Measuring Subjective Human Sentiment in Public Spaces Before and During the COVID-19 Pandemic: Using Twitter Data in Manhattan, New York City

In the mist of the COVID-19 pandemic, public spaces have taken on a more prominent role in human society. Using natural language processing techniques combined with a Difference-in-Differences approach, this study explores how subjective human sentiments expressed in tweets within either publicly owned public spaces (Pub-OPS) or privately owned public spaces (POPS) vary before and after the onset of the COVID-19 pandemic, as compared with those generated outside of the public spaces in Manhattan, New York City. Findings of this study reveal that while the COVID-19 outbreak has negatively affected people’s expressed sentiments, there have been more positive changes in subjective human sentiments within both Pub-OPS and POPS than the outside of public spaces during the pandemic, with a higher magnitude in Pub-OPS. The study provides guidance to policymakers and urban planners in the design and management processes for creating healthy urban built environments, particularly amid such unprecedented times, based on the potential for public spaces to promote people’s subjective mental well-being and positive emotional experiences.