Modeling Homelessness: An Agent-Based Simulation of Social Dynamics and Housing Instability
Disciplines
Public Health Education and Promotion | Social Work
Abstract (300 words maximum)
Homelessness is a pressing social issue with implications for public health, social cohesion, and economic stability. Poverty and homelessness have a direct negative relationship with community health outcomes, making it critical to implement preventative measures that reduce homelessness. This project develops an agent-based model (ABM) to simulate homelessness in a community, incorporating economic factors as well as social network structures that influence housing stability.
The model is built using NetLogo, a multi-agent programmable modeling environment. In this framework, households are represented as agents ("turtles"), social connections as links, and the environment as patches. Each household is assigned a randomized income and rental cost, drawn from a power-law distribution to reflect real-world income disparities. Rent burden is calculated as a function of income, and households are categorized into groups based on their probability of becoming homeless. A Bernoulli trial determines whether a household transitions into homelessness over time, influenced by economic shocks and social ties.
Additionally, the model explores the role of systemic factors, including racial disparities and policy interventions, in homelessness trends. By incorporating social ties, the model examines how homelessness weakens community networks, making it more difficult for affected individuals to regain stability. The simulation can test policy interventions such as rent control, income subsidies, and emergency housing assistance, allowing researchers and policymakers to evaluate their potential impact on homelessness rates.
This model serves both short and long term research objectives. In the short term, it aims to challenge misconceptions about homelessness, identify data gaps, and highlight policy areas requiring intervention. In the long term, it can inform local, state, and federal policies, drive data-driven decision making, and enhance research on structural inequalities that contribute to housing instability. Through this work, we seek to advance understanding of homelessness as a systemic issue and provide effective policy changes.
Academic department under which the project should be listed
WCHHS - Health Promotion and Physical Education
Primary Investigator (PI) Name
Matthew Lyons
Additional Faculty
Monica Swahn, Health Promotion and Physical Education, mswahn@kennesaw.edu
Modeling Homelessness: An Agent-Based Simulation of Social Dynamics and Housing Instability
Homelessness is a pressing social issue with implications for public health, social cohesion, and economic stability. Poverty and homelessness have a direct negative relationship with community health outcomes, making it critical to implement preventative measures that reduce homelessness. This project develops an agent-based model (ABM) to simulate homelessness in a community, incorporating economic factors as well as social network structures that influence housing stability.
The model is built using NetLogo, a multi-agent programmable modeling environment. In this framework, households are represented as agents ("turtles"), social connections as links, and the environment as patches. Each household is assigned a randomized income and rental cost, drawn from a power-law distribution to reflect real-world income disparities. Rent burden is calculated as a function of income, and households are categorized into groups based on their probability of becoming homeless. A Bernoulli trial determines whether a household transitions into homelessness over time, influenced by economic shocks and social ties.
Additionally, the model explores the role of systemic factors, including racial disparities and policy interventions, in homelessness trends. By incorporating social ties, the model examines how homelessness weakens community networks, making it more difficult for affected individuals to regain stability. The simulation can test policy interventions such as rent control, income subsidies, and emergency housing assistance, allowing researchers and policymakers to evaluate their potential impact on homelessness rates.
This model serves both short and long term research objectives. In the short term, it aims to challenge misconceptions about homelessness, identify data gaps, and highlight policy areas requiring intervention. In the long term, it can inform local, state, and federal policies, drive data-driven decision making, and enhance research on structural inequalities that contribute to housing instability. Through this work, we seek to advance understanding of homelessness as a systemic issue and provide effective policy changes.