Observatory on Social Media (OSoMe)

Tracking public opinion about unsupported narratives in the 2020 Presidential election

Wave 1, 20th Aug - 7th Sep 2020

All elections feature claims and counterclaims by contending parties. The 2020 election seems to be an especially salient example, as conspiracy theories and a variety of misleading narratives, many of which can be debunked with professional fact-checking, are rampant.

At Indiana University’s Observatory on Social Media (OSoMe), our team is tracking widely circulated narratives throughout the 2020 election, with the goal to assess the public’s awareness of them, the extent to which they are believed, and whether one’s political leaning is related to vulnerability to these narratives.

This document reports data from Wave 1 of our study, conducted 20th Aug- 7th Sep 2020.


For the first wave, we selected five narratives through a deliberative process of monitoring issues that were becoming relevant in the weeks leading up to data collection and using tools developed by OSoMe to look at the diffusion of narratives across social media.

We showed respondents a screenshot from social media that represented a trending narrative. We asked: “Have you encountered this, or similar stories about this issue, on social media or the internet?” In addition to asking whether respondents had seen the story, we asked, about each story: “To what extent do you believe the following statement is true?”

The statements were:

Joe Biden is not mentally fit to be President.1

Kamala Harris is not a natural-born US citizen.2

Dr. Anthony Fauci funded a lab in Wuhan to develop the coronavirus.3

Joe Biden’s family has illegal business ties with China.4

Mail-in ballots cause election fraud.5

Figure 1: Network diagram of belief in unsupported narratives


Our results show that large segments of the population have encountered these narratives through media exposure, and find them plausible.  For example, 78.9% of participants in our survey were aware of at least one of these five stories, whereas only 21.1% had not heard about any of them. It is also worth noting that a majority (61.1%) believe at least one of the five narratives; 38.9% do not believe any of them.  Figure 1 is a network diagram, showing similarities across all respondents when all five narratives are considered. Respondents at the extreme left of the figure rejected all the narratives; those who believed all five narratives are shown on the extreme right.  In between are clusters of respondents who believed one to four stories.

The five stories were not equally known or believed among the participants. Figure 2 shows that the Kamala Harris birther narrative was one of the two most widely known (54.7% aware), having been frequently covered during the survey period, but also the least believed (27.7%). Mail-in votes causing voter fraud was the most believed narrative (46.4%), and was as widely known as the Harris narrative. In two of the five narratives, the number of participants who believed the stories was more than the number who reported to have encountered them online. For example, 46.1% of participants believed that Joe Biden is cognitively compromised, whereas 43.1% reported that they had seen the story.

Figure 2

Comparison by party affiliation

More self-identified Republicans and Independents believed all five narratives than Democrats. Most striking are the differences between Republicans and Democrats when it comes to endorsing the two narratives about Biden and mail-in fraud.

Figure 3


This is the first of a six-part series of reports tracking the diffusion of misinformation in the 2020 U.S. presidential election. Data in this wave were collected from an online panel of American adults, collected by Qualtrics. Data were collected from 20th Aug to 7th Sep. The sample size was 597 (margin of error ≈ 4%). The sample was 52% female. The average age was 47.45, with a range of 18 to 88. The sample was 63% white, 13% African American, and 15% Latino.

View the data


The Observatory on Social Media is a joint project of the Network Science Institute (IUNI), the Center for Complex Networks and Systems Research (CNetS) at the Luddy School of Informatics, Computing, and Engineering, and the Media School at Indiana University.