The Future of Children, Princeton - Brookings: Providing research and analysis to promote effective policies and programs for children.

Since the early 1980s, estimates of the size of the overall homeless population have ranged from 192,000 to three million.⁶ An important reason for this variance in the estimates is different definitions of homelessness.

Two broad groups of individuals can be included in a definition of the homeless. The first is often called the literally homeless, people who have no permanent homes and spend the night in places such as the street, cars, or emergency shelters.⁷ The other group sometimes considered homeless is the precariously housed population. People who are precariously housed are in danger of becoming literally homeless because they have no place of their own to live or their current housing situation is tenuous. This group includes, among others, people who are doubled up—those who are living for short periods of time with friends or relatives and thus lack a fixed, regular nighttime residence. Children often appear among the precariously housed population because parents who become homeless may place their children with friends or relatives in order to avoid literal homelessness for them. Because some individuals and families choose to share housing as a regular, stable, and long-term arrangement, distinguishing the precariously housed from those in stable sharing arrangements is difficult.

The official definition of homelessness used by government agencies for determining eligibility for services is found in the 1987 Stewart B. McKinney Homeless Assistance Act, the first comprehensive federal law dealing with emergency assistance for the homeless. The act defines a homeless person as:

(1) an individual who lacks a fixed, regular, and adequate nighttime residence; [or]

(2) an individual who has a primary nighttime residence that is—

• a supervised or publicly operated shelter designed to provide temporary living accommodations (including welfare hotels, congregate shelters, and transitional housing for the mentally ill);



• an institution that provides a temporary residence for individuals intended to be institutionalized; or



• a public or private place not designed for, or ordinarily used as, a regular sleeping accommodation for human beings.⁸

The McKinney definition includes many of the literally homeless population, although it excludes individuals in jails. It may or may not include the precariously housed, depending on the interpretation of "fixed, regular, and adequate nighttime residence." This ambiguity makes the McKinney definition difficult to use for data collection.

The existence of an official definition does not mean that the government has assumed responsibility for counting the homeless. In fact, other than making administrative counts of the homeless in some jurisdictions, no agency is charged with keeping track of the size of the homeless population. The McKinney programs are largely designed to address emergency housing needs and have no mandate to monitor the size of the homeless population. Therefore, many groups, each with its own political and social agenda, have attempted to fill this measurement gap.

Every attempt to measure the homeless population requires practical decisions about specific subgroups to include or not include in the estimates. People living in single-room occupancy (SRO) hotels, minors in shelters for runaways, people in nursing homes, and women in battered-women's shelters might all be defined as homeless under the McKinney definition, but they are often excluded from counts of the homeless population.⁹ As a practical matter, many studies use participation in programs and services as the basis of their estimates, but this approach runs the risk of not counting the homeless who do not contact those programs and services. When counting homeless families and children, even more definitional issues are involved. Depending on the definition, an 18-year-old can be counted as a homeless adult or a homeless child, and a mother and father in a shelter without their children present may or may not be considered a family.

Aside from differences in definitions, another source of variability in published estimates of the size of the homeless population is the diverse statistics used to describe the population. The two most frequently reported statistics are nightly counts (or point prevalence) and annual prevalence. Nightly counts measure the number of people who are homeless at one point in time. This type of estimate presents a "snapshot" of the size and composition of the homeless population. The point in time chosen for the nightly count can affect an estimate because there are typically more homeless people at the end of the month (when government assistance money runs out) and in the summer months.¹⁰ A nightly count can, however, provide a useful estimate of the number of homeless people who may need services on a given day.

Annual measures provide estimates of the number of people who are ever homeless over the course of a year. Because homelessness tends to happen intermittently for many people, the annual counts result in much higher estimates of the number of homeless people than the nightly counts. The duration of homelessness can have a large impact on estimates of annual prevalence. Efforts to count the homeless at different points in time may count the same people twice if people tend to be homeless longer than the time that elapsed between the counts.¹¹

Variations in the duration of homelessness can also affect the mix of services that families need. A family experiencing a short, one-time bout of homelessness is likely to be affected in other ways and to need different services than one that is homeless repeatedly and intermittently. Families that have been homeless continuously for an extended period may need still different services. However, these key data about homeless families, the duration of homelessness and number of homeless episodes, are difficult to estimate because of the lack of tracking systems for the homeless and the anonymous nature of many of the services that are provided.