Skip to content

"News you don't believe": Audience perspectives on fake news

Authors: Rasmus Kleis Nielsen, Lucas Graves

Venue: Reuters Institute for the Study of Journalism, October 2017

TL;DR

Using focus groups and surveys across the US, UK, Spain, and Finland, Nielsen and Graves find that ordinary people see fake news as a spectrum rather than a binary distinction—encompassing poor journalism, political propaganda, misleading advertising, and fabricated content. Audiences experience the "fake news" problem as one of degree, shaped by low generalized trust in news media, politicians, and platforms alike.

Contributions

  • Empirical evidence that audiences define "fake news" much more broadly than academic definitions (which focus narrowly on fabricated reports)
  • Finding that the perceived fake news problem spans poor journalism, propaganda, some advertising, and false content—implying solutions targeting only fabricated content miss the larger audience concern
  • Documentation of generalized skepticism: people are distrustful of both platforms and publishers, and low trust in one is correlated with low trust in the other, not opposed
  • Evidence that specific trusted outlets exist for most people, but there is no consensus on which outlets are trustworthy

Method

Mixed-methods study combining: - 8 focus groups (6–8 participants each, ~2 hours, conducted Feb–Mar 2017) in New York, London, Madrid, and Helsinki, with stratification by age (20–34 and 35–54) - Survey data from online news users in the four countries, including questions on trust in news media and social media's role in distinguishing fact from fiction - All participants recruited via Kantar Media; focus groups used pre-task methodology to investigate behaviors and attitudes toward news

Results

How audiences define fake news. Focus group participants do not use a binary real/fake distinction; instead, they view fake news as existing on a spectrum: - Not seen as news: satire, parody (recognized as intentionally amusing/untrue) - Poor journalism: sensationalized, unreliable reporting (e.g., celebrity, health, sports coverage) - Propaganda: politicians lying or spinning; partisan news coverage that omits or distorts facts - Some advertising: clickbait pop-ups, sponsored content ("around the web" links), misleading ads - False news: for-profit fabrication and politically motivated misinformation campaigns

Participants see the boundary between fake news and legitimate news as a matter of degree, not kind.

Main content categories blamed for fake news. When asked for examples, people point to: 1. Poor journalism more frequently than fabricated stories 2. Political propaganda (domestic and international politicians; partisan outlets they distrust) 3. Misleading advertising and sponsored content 4. False content narrowly defined (the least frequently cited category)

Source attribution. People associate the fake news problem with some combination of news media, politicians, and platform companies—though they rarely directly blame platforms, the proliferation of false information is often tied to social media distribution.

Trust context. Survey data shows: - Less than half of online news users in the US and UK believe you can trust most news most of the time (38–43%) - Higher trust in Spain and Finland (51–62%), but still substantial skepticism - Only modestly higher trust in news people consume themselves vs. news in general - Generalized skepticism: low trust in news media and low trust in social media are correlated, not opposed; both are doubted

Awareness of the "fake news" term. Most participants in the US and UK are aware of the fake news discussion, but many see it as a politicized buzzword—especially as used by politicians (Trump cited frequently) to attack journalism.

Connections

Notes

Strengths. This is perhaps the most widely-cited audience study of the "fake news" era. The key insight—that audiences see fake news as a spectrum of credibility problems rather than a discrete category—has proven enduring and frequently invoked in policy and platform-design contexts. The multi-country design (US, UK, Spain, Finland) provides cross-cultural grounding. The distinction between narrow (fabricated stories) and broad (poor journalism + propaganda + misleading ads) framings of the problem has proven analytically powerful.

Limitations. Focus group methodology means findings do not represent population-level proportions; participants in the US and UK discussed fake news much more than those in Spain and Finland, suggesting attention is not uniform. The sample is drawn from those who already consume online news, not the general population. The 2017 timing, just post-2016 US election, may have inflated salience of the term.

Follow-up questions. How do audience definitions of fake news evolve in response to legislative and platform action taken in the years following this study (2017–present)? Do people's perceptions of propaganda, poor journalism, and false content shift as generative AI makes synthesis and distribution easier?