Adapted from “Is Journalistic Truth Dead? Measuring How Informed Voters Are About Political News,” by Andrea Prat, of Columbia Business School; and Charles Angelucci, of MIT Sloan School of Management.
Key takeaways:
- Approximately 47 percent of the study participants were able to confidently identify true news stories, about 3 percent believed fake stories, and the remaining half were uncertain.
- Socioeconomic factors such as income, education, age, gender, and ethnicity predicted participants’ ability to distinguish truth more significantly than political alignment. These factors alone accounted for up to 18 percent of the variation in discernment.
- While partisan bias had a measurable effect on news selection, its impact on participants' ability to assess accuracy was relatively minor (around 2 percent), suggesting that political leanings are less influential in determining beliefs about news accuracy compared to socioeconomic status.
Why the research was done: At the 2024 World Economic Forum meeting in Davos, risk specialists identified misinformation as one of the greatest threats currently facing humanity. With debates swirling around the “death of journalistic truth” and the polarization of factual realities, there’s mounting need for empirical data on the public’s ability to distinguish true news from false news.
“The ʻparallel universes’ hypothesis of journalism is the idea that, depending on whether you lean left or right politically, you have different facts,” says Andrea Prat, the Richard Paul Richman Professor of Business in the Economics Division at Columbia Business School. “This has created a lot of concerns. Some law scholars have even said that press freedom is perhaps something we should rethink in light of it.”
Prat’s research aimed to test the death of truth and the parallel universe hypotheses by measuring individuals’ ability to assess news accuracy across various demographics, thus contributing to a more nuanced understanding of misinformation’s impact.
How it was done: Over a period of 11 months, researchers conducted a series of 11 surveys, engaging a total of 15,000 participants through the polling company YouGov, which is renowned for its focus on political data. Each survey presented subjects with six news stories — three true and three false — and participants were given a short time to read the material.
To ensure the authenticity and relevance of the news articles, true stories were selected by a panel of professional journalists, while false stories were sourced from Snopes or crafted by journalists to be plausible yet fictitious.
Participants were incentivized by monetary rewards to encourage earnest participation. This design aimed to minimize the “cheerleading effect,” ensuring responses were genuine and not merely expressions of partisan preference.
What the researcher found: The study revealed significant insights into how well different demographics of Americans can distinguish true news from misinformation. Notably, about 47 percent of participants could confidently tell the difference between true and false stories. Conversely, only 3 percent were likely to confidently choose false stories. A considerable portion — the remaining 50 percent — were uncertain.
“Our results cast doubt on the ʻdeath of truth’ narrative,” says Prat. “For at least half of Americans, it’s simply not true that they don’t know what’s going on. Only very few people believe fake stories. This suggests that active misinformation is not the real problem, or at least doesn’t seem to be the biggest problem here. A lack of information seems to be the bigger issue.”
Interestingly, the research indicated that socioeconomic factors were significantly more predictive of an individual’s ability to identify true news than partisan alignment, which accounted for only about 2 percent.
Specifically, the probability that a subject would select a true news story could be predicted in order of impact by age (above median), education (college or higher), gender (male), income (above $60,000), and ethnicity (white). In other words, the likelihood that an older, high-income, college-educated white man could identify a true news story was 18 percentage points greater than participants with five complementary characteristics — nine times the “parallel universe” effect.
Why it matters: This research addresses pressing concerns around misinformation and its impact on democratic processes — particularly the ability of citizens to make informed electoral decisions based on their media diet.
By systematically evaluating how well voters can differentiate between true and fake news, the study provides empirical evidence that challenges the prevailing notion of a widespread “death of truth.” This insight is vital for policymakers, educators, and media organizations, as it shifts the focus from combating misinformation alone to also enhancing access to reliable information across diverse socioeconomic groups.
“The key message is that there is brutal information inequality in US society: Some people are informed; others are not. That doesn’t really correspond to the country’s ideological divide,” says Prat. “As a society, we’re devoting enormous resources now to fix misinformation and fight fake news. We should also devote resources to make sure that everybody gets access to real news.”
Read more: Can We Curb Fake News? Smart Research May Provide the Answer