Experimental Data on AI vs. Human News Credibility and Selection

A 2026 study by Jun Luo presents data from a preregistered 2x2 between-subjects experiment with 1,012 participants. It examines how source attribution (AI vs. human journalist) and news topic (economy vs. immigration) affect perceived news credibility and future selection. The data is shared under a CC-BY-4.0 license on figshare.

Use Cases

• Analyzing the relationship between source attribution and perceived credibility based on experimental conditions.
• Investigating the moderating role of AI familiarity and personal topic involvement on credibility ratings.
• Modeling how credibility perceptions predict future AI news selection among different participant groups.
• Studying the normalization of AI authorship in news media using survey and experimental data.

Strengths

• Data originates from a preregistered experiment with a substantial sample size of 1,012 participants.
• Experimental design includes two manipulated factors (source and topic) and measured moderators (AI familiarity, involvement).
• The study directly addresses a timely topic in AI-powered journalism and transparency.

Limitations

• Row count and column-level documentation are absent; field semantics must be inferred after download.
• The dataset is very small (61.6 KB), indicating limited scope, likely containing summary results or survey responses rather than raw individual-level data.
• Description metadata is limited; actual data structure and granularity require manual inspection after download.

Provenance

Source

Jun Luo via figshare.

Collection Method

Preregistered 2 (source: AI vs. human) × 2 (topic: economy vs. immigration) between-subjects experiment.

Freshness

Last updated 2026-06-02 09:06:46; freshness should be verified.