Back to Data Equity Hub

National Alliance to End Homelessness: Demographics of Homelessness

Examining the role of race in predicting one's chance of experiencing homelessness across the United States.



The data

The data used for this visualization include homeless population counts from the U.S. Department of Housing & Urban Development’s AHAR: 1) Estimates of Homelessness in the U.S. and 2) general population counts from the U.S. Census Bureau’s American Consumer Survey.  These two data sets allow us to calculate per capita rates of homelessness.

The visualization

The design needed to convey demographic information of multiple groups experiencing homelessness within all 50 States at a glance.  Within the tile map above, the box for each state includes a bar chart which allows users to compare a demographic group’s rate of homelessness to the state’s overall rate and the national rate.  

The historical context

In America, all levels of government have historically worked to segregate and limit the places where people of color can live. Decades of exclusionary zoning and disinvestment still affect the housing outcomes. Furthermore, people of color disproportionately lack sufficient access to mental and physical health services, unemployment insurance, living wages, and other needs. The results of these overlapping failures help explain the disparities in our homeless population data.

The current implications

Beyond helping jurisdictions identify if they have a problem, this viz can help them locate potential solutions.  Stakeholders can see which jurisdictions have better numbers and, therefore, potential best practices worth emulating.  


Data Deep-Dive

Key takeaways to guide analysis

Who is most affected

Only two groups (Black people and men) have above average rates of homelessness in all 50 states

A need to focus on Black men

This reality suggests an imperative to center Black men in national and local solutions and policies

Understanding the LatinX experience

LatinX rates of homelessness vary significantly by region—pointing to a need for stronger data and understanding what's driving the differences