Digital literacy is associated with more discerning accuracy judgments but not sharing intentions

Key claims

• Digital literacy and accuracy discernment: Both measures of digital literacy (self-reported internet familiarity and understanding of social media algorithms) significantly predict the ability to rate true headlines as more accurate than false headlines. Effect sizes are comparable to analytic thinking and procedural news knowledge.
• Digital literacy and sharing discernment: null finding: Neither digital literacy measure consistently predicts sharing discernment (tendency to share true more than false headlines) or the quality of shared content. This holds across multiple outcome measures and subgroup analyses.
• The accuracy-sharing gap: Participants show fair discernment when asked to judge accuracy (true: 67.7% rated accurate, false: 40.9%) but this discernment largely disappears when deciding what to share (true: 54.1%, false: 48.3% — no significant difference).
• Robustness across contexts: Results hold for both political and COVID-19 headlines, regardless of participants' partisanship, and with consistent effect patterns across demographic groups.
• Procedural news knowledge predicts sharing quality: While digital literacy does not, procedural news knowledge (knowledge of journalism practices) is significantly associated with sharing higher-quality content, particularly among Republican participants.

Sources cited in the article

The article reviews and builds on prior work examining:

• The relationship between digital-literacy and fake news susceptibility (Jones-Jang et al. 2019; Guess & Munger 2020)
• Media literacy interventions and their effects (Guess et al. 2020; McGrew et al. 2019; Breakstone et al. 2021; Hameleers 2020) — see media-literacy for overview
• The accuracy-sharing gap: distinct patterns between accuracy judgments and sharing behavior (Epstein et al. 2021; Pennycook et al. 2020, 2021)
• Cognitive predictors of misinformation resistance: cognitive-reasoning and cognitive reflection

Notes

Well-designed survey experiment with large sample (N = 1,341) and careful experimental controls. The main contribution is demonstrating a striking disconnect: digital literacy helps people identify accurate information but does not translate into more careful sharing behavior. The authors test two distinct conceptualizations of digital literacy (technical skills vs. platform-specific knowledge) and find similar null effects for sharing, emphasizing the robustness of the finding.

The study is limited to stated sharing intentions in a hypothetical survey setting rather than actual sharing behavior, though the authors note that prior work suggests reasonable correlation between these measures. An important implication for practitioners is that identifying digitally literate users may not be an effective strategy for reducing misinformation spread; instead, interventions targeting attention to accuracy at the point of sharing may be more promising.

The distinction between accuracy and sharing outcomes is striking and reflects broader patterns documented in the literature: people's capacity to evaluate truth and their motivation to share quality content are partially decoupled, suggesting different psychological mechanisms underlie these behaviors.