Neighborhood disadvantage, residents' distress, and online social communication: Harnessing Twitter data to examine neighborhood effects.

This study aims to examine the role of place-based online social communications in buffering the relationship between neighborhood disadvantage and psychological distress using geo-located Twitter feeds. We collected 1,981,802 geo-tagged tweets posted within the physical boundaries of 78 neighborhoods located in Pittsburgh for 16 consecutive months, beginning July 15, 2013. Psychological distress was measured using sentiments expressed in tweets, and within- and between-neighborhood online social communications were assessed using users' Twitter activity. This study reveals differential effects of within- and between-neighborhood online social communications. More online social communications within a neighborhood were associated with increased psychological sadness, anxiety, and distress in high-poverty neighborhoods; a relationship that was not found in low-poverty neighborhoods. This study suggests the buffering effects of online social communications may follow the same patterns as off-line social support in disadvantaged neighborhoods. Considering the interactions between online spaces and neighborhoods may be important for planning and implementing neighborhood-level interventions.