pretty sure” it was false and 8.9% reported being “extremely” or “pretty sure” it was true.

A full deck with findings is available here.

Recommendations

Though there is much more to explore when it comes to trusted messengers and disinformation, this test’s findings can serve to complement Latino-serving grassroots organizations’ existing strategies for countering disinformation. In particular, these findings suggest organizations would benefit from:

1. Harnessing the voices of journalists in engagement strategies by:
   a. Prioritizing promoting positive news stories from journalists
   b. Amplifying coverage done by trusted local journalists and community journalists
2. Helping journalists to approach coverage in ways that emphasize “prebunking” disinformation narratives or “explaining” tough issues relevant to Latino audiences
3. Using messengers from organizations and/or people from the community as a way to supplement the news promotion strategy
4. Finding and sticking to a lane — continuing to focus on filling information voids with “good messaging,” and assuring that you and your fact-based, contextualized messaging are present where Latinos consume information (YouTube, Facebook, WhatsApp, Instagram).

Future Tests

The findings of this trusted messenger test are only the beginning of what is out there to understand about trusted messengers’ roles in countering disinformation.

It is important to recognize that this test’s findings are centered around one particular claim, and that further testing is needed to increase our understanding of how these messengers may influence uncertainty and belief about other claims and narratives.

Looking to the future, Equis recommends not only replicating this test with other narratives, but also testing known messengers, including Latino and non-Latino journalists with recognizable names, as well as different types of messengers at the national and local levels overall.