Research > 11 > 98

Homeless Families in Washington State

A Study of Families Helped by Shelters and Their Use of Welfare and Social Services.

Executive Summary

This study was conducted by the Washington State Department of Social and Health Services (DSHS), to provide state policy makers and program executives with basic information about homeless families relying on shelters throughout the state.

This report draws together four kinds of data:

1. A census of all shelters, including private shelters, that serve families in Washington State. The one-night rolling census obtained the number of families each shelter had provided a place to stay the previous night, from late June through late September 2000.

2. Interviews with one parent from each of 411 families using those shelters.

3. Interviews with 70 shelter providers and 27 local welfare office administrators and their staffs.

4. DSHS administrative records of welfare eligibility and actual grant and Food Stamps issuances for most of the 411 interviewed families. We similarly obtained DSHS medical insurance and social service data where these existed for those respondents and their minor children. We received permission from the parents we interviewed to look at the DSHS records for 81 percent of these parents plus the children living with them. We found one or more DSHS records for 98 percent of these persons.

Numbers of families relying on shelters
Statewide, an estimated 750 families were relying on shelters during one night in mid-2000. The 2,529 children and adults that made up these 750 families represent 4.3 out of every 10,000 children and adults in the state's population, proportionally more in the eastern part of the state (5.3 per 10,000 population) than in the west (4.0 per 10,000). These 2,529 children and adults also represent just under two out of every 100 children and adults receiving Temporary Assistance for Needy Families (TANF) in July of 2000.

Demographic characteristics
Two-thirds of the families were headed by one adult, usually a woman, and one-third were two-parent families. Two-thirds of the families had either one or two children; ten percent included a pregnant woman. Thirty-seven percent of the children were under five years of age, 42 percent were five to eleven years, and 21 percent were adolescents, aged 12-17. Twenty-eight percent of the respondents had at least one child not living with them at the shelter.

Respondents represented diverse racial and ethnic groups: 51 percent were non-Hispanic white, 19 percent non-Hispanic black, 12 percent Hispanic, eight percent non-Hispanic American Indian/Alaska Native, and three percent non-Hispanic Asian or Pacific Islander. More than one race was reported by nine percent.

Compared to heads of TANF families, homeless parents were somewhat less likely to be non-Hispanic white (51 versus 62 percent) and more likely to be black (19 versus 12 percent) or American Indian (8 versus 5 percent). In contrast, the state's population was more likely than either homeless or TANF parents to be white (79 percent) and much less likely to be black (3 percent), American Indian (1 percent), or Hispanic (7 percent).

When only homeless parents who were receiving TANF were compared to TANF parents in general, they were found to be similar on several characteristics. In both groups: about ninety percent were female, slightly over eighty percent were one-parent households, just under 20 percent were currently married and living together, and about 45 percent had a high school diploma or GED as their highest level of education.

Homeless parents on TANF differed from TANF parents, however, in their age and length of time on TANF. Homeless parents were slightly older than TANF parents in general. Also, homeless families had been receiving TANF for less time than TANF families in general since homeless families were more likely to have received TANF for a short time span (36 versus 23 percent under six months) but less likely to have received TANF in the longest time span (5 versus 15 percent between three and four years).

Living arrangements during this homeless period and before
On average, the families already had been continuously homeless for nearly four months: 39 days at the shelter and 77 days before the shelter. At the time of the interview, their homelessness had not yet ended.

While homeless, the families moved often: 80 percent had lived in two or more other homeless places before the shelter and 54 percent had lived in three or more prior homeless places.

Temporary shared living was the most common homeless living arrangement. Thirty-nine percent of the families came to the shelter from a shared-living arrangement. The families' second most frequent living arrangement was other shelters. Twenty-two percent of the families came to the shelter from another shelter. For every 100 admissions to their present shelters the families had had 68 previous shelter admissions during the past year.

More than half the respondents had been homeless previously. For 42 percent, this was their first homeless experience. During the last twelve months, 26 percent of the families had been homeless, then housed, then homeless again. Forty-four percent had been homeless before the last year.

Sources of money and access to welfare benefits
Before the families were homeless, in any given month, between 30 and 35 percent were getting cash assistance from welfare programs, mostly TANF, and between 40 and 45 percent were getting food stamps. Access to welfare benefits increased sharply soon after the families became homeless. Within three months after becoming homeless, between 60 and 65 percent were receiving cash assistance in any given month and between 70 and 75 percent were getting food stamps. Since becoming homeless, 44 percent of respondents had gotten some money from paid work and 22 percent had received financial help from their families.

Supplemental emergency housing grants, called Additional Requirements for Emergent Needs, or AREN, were received by 20 percent of the respondents in the 12 months before becoming homeless. Between the onset of their most recent homeless period and the end of calendar year 2000, 32 percent received an AREN grant.

Nearly all (97 percent) of the respondents had been to a Community Services Office (CSO), often called the "welfare office," at some time in their lives, but only 73 percent said they had been to a CSO since becoming homeless. Most (85 percent) of the administrators and lead staff workers we interviewed at these offices reported that homeless families are given priority or expedited service when applying for welfare benefits. Lack of necessary documents is a common problem for homeless families when applying for welfare benefits, cited by over 40 percent of the welfare office administrators and shelter providers we interviewed.

Work and participation in WorkFirst
In the week before the interview, 21 percent of respondents and their spouses or partners had worked 20 hours or more. Thirty percent said they could not work due to illness, disability, treatment, or counseling.

Fifty-eight percent of the respondents participated in WorkFirst in the month before the interview, based on DSHS records and the state-defined participation rate which includes working 20 or more hours in the prior week, being employed in a work study position, looking for work, preparing for work, or being under a short-term (three months or less) sanction. This rate is lower than the 93 percent found for TANF recipients in general in August 2000, mostly due to fewer homeless families working or preparing to work compared to TANF recipients in general. Twenty-five percent of the homeless respondents who were receiving TANF in the month of our interview were exempt or deferred from work-related WorkFirst activity, roughly the same percentage (28 percent) as TANF families in general. Homeless families, however, were much more likely to be deferred while they resolved issues related to homelessness (12 versus 2 percent). Eighty-one percent of the CSO administrators we interviewed said their CSOs deferred homeless families from WorkFirst work preparation requirements for limited time periods to give the families time to find a place to live.

Drug and alcohol use
Rates of drug and alcohol use by the homeless respondents in this study were compared to rates for women aged 18 to 54 living below 200 percent of the Federal Poverty Level based on a conveniently available statewide household survey in the mid-1990s. The homeless respondents and women in poverty had the same rates of drinking in their lifetime (93 percent) and binge drinking in the last 18 months (23 percent). For overall rates of drinking in more recent periods, however, the homeless respondents reported lower rates of drinking than women in poverty. Sixty-eight percent of homeless respondents reported drinking in the last 18 months compared to 74 percent of women in poverty. In the last 30 days, only 20 percent of the homeless respondents reported drinking, compared to 60 percent of women in poverty. These lower recent alcohol use rates among homeless respondents could reflect a change in drinking patterns while living at shelters where the use of alcohol is normally prohibited, under-reporting, or successful efforts to reduce their use of alcohol.

Comparisons between self-reported drug use of homeless respondents and that of women in poverty produced mixed results. Lifetime rates of drug use were about the same for many drugs: hallucinogens (25 versus 24 percent), stimulants (33 versus 30 percent), and opiates other than heroin (9 versus 8 percent) but were higher among homeless respondents for other drugs: marijuana (72 versus 53 percent) and cocaine (38 versus 21 percent). Differences, however, were not tested for statistical significance and could be due simply to chance or measurement.

Drug use in recent periods (past 18 months and last 30 days) was determined for two general categories: marijuana and any illicit drug other than marijuana. Differences between homeless respondents and women in poverty were small and could have been due to chance. In the last 18 months, marijuana use was reported by 10 percent of homeless respondents and 15 percent of poor women while use of other illicit drugs was reported by 13 percent of homeless respondents and 10 percent of poor women. Rates of marijuana use in the last 30 days dropped to five percent for the homeless and nine percent for women in poverty, and past-month use of other illicit drugs was only three and five percents, respectively.

Recent indicators of need for chemical dependency treatment appeared to be quite similar for homeless respondents and women in poverty. Of the homeless respondents, 17 percent met screening criteria for substance abuse or dependence in the last year, whereas, of the women in poverty interviewed in the mid-1990s, 14 percent had an alcohol or drug use disorder in the last 18 months. Homeless respondents were more likely than women in poverty, however, to have received treatment, counseling, or assistance from self-help groups (e.g., Alcoholics Anonymous) for drug or alcohol use at some time in their lives: 29 percent versus 11 percent. According to records from the DSHS Division of Alcohol and Substance Abuse, 21 percent of the homeless respondents had received publicly funded alcohol or drug abuse treatment (inpatient, outpatient, or methadone) within a recent 2½ year period (July 1998-December 2000).

Mental health
One-third of the respondents had indications of major depressi/ms/rdaon in the last year, and one-third had panic disorder in the same period. In contrast, a 1994 survey of households in Washington State resulted in much lower estimated rates for each of these mental health problems among women in poverty aged 18 to 54: 12 percent for major depressi/ms/rdaon and six percent for panic disorder. Almost half (45 percent) of the homeless respondents reported getting mental health treatment at some point in their lives, and a quarter had received some form of publicly funded treatment in a recent 2½ year period according to DSHS records from the Mental Health Division.

Domestic violence
Ninety percent of the respondents answered questions about domestic violence by intimate partners in the last year. Emotional abuse was reported by 44 percent of them, physical abuse by 27 percent, and sexual abuse by 10 percent. Of those who had experienced some form of domestic abuse, one quarter had gone to a medical care provider to seek care as a result of the violence, half had law enforcement involvement in their domestic situation, and one third had received a court-ordered protective order.

Family services
Over a third of the families had received services from DSHS Children's Administration during a recent 2 ½ year period including Children's Protective Services' (CPS) case management, risk assessments, and counseling as well as other family reunification services and support for basic needs. During this same period ten percent of the homeless respondents were involved in Children's Administration cases in which at least one child was removed from their home.

Twenty-eight percent of the respondents had children who were not living with them at the shelter at the time of our interview. Three quarters of these children were living with another family member, 11 percent were in foster care, and five percent were living with their adoptive parents.

How shelters operate
Within Washington State we identified 152 programs that provided shelter to homeless families, and we interviewed staff at 70 of these shelters. Of the 152 shelters, 130 received state funding: 86 emergency shelter funds only, 36 both emergency and domestic violence funds, and eight domestic violence funds only. Twenty-two shelters got no state funds to operate their programs. At the shelters where we interviewed lead staff, the types of living accommodations they provided families included rooms or apartments in one building (77 percent), vouchers for motel or hotel rooms (46 percent), and scattered rooms or apartments (13 percent).

Of the 70 providers interviewed, 60 said they had a rule stipulating the maximum length of stay, ranging from two days to two years, with 23 percent using a 90-day maximum, 20 percent using 60 days, and the rest providing some other time limit. Limits were somewhat flexible, however, with providers who had maximum stay rules estimating that about 26 percent of their families were likely to stay an extra week or so.

In addition to providing shelter, these programs provide many other services to families as well. These include help in finding housing (96 percent of the providers), clothing (93 percent), case management (90 percent), food or meals (86 percent), and help in getting welfare benefits (84 percent).

Almost half (47 percent) of the families we interviewed got into their current shelter without any delay, and another third got in within a week. Twenty percent waited longer than a week. Of those who had to wait at least one day, some were given motel vouchers by the shelter or another temporary place to stay, but most had to remain wherever they were living, usually a place shared with family or friends

Future directions
This study provides a comprehensive description of the characteristics of families who are being helped by shelters and the types of social services they have received. This information may contribute to future policy questions by showing both what we do and do not understand about homeless families. The study also provides information about the issues that local welfare offices and shelter providers face in trying to meet the needs of homeless families. The two state agencies most responsible for serving these families through public assistance as well as shelter and housing programs are DSHS and the Office of Community Development (OCD). Through the Homeless Families Plan developed by these entities, greater cooperation has been achieved at the state and local levels. Despite the efforts and successes of the past, more work remains to be done to bring state, local, and private resources to bear on the problems faced by homeless families within our state.

Modified: Thursday August 16 2007

For more ways to get in touch with the Department of Social and Health Services, go to the DSHS Contact Information Web page. Technical Site Comments: DSHS Webmaster. Copyright 2004 Washington State Department of Social and Health Services.