Social Support of Homeless and Housed Mothers: A Comparison of Temporary and Permanent Housing Arrangements* Bethany L. Letiecq,** Elaine A. Anderson, and Sally A. Koblinsky This study compared the social support of 115 low-income housed mothers and 92 homeless mothers residing in emergency shelters (n = 31), transitional housing units (n = 44), and doubled-up arrangements (n = 17). All mothers had a preschool child in the Head Start program and the majority of the sample was African American. Results revealed that homeless mothers in emergencyshelters and transitional-housing had significantly less contact with friends and relatives, could count on fewer people in times of need, and received less help from their families over a six-month period than housed mothers. Findings suggest that family service workers should devise strategies to help homeless families access and mobilize existing familial supports and develop new social support networks. Additional implications for policy development and practice are discussed. Key Words: Head Start, homeless families, social support. Homeless families, generally headed by single mothers, are the fastest growing population among the homeless. Although accurate statistics are difficult to obtain, one study found that homeless families accounted for 43% of the sheltered homeless population in 1993 (U.S. Conference of Mayors, 1993). The Institute of Medicine (1988) further estimated that as many as one hundred thousand children are homeless every night. The recognition of family homelessness as a serious and growing social problem has resulted in several studies documenting the root causes of family homelessness (e.g., McChesney, 1992; Rossi, 1989; Wright, 1989) and the characteristics of homeless mothers and children (e.g., Bassuk & Rosenberg,1988; Shinn & Weitzman, 1996). Although previous studies have found that individual factors such as mental illness and substance abuse enhance families' likelihood of experiencing homelessness (e.g., Buckner, Bassuk, & Zima, 1993; Robertson, 1991), less is known about the relationship between homelessness and family variables such as social support. In general, social support is believed to enhance an individual's subjective well-being, buffer the negative effects of stress, facilitate positive coping, and strengthen family functioning (McLoyd, 1990; Rook, 1984). When social support is lacking, families who are vulnerable to homelessness may be left without a critical safety net to catch them before they lose their permanent housing (Bassuk & Rosenberg, 1988; McChesney, 1992; Shinn, Knickman, & Weitzman, 1991). The experience of homelessness may also strain existing social support because the sheltered homeless may live at great distances from their home communities and families, and may be more demanding of their friends' and families' resources. Finally, limited social support may hinder the sheltered homeless family's efforts to return to permanent housing. To provide a more precise framework for investigating social support, Barrera (1986) identified three concepts. The first, social embeddedness, refers to an individual's connections with significant in a social network, including the size, proximity, and/or amount of contact with network members. A second concept, perceived social support, is defined as the cognitive appraisal of being reliably connected to others (p. 416), and incorporates the perceived availability and adequacy of support. The last concept, enacted support, addresses the actual assistance provided by members of one's social environment. Over the past decade, several studies have attempted to investigate the relationship between social support and family homelessness. In one study of Boston homeless families, Bassuk and Rosenberg (1988) found that homeless families had significantly smaller social networks than housed families, as well as fewer extended family members living nearby. …