Sociology 498: Homelessness

Rossi: Down and Out in America

What Have We Learned About Homelessness?

Methods

• You can do surveys of the homeless
• But Rossi's area probability sample not the best way
• Instead: interview at homeless service providers (e.g., soup kitchens
• You need a comparison group
• You must have comparable questions
• Correlations do not necessarily mean causation
• Homelessness is both a cause and a consequence (e.g., depression)
• Homelessness and correlate may be joint consequences of some third cause

Types

• Fuzzy line: shifting back and forth in and out of homelessness
• Many short term homeless; fewer long term homeless, but they are the most of the problem

Causes

• Disabilities: homeless have higher rates than other poor of:
• incarceration
• drug use
• alcoholism
• mental illness
• But:
• Especially high rates of multiple disabilities
• Not higher rates of alcoholism & mental illness alone
• Policy implications: some homeless are very difficult to help
• Poverty: homeless are part of the extremely poor. But:
• little or no difference in work rates between homeless poor and domiciled poor
• Not always poorest groups who are most homeless:
• women
• young
• old
• Policy implications: government support, not jobs, often the best protection against homelessness
• Connections with family and society
• Extraordinarily high rates of non-marriage
• Suggestive evidence of low contact with family & friends
• Theoretical implications: Low social capital (Bowling Alone)
• These causes are not unrelated to each other but are combined in a causal path that better explains how people become homeless.