About This Project

Trust in U.S. news media has dropped to 28%, while 45% of Americans now identify as independents. Research question: Does perceived algorithmic relevance explain the trust gap between independents and partisans? Hypothesis: Algorithmic feeds optimized for partisan engagement fail non-major-party audiences, eroding their institutional trust. A national survey of 1,000 adults and structural equation modeling will test this.

Ask the Scientists

What is the context of this research?

Trust in U.S. news media has fallen to 28%, a historic low, while 45% of Americans now identify as independents, up from 33% in 2005. News distribution has shifted from human editors to algorithmic systems optimized for engagement rather than journalistic norms. Algorithms categorize users by partisan signals, so individuals outside the major-party binary may receive content that feels irrelevant.

When algorithms fail to serve relevant content, audiences may attribute the failure to news organizations themselves. Prior work has examined algorithmic effects on trust in specific content, but no study has quantified whether perceived algorithmic relevance explains the trust gap between independents and partisans. I hypothesize that it does: independents report lower perceived relevance, which in turn predicts lower trust in their referent outlets.

What is the significance of this project?

This project addresses a critical gap by quantifying how algorithmic gatekeeping shapes institutional trust among an understudied and growing population.

Academically, it extends Gatekeeping Theory into algorithmic environments by testing a statistically modeled indirect association between political affiliation and media trust. No prior study has modeled whether perceived algorithmic relevance accounts for the trust gap between independents and partisans.

Practically, findings will inform news organizations seeking to maintain credibility with audiences whose identities lack clear partisan signals. Organizations must consider how their content is perceived in algorithmic feeds, not just how it is produced.

The 45% of Americans identifying as independent represent a substantial segment whose trust formation remains poorly understood. Understanding how algorithms shape their relationship with news institutions has implications for democratic engagement.

What are the goals of the project?

This study asks: Do political independents experience algorithmic news feeds differently than Democrats and Republicans, and does that difference erode their trust in news organizations?

When you scroll through social media or news apps, algorithms decide what content reaches you. These systems categorize users based on political signals. But nearly half of Americans do not fit neatly into partisan categories. If algorithms cannot categorize them effectively, these individuals may receive content that feels irrelevant.

The goals are to survey 1,000 U.S. adults and use structural equation modeling to determine whether independents and partisans report different levels of trust in their primary news outlet, whether independents perceive their feeds as less relevant, and whether that perceived irrelevance explains the trust gap. This may identify how algorithmic systems are reshaping nearly half the public's relationship to democratic institutions.