Twitter sentiment in New York City parks as measure of well-being

• Published in: Landscape and urban planning Vol. 189; pp. 235 - 246
• Main Authors: Plunz, Richard A., Zhou, Yijia, Carrasco Vintimilla, Maria Isabel, Mckeown, Kathleen, Yu, Tao, Uguccioni, Laura, Sutto, Maria Paola
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.09.2019
• ISSN: 0169-2046; 1872-6062
• DOI: 10.1016/j.landurbplan.2019.04.024

•Twitter as an effective measure of people’s expression of sentiment.•Geo-located twitter data as well-being indicator associated with urban park space.•Urban park sentiment varies for spatial context and user characteristics.•Twitter data as a potential resource for urban design and planning. While there is an extensive literature regarding the benefits of natural environments within urban settings, there is relatively little statistical research on the correlation of well-being with urban green space. This research uses social media to develop a methodology for understanding the varying levels of feelings in urban green space. Using a geolocated Twitter database, this research correlates quantified sentiment levels inside parks in New York City. It addresses the following: are people more positive when they are in parks as compared to when they are in other places? Specifically, among Twitter users in New York City do people who visit parks have more positive Twitter-sentiment expression compared to their sentiment in other places? Our results show that sentiment expressed in tweets varies between areas inside and outside of parks. We find that in Manhattan in-park tweets express less positive sentiment as compared to tweets outside of parks, but park visitors in the other boroughs of New York City generate more positive in-park tweets as compared to those outside of parks. We discuss the use of tweets as an indicator of the public expressed sentiment and derive suggestions for further research.