What is Homelessness?

Literal homelessness refers to the status of people who are either sheltered homeless — using emergency or transitional housing — or unsheltered homeless — living on the streets or in parks, abandoned buildings, cars, subway tunnels or other places not meant for human habitation). Chronic homelessness occurs when people have long or repeated episodes of homelessness, such as living continuously in a shelter for one year or or on at least four separate occasions in the past three years.

Much of the data presented on this website were collected based on a formal definition of homelessness, which has been used for many years by the U.S. Department of Housing and Urban Development. The definition excluded people who are “doubled-up” with friends or relatives or who reside in transient or non-transient motels and hotels. Many of these people experience literal homelessness temporarily. An expanded definition of homelessness used by the U.S. Department of Education to track homelessness among school-age children includes sheltered and unsheltered children, but also those who are “doubled-up,” living in hotels or motels and those in unknown residences (McKinney-Vinto Act, Title X, Part C of the No Child Left Behind Act). This expanded definition has also been used by a variety of safety net programs, such as Head Start, and in legislative acts. 

In 2009, Congress passed the Homeless Emergency Assistance and Rapid Transition to Housing Act, which sought to align definitions for homelessness. The HEARTH Act defined the homeless as “an individual or family who lacks a fixed, regular, and adequate nighttime residence,” including not only the sheltered and unsheltered homeless, but also those who will “imminently lose their housing,” have no subsequent residence identified, and lack the resources or support network to obtain permanent housing. The HEARTH Act, which also included an expanded definition of homelessness for unaccompanied youth and families with children and youth, called for federal agencies to examine the feasibility of adopting a unified definition of homelessness. The HUD statistics presented on this website were collected before this legislation and new rates of homelessness may be calculated if and when new definitions for homelessness are incorporated in surveys.

Even broader definitions of homelessness exist. Although precarious housing, such as substandard living conditions and housing cost burdens, are distinct from literal homelessness, they put individuals and families at greater risk of becoming homeless and create unstable and potentially damaging living conditions. Some make the distinction between “houselessness” and homelessness to emphasize the importance of disconnected relationships, supports, neighborhood and community that enter into the broader meaning of a home.

The difficult task of measuring homelessness has been attempted for two decades, but a consistent set of serial data is not available. Methods for measuring the prevalence of homelessness are imprecise. The most rudimentary method has been the “point-in-time” count. Estimates are derived by surveying providers to identify the number of persons at an emergency shelter or transitional housing program on an identified night. Estimates of the unsheltered homeless are based on street counts that employ a variety of methods, most of which are of limited accuracy and consistency. Assessments are sometimes made where homeless people congregate, ranging from service centers to parks, encampments and steam grates. In other studies, enumerators canvass every street in their jurisdiction. Some communities conduct interviews at non-shelter service locations, such as soup kitchens.

A better method for estimating the number of homeless persons involves longitudinal administrative databases for clients served by homeless services, which can produce an accurate “unduplicated” count of homeless persons over time. In legislation passed in 1999 and again in 2001, Congress directed HUD to collect more definitive data on the prevalence of homelessness based on local Homeless Management Information Systems. However, data obtained from service providers miss homeless people who do not use such programs. The generalizability of HMIS data published in the Annual Homelessness Assessment Report is limited by the number of communities and shelter providers that participate and provide data.

For more information about these and other terms, visit the [[A-Z Glossary]].

Further reading about definitions of homelessness, the Annual Homelessness Assessment Report, and other past efforts to measure the prevalence of homelessness: