Senior Honors Thesis: Greenspace and Health

By Lizzie Tong

In the decades to come, cities must grapple with a myriad of challenges – climate change, increasing population density, rising inequality – and develop mitigation strategies through smart urban design. Cities around the world, including Singapore, Vienna, and Shenzhen, have turned to greenspace as a way to address these concerns and improve overall quality of life for its citizens. 

Greenspace can mitigate challenges like excessive urban heat. The urban heat island effect is a phenomenon in which the heat trapped in densely built environments raises summer city temperatures. This exacerbates air pollution and heat-related illnesses, which can disproportionately affect minorities and other vulnerable populations. Increased greenery lessens urban heat by providing shade to city residents, reducing dark, heat-absorbing surfaces, and contributing to overall lower city temperatures. Additionally, more tree canopy better filters particulates in the air, improving air quality and reducing the risk of respiratory-related illnesses. While it is clear that the benefits of greenspace improve the natural environment for city dwellers, greenspace is also linked to improving an individual’s physical and mental health by affecting one’s mood and behavior.

“Faces in the Fountain” Photo Credit: Creative Commons

Recent studies find a slight positive correlation between greenspace and overall health; however, the research fails to shed light on whether greenspace causes improved health. And if it does, what aspect of health – physical or mental – is it improving? My honors thesis research applies an econometric, fixed-effects analysis of cross-sectional data to determine the effect of greenspace on physical activity rates and social cohesion in Baltimore, MD. Exercise is an indicator of physical health while relationships, like social capital, social networks, and social cohesion, are indicators for mental health and well-being. The paper not only establishes a causal link between greenspace, physical activity rates, and social cohesion, but also seeks to address the inverse relationship between heat and income by analyzing how greenspace can be introduced to minimize health disparities and improve living conditions for Baltimore’s most vulnerable populations.

“Row of Houses in Baltimore” Photo Credit: Creative Commons

Why is a fixed-effects analysis important?

Traditional Ordinary Least Squared (OLS) regressions are not able to determine a causal link between greenspace and health because it fails to eliminate residential self-selection, non-random greenspace, and reverse causality, which are various types of estimation bias. These three estimation biases present endogeneity — internal factors that can be biasing the effect of greenspace on health. For example, residential self-selection assumes that individuals who prefer to exercise could intentionally move to areas of increased greenspace that facilitate exercise, causing a bias that overestimates the effect of greenspace on physical activity. Similarly, intentionally targeting increases in greenspace in certain neighborhoods (non-random greenspace) overestimates the benefits to health, further introducing bias that traditional OLS cannot capture. The Baltimore Sustainability Plan specifically states that the City of Baltimore intentionally targets greenspace development in low-income, African-African, and minority residential neighborhoods, which proves that non-random greenspace is an endogenous factor that makes it difficult to determine the causal link between greenspace and health. Finally, strong social cohesion levels could also encourage communities to band together to improve their own neighborhood’s greenspace. Thus, reverse causality, or the effect of social cohesion on greenspace is another example of bias that needs to be accounted for. Applying a fixed-effects analysis at the zip code and census tract level accounts for this type of endogeneity. Fixed-effects is an estimation technique that controls time-invariant, unobserved characteristics at a given geographic area, like community culture, an individuals’ perception of safety, retail environment, or features of the natural environment. Holding these unobserved variables constant reveals the causal link between the effect of greenspace on physical activity and social cohesion, removing bias caused by residential selection, non-random greenspace, and reverse causality.

What is the impact of greenspace on health?

The results find a small, positive association between greenspace and physical activity rates for individuals in Baltimore. This confirms the findings of previous research; however, the results of my research do not indicate a statistical significance. There are more promising results for social cohesion; there is a positive, statistically significant relationship between greenspace and social cohesion rates. Social cohesion is a composite variable measured by individual responses to the following characteristics on a scale from 1 to 5, from strongly disagree to strongly agree: whether the neighborhood is close knit, trusting, willing to help, and has strong problem solving abilities. Based on the individuals’ composite responses, the results find that a 10% increase in greenspace in a given zip code can cause a 1 point increase in the social cohesion index, suggesting that a moderate increase in greenspace can translate to a tangible change in neighborhood social cohesion.

Who benefits from increases in greenspace?

Not only do the results suggest that greenspace improves social cohesion rates broadly, but the results indicate significantly more beneficial effects for low-income populations. Individuals with lower income, lower levels of educational attainment, that do not identify as White exhibited greater increases in social cohesion rates than other demographics. The results confirm a hypothesis by Gordon-Larsen (2016) that suggests that an increase in the accessibility and availability of public amenities, like recreational areas, can improve health outcomes for lower-income, minority populations. In my research, a similar trend finds that increasing greenspace could be even more beneficial for low-income, minority demographics than for wealthier, Whiter demographics.

City Livability

Given these findings, cities can be more confident in their decisions to implement greenspace as a means to improve its residents’ health. Not only is there a direct causal link, but cities can evaluate a quantitative impact on their populations – a 10% increase in greenspace increases public perception of social cohesion by one index value. Further, the fixed-effects analysis eliminates any concerns of overestimating the effect of greenspace on health. Once all endogeneity is removed, the results still indicate a large, statistically significant increase in social cohesion rates. Thus, greenspace, which is already a viable solution in megacities internationally, is undoubtedly a key pillar of urban design that not only improves the natural environment, but also improves the final frontier of city liveability: human physical and mental well-being. 

Featured Image: “Luiseninsel Großer Tiergarten Berlin” Photo Credit: Tobias Nordhausen, Flickr Creative Commons

About the Author: Lizzie Tong is a senior studying economics and computer science at UNC, with an interest in applying data science to solve challenges tied to urban sustainability. After graduating, she will be working as a research assistant for the Community Development and Policy Studies Team at the Federal Reserve Bank of Chicago. In her free time, she enjoys trying her hand at oil painting, running, and new Bon Appetit recipes.