Misinformation in action: Fake news exposure is linked to lower trust in media, higher trust in government when your side is in power

Katherine Ognyanova; David Lazer; Ronald E. Robertson; Christo Wilson

doi:10.37016/mr-2020-024

In brief

Using survey data combined with browser history tracking from 3,000 U.S. respondents during the 2018 midterm elections, this study finds that exposure to fake news is associated with a 5% decrease in mainstream media trust overall but an increase in political trust. Critically, the effects vary sharply by ideology: strong liberals who consumed fake news reported lower trust in government, while moderates and conservatives reported higher trust — a pattern attributable to the predominantly right-leaning composition of fake news sources in their sample.

Key findings

Finding 1: General decline in media trust. Respondents exposed to fake news showed lower trust in mainstream media institutions across all ideological groups. The association implied a 5% drop in media trust during the one-month period around the election.

Finding 2: Paradoxical increase in political trust. Despite declining media trust, fake news exposure correlated with increased trust in political institutions overall, particularly Congress (4% increase) and more strikingly among those highly exposed (8% increase for trust in Congress). This contradicts simple hypotheses that misinformation uniformly erodes institutional confidence.

Finding 3: Ideological divergence. The relationship between fake news consumption and political trust is strongly moderated by ideology:

Strong liberals: Fake news consumption predicted lower political trust; they appear to recognize and reject the pro-Republican bias in these sources.
Moderates and conservatives: Fake news consumption predicted higher political trust; they appear to accept claims sympathetic to their political orientation.

This finding suggests that the impact of misinformation cannot be examined in isolation from the ideological alignment of the source and the viewer.

Method

Data collection: Survey of 3,000 U.S. respondents in two waves (late October and late November 2018, around the midterm elections)
Digital tracking: 227 respondents (8%) installed a browser extension logging their web browsing history between survey waves
Fake news exposure: Constructed from browser history using a curated list of fake news sources (Grinberg et al., 2019)
Trust measurement: Items adapted from the General Social Survey (GSS), measuring confidence in mainstream media (newspapers), the White House, Congress, the Supreme Court, the military, and the justice system
Statistical approach: Multiple regression models controlling for demographics (gender, race/ethnicity, age, education, family income), party affiliation, ideology, interest in politics, national news consumption, and number of visited news URLs
Interaction effects: Separate models examining how ideology moderates the relationship between fake news exposure and political trust

Key methodological considerations

The study relies on a subsample of 227 respondents for browser tracking, which is relatively small but detailed. The fake news list used was predominantly composed of right-leaning sources, which shapes the interpretation: fake news exposure in this context means exposure to right-leaning misinformation. The timing around the 2018 midterms (with Republicans controlling the White House and Senate) provides a natural context for understanding why pro-government messaging would increase trust among conservatives.

Notes

This work exemplifies the complexity of misinformation effects: exposure to fake news does not uniformly damage institutional trust but instead amplifies and polarizes pre-existing attitudes. The ideological interaction is theoretically important—liberals may dismiss right-wing misinformation as biased and actually update away from trusting a government they already distrust; conservatives may find confirmation in these sources and reinforce confidence in their political leaders. The study highlights the necessity of attending to the current political context (who is in power) and the ideological lean of the news sources in predicting how misinformation will reshape public opinion.