Skip over navigation

Journal Issue: Financing Child Care Volume 6 Number 2 Summer/Fall 1996

CHILD INDICATORS: Homeless Families and Children
Eugene M. Lewit Linda Schuurmann Baker


Public concern about homelessness in the United States has increased in recent years. A late 1995 Gallup poll found that 86% of Americans feel sympathy for the homeless, and 33% report that they feel more sympathy now than they did five years ago. According to the same poll, one reason for this apparent increase in sympathy is that 17% of Americans, primarily women and young adults, believe that they could become homeless.1 The fact that these groups are concerned about homelessness reflects, in part, two decades of increases in the visibility of homeless women and children in the United States. Published reports suggest that most homeless families with children are headed by single women between the ages of 26 and 30 who have never been married and have two children.2

Because shelter is a basic human need, it is not surprising that the effects of homelessness on children and families appear to be harsh and multifaceted. According to one study, homeless women are significantly more likely to have low birth weight babies than are similar poor women who are housed.3 Others report that, compared to the general population of children, homeless children have twice as many health problems, are more likely to go hungry, and have higher rates of developmental delay; and although findings have not been consistent, higher rates of depression, anxiety, and behavior problems have been reported for homeless children.4 Because, however, as discussed in this article, collecting reliable and comprehensive information about the population of homeless families with children is very difficult, accurately estimating the size, scope, and impact of homelessness among families with children in the United States has been almost impossible. 5 Estimates of the size and composition of the population of homeless families and children are important, however, to understand the etiology and consequences of homelessness, to design effective programs and policies to address the problem, and to evaluate whether interventions are working.

This Child Indicators article focuses on available data on homeless families and children. First, it reviews different definitions of homelessness and the most common methods used to estimate the size of the homeless population. It then examines data on subgroups of homeless children and youths in the United States and considers the duration of homelessness for families with children that use shelter services. Finally, it examines trends in the numbers of families who are at risk of losing their housing.

Estimates of the size of the homeless population vary, depending on the definition of homelessness used. Even when definitions are clear and consistent, the methods used to count the homeless differ widely. Estimates of the number of homeless at one point in time or for a period of time can be made. In practice, homeless families and children are a difficult group to find and track, and few estimates that focus on children specifically have been made. Using a variety of techniques, the U.S. General Accounting Office (GAO) estimated that between 80,000 and 400,000 children were likely to be homeless or doubled up, living with friends and extended family, on any given night in 1988. Based on the GAO's "best" estimates, many more children were doubled up (186,000 in 1988) than living in shelters or other community settings provided for homeless families (68,000 in 1988). The length of homelessness for families tends to be short (less than three months), although there is evidence that a small group of families is homeless for years. Data on trends in actual homelessness are not available, but trends in the number of single-parent families in extreme poverty, conditions that tend to precede homelessness for families, suggest a large increase in the population of potentially homeless families since 1975.