The value of Twitter data for determining the emotional responses of people to urban green spaces: A case study and critical evaluation

Research output: Contribution to journalArticlepeer-review


Colleges, School and Institutes


Interactions between humans and nature are understood to be beneficial for human well-being. In cities, urban green spaces are believed to provide many benefits to urban populations in terms of mental and emotional well-being. Through a case study of 60 urban green spaces in Birmingham, United Kingdom, this article investigates the spatial and temporal variation of the emotions experienced by individuals whilst using urban green spaces. Using a dataset obtained from Twitter as the basis for emotional explorations, sentiment analysis was performed on over 10,000 tweets to ascertain the positivity/negativity of individuals. Positive responses were more common than negative responses across all seasons, with happiness and appreciation of beauty being the common positive emotions identified. For the negative responses, fear and anger were present in similar amounts, with fewer tweets indicating sadness and disgust. Our findings show that Twitter data is a viable source of information to researchers investigating human interaction and emotional response to space in cities. Such information has implications for urban planners and park managers, enabling the creation of evidence-based spaces which enhance positive outdoor experience. Limitations in using Twitter data are discussed and these should be considered in future research.

Bibliographic note

Article initially accepted under the working title: 'Investigating the Emotional Responses of Individuals to Urban Green Space Using Twitter Data: A Critical Comparison of Three Different Methods of Sentiment Analysis'


Original languageEnglish
JournalUrban Studies
Early online date8 Feb 2018
Publication statusE-pub ahead of print - 8 Feb 2018


  • built environment, method, public space, Twitter, urban green space

Sustainable Development Goals