When digital literacy meets negative bias: Understanding the adoption of online information and sentiment on X

Abstract

Digital platforms increasingly mediate how citizens engage with public policy, shaping not only access to information but also the formation of collective interpretations and sentiments. In the context of Indonesia’s Free Nutritious Meal (Makan Bergizi Gratis: MBG) program, this study examines the relationship between individual digital literacy and online information adoption, while simultaneously mapping public sentiment and discourse dynamics on X. Employing a concurrent mixed-methods design, the study combines an online survey of 385 social media users drawing on the DigComp 2.2 framework and the Information Adoption Model with large-scale social media analysis of 87,589 MBG-related mentions on X collected between April and August 2025 using Drone Emprit. Regression analysis revealed that digital literacy significantly predicts information adoption regarding the MBG program (β = 0.581, R² = 0.549, p < 0.001), indicating that digital literacy explains 54.9% of the variance in information adoption and substantially enhances individuals’ capacity to evaluate and accept policy-related information. While digital literacy supports information adoption at the individual level, the overall online discourse surrounding MBG remained predominantly negative, suggesting a disconnect between individual informational competence and collective sentiment formation. Sentiment analysis further revealed that public discourse was predominantly negative (56%), compared with positive (43%) and neutral (1%) expressions, with significant surges observed in May and July 2025. Network analysis additionally revealed that negative sentiment is concentrated around accounts with high centrality and influence, thereby amplifying critical narratives through networked interactions. These findings highlight a structural tension between individual informational competence and collective sentiment formation in digital communication environments. This research contributes to the existing body of work on digital communication and public policy by illustrating that initiatives aimed at improving digital literacy should be complemented by network-aware communication strategies, such as influencer engagement and real-time discourse monitoring, to bolster public trust and legitimacy in extensive policy interventions.