Empowerment in a Bureaucracy? Survivors’ Perceptions of Domestic Violence Shelter Policies and Practices

Abstract

Ten qualitative focus groups were conducted with 31 survivors of domestic violence living at an emergency shelter to study survivors’ perceptions of whether shelter policies and practices felt empowering or restrictive. Five themes emerged: (1) shelter monitoring and privacy; (2) transportation and mobility; (3) managing chores, cleaning, and shopping; (4) relationships with staff; and (5) feelings around leaving the shelter. Themes were shaped by larger economic and social forces, including (1) financial constraints and (2) the bureaucratization and professionalization of the shelter system that led to hierarchical relations between staff and residents. Recommendations for improving shelter experiences are addressed.

One in three women in the United States will be a victim of some form of physical violence by an intimate partner within her lifetime (Black, 2011). In the 1970s, women’s rights activists began to organize services and open spaces to shelter women who were fleeing abusive intimate partners (Davis, Hagen, & Early, 1994; Krishnan, Hilbert, McNeil, & Newman, 2004; Mantel, 2013). These grassroots organizations had limited budgets and were invested in second wave feminist ideals of empowerment, which included collective decision-making and a shared power structure. These ideals manifested themselves in service delivery that reflected a commitment to empowering survivors. As shelters faced financial challenges and sought additional outside funding sources, they were held accountable to government and private funders. This accountability led to the adoption of a professional service delivery model in which staff were hired with professional degrees and certifications, and many shelters became more bureaucratized to meet these funding requirements (Gengler, 2012). While bureaucratization had the potential in theory to enhance the services provided to survivors, many believe that in practice, it resulted in “a compromise of earlier feminist commitments” by “coming to replicate the controlling environments women are attempting to flee” (Gengler, 2012, p. 501), though other scholars point out that it is possible to create feminist bureaucracies that lead to the empowerment of all members of the organization (Martin, 2012).

Financial Constraints and Unmet Needs

Demand for domestic and intimate partner violence services outpaces the financial resources available. The 2015 National Census of Domestic Violence Services found that local and state hotlines answer a total of more than 21,000 calls a day, or about 14 calls every minute, on average (NNEDV, 2015). Additionally, 40,302 child and adult survivors utilized emergency shelters or transitional housing, and another 31,526 people received nonresidential assistance and services, including counseling, legal advocacy, and services for children (NNEDV, 2015). The study found 12,197 unmet requests for services in 1 day, of which 7,728 were for housing services, because programs did not have the financial resources to provide them. The report states that 1,235 staff positions were eliminated in the past year, most of which had been filled by direct service providers, due to funding reductions (NNEDV, 2015).

The current service model faces significant funding challenges which impact the experiences of survivors, and funding continues to be politically sensitive and contentious. For example, in the latest reauthorization of the Violence Against Women Act (P.L. 103-322), the first federal policy in the United States designed to combat intimate partner violence, spending of up to USD$660 million each year for 5 years was earmarked for domestic violence programs. However, this represented 17% less than when the act was reauthorized in 2005 (Helderman, 2013). This larger economic context has contributed to the shift toward the bureaucratization and professionalization of domestic violence shelters.

Feminism and Bureaucracy Within Domestic Violence Shelters

Research shows that domestic violence shelters with feminist ideologies have better outcomes for victims/survivors than shelters that do not have women’s empowerment as the central goal (Nichols, 2011). A decade or so after the first battered women’s shelters were founded in the United States, Rodriguez (1988) published an ethnographic account of a shelter in Hawaii that had “developed a counterbureaucratic structure to empower the women” (Rodriguez, 1988, p. 214). The shelter promoted self-help and self-reliance among survivors, and residents did the meal planning, cooking, cleaning, and childcare. There were several main features that stood out in the operation of the shelter including (1) eliminating professional hierarchy and hiring staff who were not formally trained human service professions; (2) consensual decision-making; (3) equal pay for all workers, which was identified as the most radical aspect of the structure, since it meant that full-time staff earned $10,600 per year as of mid-1987 (roughly equivalent to $22,000 per year in 2016); and (4) a libertarian structure, which meant few restrictions and rules (e.g., no curfew, residents sit on every committee to make sure their needs are taken into consideration). When conflicts arose, parties were required to negotiate solutions among themselves or with the help of mediators (Rodriguez, 1988, p. 224).

As sociologist Patricia Martin noted, “Bureaucracy and feminism are long-time adversaries” (2012, p. 283). While feminism “is a social philosophy and social movement aimed at securing equality (including physical safety, economic security, etc.) for women and girls” (Martin, 2012, p. 281), bureaucracy “refers to the structural and procedural dimensions of organizations that, as Max Weber specified, make it possible to coordinate the labor of many people to achieve ‘rationally’ identified goals” (Weber, 1947 in Martin 2012, p. 283), including a hierarchy of authority that specifies who has power over whom (Martin, 2012, p. 283). It is this bureaucratic feature that Martin says feminists are most uneasy about, since it “gives some members power over others” (Baker, 1982 in Martin 2012). Scholars have argued that this institutionalization and professionalization has led to a depoliticization of the battered women’s movement and to the potential for “abuse of power and control within the feminist movement against domestic violence” (Koyama, 2006, p. 209), which excludes many women from full participation, particularly women of color and poor women (Koyama, 2006) and immigrant and refugee women (Murdolo, 2014). Patriarchal policies include mandatory attendance in therapy, classes, and meetings as a condition of staying in the shelter. To be consistent with feminist gender-based advocacy that relies on survivors’ choice and agency, feminist scholars argue that these services should be available but optional (Nichols 2011). Ultimately, rather than blame individual direct service workers, Koyama (2006) argues that structural changes are needed to “make domestic violence activists and shelter workers more accountable to the actual needs and demands of survivors” (p. 210).

This Study

This article explores how feminist ideals of empowerment interact with bureaucratic realities at a domestic violence shelter that is referred to in this article as “Harbor.” It explores the question: How do shelter residents perceive shelter policies and practices and to what extent do they experience the shelter system as empowering or restrictive? By focusing on the voices and experiences of survivors, this research sees survivors as allies in the process of improving the system, while acknowledging that some survivors have less power and agency than others. This research is mindful of how gender, race, immigration status, disability, and the intersection of multiple identities shape experiences of survivors living in a domestic violence shelter. Intersectionality is the theoretical framework that sees multiple dimensions of identity as interlocking (Collins, 2002) and understands that a single-axis framework (e.g., only focusing on the experiences of “women” to the exclusion of how gender intersects with other identities) marginalizes those who are “multiply-burdened” (Crenshaw, 1989, p. 140). Intersectional work advances social justice by “addressing institutional-level power and privilege by challenging the system from within” (Kapur, Zajicek, & Gaber, 2015). The present study finds that intersections with gender and disability, and gender and language and immigration, are particularly salient in the context of this shelter.

Method

Procedure

Ten focus groups were conducted over the course of 12 months, from January 2016 to December 2016. Focus groups were held every 4–6 weeks (in January, February, March, May, June, July, August, October, November, and December). Group interviews are particularly well suited to exploring issues that are relevant to the person in context and to shifting the balance of power from the researcher to the research participants; therefore, scholars have noted that focus groups are a useful methodological tool for feminist research (Wilkinson, 1999). One of the strengths of this method is “the explicit use of the group interaction to produce data and insights that would be less accessible without the interaction found in a group” (Morgan, 1988, p. 12).

Participants were recruited for participation in the study from a shelter and transitional housing program that serves victims/survivors of domestic violence and their families in a mixed urban/suburban area in New England. There was an average of four participants per focus group, with a range of 2–10 participants. There were 31 unique participants but nearly one third were interviewed more than once (22 participated in one focus group, 7 participated in two focus groups, 2 participated in three focus groups). Not every participant was interviewed more than once, since length of stay at the shelter varied from a few days to a few months.

Focus group interview questions followed the trauma-informed care model (Substance Abuse and Mental Health Services Administration, 2014), which the shelter itself advocates, since residents of a domestic violence shelter constitute a vulnerable population (Seedat, Pienaar, Williams, & Stein, 2004). Consistent with trauma-informed care, the questions were related to safety; trustworthiness and transparency of staff; peer support; collaboration and mutuality; empowerment, voice, and choice; and cultural, historical, and gender issues. Before each interview began, the author emphasized to participants that she did not work for Harbor and was a survivor of domestic violence herself. The researcher also met with the entire staff at Harbor to get their feedback on the questions early in the project, and anonymous focus group data were shared with executive staff members at two additional in-person meetings throughout the year and with the entire staff at the conclusion of the research (strictly redacted to protect confidentiality), with the goal of identifying ways to improve services. Ultimate responsibility for the study methodology and data collection belongs to the researcher alone. The study was approved by the Eastern Connecticut State University Institutional Review Board (IRB Protocol #1548).

Anyone living at the shelter who spoke English or Spanish was eligible to participate in the focus group interviews, since the researcher is fluent in both languages. No one was turned away due to language restrictions. Four of the Latina interviewees did not speak English; therefore, 6 of the 10 interviews were done exclusively in English, and four interviews were done by the bilingual author translating English and Spanish throughout the interview for the benefit of all participants. Typically, focus groups would be conducted in one language; however, participants expressed a strong desire to be interviewed all together and have the interviewer translate. Despite language differences, residents wanted to have an inclusive conversation and to hear what others had to say. For many, this was the first opportunity to listen to and communicate with those who spoke a different language about their thoughts and experiences at the shelter.

Residents read and signed an informed consent form (in either English or Spanish) that included the purpose, procedures, risks and discomforts, their voluntary participation, and that they may discontinue their participation at any time and may decline to answer any question. One resident with low vision who could not read the informed consent had it read aloud to her and gave oral consent. All interviewees agreed to have the interviews audio recorded, and although there was no direct compensation for participation, food was provided at each focus group. Interviews were conducted in a space within the shelter where staff were not present. Interviews lasted 1–2.5 hr, except for the first interview, which ended after 45 min, when one resident who was not participating in the focus group had a medical emergency and an ambulance was called. All participants’ names were changed to protect confidentiality.

Data Analysis

All interviews were transcribed by the author or by an undergraduate student who completed National Institutes of Health (NIH) training in human subjects protocol. Analysis was done through analytic induction, a process whereby categories are developed as they emerge from the interview data (Glaser & Strauss, 1967). Preliminary data analysis and memo writing began as the researcher entered the field. Coding categories evolved over the course of data collection and were updated and expanded throughout the process (Knodel, 1993). Line-by-line coding was used for all 10 interviews, which involved a close reading of each transcript and allowed for new themes to emerge and be catalogued, a process that is consistent with a grounded theory approach to data analysis (Emerson, 1983; Glaser & Strauss, 1967).

Findings

At Harbor, official policies are in line with a feminist ideology of empowerment (e.g., no curfew and no mandatory chores, meetings, or therapies), but residents pointed out that direct service workers within the bureaucracy sometimes did not follow these policies, which will be further explored below. The two primary reasons this happened were (1) funding limitations that meant Harbor had fewer paid staff members and other resources that were needed to serve residents and (2) wanting to maintain a professional distance from residents as part of a larger trend toward bureaucratization and professionalization in the field. Following discussion of sample demographics, this article presents five main themes that emerged from focus group interviews with shelter residents: (1) monitoring by staff and lack of privacy; (2) challenges with transportation and mobility; (3) managing chores, cleaning, and shopping; (4) relationships with staff; and (5) rules around how long someone can stay in the shelter. Overall, although official policies are designed to be supportive and empowering, in practice, they can be experienced as restrictive and disempowering when programs face budget shortfalls or when staff members maintain a strict relational separation and hierarchy between themselves and residents.

Survivor Demographics

As stated above, the sample consisted of 31 residents of a domestic violence shelter and transitional living program. There were 30 women and 1 man who participated. Most participants had fled abusive opposite-sex intimate partners, though four younger women were in the shelter after leaving a household with an abusive father. The average age of participants was 31, with a range of 18- to 53-years old. No minors were interviewed. About 10% of participants were working in the paid labor force. Among those who reported an income (from paid work or other sources), only one person earned over $10,000 per year, and no one earned more than $20,000 per year. Close to 60% of participants were parents, though only a few were living with their children in the shelter or transitional living house; the rest of their children were either grown or in the care of someone else—typically family, friends, or in some cases, the Department of Children and Families (DCF). Ten percent reported a physical disability that significantly limited mobility. Two-thirds of those who reported their education had a high school degree or less, while one third had completed some college. No participants reported having a college degree or higher. There were 8 black participants (26%), 13 Latina (42%), 9 white/non-Hispanic (29%), and 1 multiracial participant (black, white, Native American; 3%). Thirteen percent spoke no English. Participants were not asked to disclose immigration status; however, a couple of residents shared that they were unauthorized.

Monitoring and Privacy

Harbor did not have a curfew. Residents appreciated this policy, since it allowed them to freely work and carry on with their lives as normally as possible. Nuria said,

That’s a positive, let me tell you. That’s a very positive—because I can go and see my daughter and stuff. And take her out to [the amusement park] and not worry. I can stay there until it closes and still be able to come back here. That’s a good thing.

However, other forms of surveillance and monitoring are in place at Harbor including (1) changing door codes to monitor who is staying overnight; (2) lack of privacy in the shelter, which can be triggering for some survivors; and (3) monitoring to see who would be a good candidate for transitional housing.

First, staff members frequently change the door code to maintain a high level of security when a resident moves out. However, staff also said that lock changes were done so that they can keep track of residents who are not coming back to the shelter at night. Staff did not share this reason with residents. Tanya, who had been locked out before because of these lock changes, remarked, “it’s kind of creepy at night there.” When asked what residents do when staff changes the lock code, Pilar said they go down to the office and “you pray that they’re there. If not, you’re standing there in the cold and waiting.” Tanya added, “In the darkness, waiting for something to happen!” A practice that ostensibly was done to keep track of residents and keep them safe may have the effect of making some feel less protected.

Second, residents mentioned the lack of privacy in the house, and how staff sometimes come into bedrooms unannounced, even if the door is locked. Brenda said, “They just come in our room one day and open the lock and walked in on me and my daughter. I wasn’t too happy.” Her adult daughter Emily said, “I know, like I could’ve been naked. I could’ve just gotten out of the shower!” Another resident, Flora, found the lack of privacy triggering, given her past trauma. Two months ago, a staff member entered her room at night. Flora explained in Spanish,

I was sleeping and I felt that—this woke me up. Because I’m a light sleeper. And when they came up on my side like this, they shined the flashlight in my face. Girl, forget about it! I about died in bed, from fright! And from there, my treatment changed. I had to up my [medicine] dosage.

Flora was not told why the staff member came in unannounced, and since the staff member did not speak Spanish, she was not able to explain.

Finally, some residents were aware that if they wanted to be considered for transitional housing services once their time at the shelter was up, they were being monitored more closely by staff. For example, Christine was hopeful that she and her two children would qualify for long-term transitional housing and felt that staff were watching to see if she would be a good candidate. “They’re watching how we talk to our kids; they are observing how we take care of ourselves, our children. They are observing all of those things.” This form of monitoring was high stakes, since it could determine whether a survivor had a safe, affordable place to go next.

Transportation and Mobility

Very few of the residents in this study had their own cars with them, and only a small percentage had a reliable income with which to purchase bus passes or frequent taxi rides. Residents walked or depended on Harbor’s van, driven by staff or volunteers, to get to appointments and stores.

At times, Harbor had the funds to offer additional transportation assistance, and this made a significant difference in the freedom and mobility of residents. For one 18-year-old pregnant woman who hoped to move across the country to live with her brother, Harbor said they could pay for her plane ticket. Another resident, Christine, was in a job training program and studying for her driver’s license. Harbor does not pay for driving lessons, but they will cover part of the licensing fee. “They only pay for half. It’s the funding and that’s the thing, they are funded by the government so like why aren’t we able to grab those resources instantly? Why aren’t those given, you know?” In another interview, Pilar said that there are three people in the house, including her, who would like driving lessons. Tanya agreed and said she has asked staff about this but they replied, “‘I’m sorry we don’t have resources on that. We’re closed on that.’ So all the doors are closed on us, so how are we gonna feel?”

Bus passes were scarce but valuable resources for residents. Before Yesenia left her abusive father’s home, she had been earning just over $13 an hour at a job she enjoyed. Since there were no shelter beds available near where she lived, and she did not have a car, she could not keep the job. Once she moved to the shelter, her advocate emphasized that “we still get rights,” including the right to transportation. However, in practice, due to staffing constraints, driving requests could not always be accommodated and bus passes were not always available. She is not from the area and says she gets lost walking sometimes, “But since I don’t have a job or money, I cannot take the bus or anything.”

For survivors who had physical disabilities or were pregnant, the lack of transportation was especially challenging. Zinnia missed the last bus back to the shelter one day.

I was at the mall and I was stuck and I’m 6 months pregnant and they told me they couldn’t get me. I had to walk here from the mall because I missed the last bus. But look, I didn’t have no money on me so I couldn’t call a cab. They didn’t give a [expletive]. I had to walk. And it was raining.

Another resident, Ellen, was nearly blind—a complication from type I diabetes that was exacerbated by being strangled by her ex-boyfriend. She recounted an interaction with a staff member whom she called for a ride when she got lost. The shelter only had one van and was transporting another resident. Ellen said, “I think they got a little bit upset. But I’m like, ‘What?’ [starts to tear up, voice shaky] See, stuff about my sight makes me feel bad.” Another resident, Maria, was in a wheelchair and although staff tried to accommodate her, the transportation options were less than ideal. This was made more poignant as Maria stated that when she came to the United States 17 years ago, she spoke no English and her abusive husband rarely allowed her to leave their apartment. In her words, home was “a prison,” and her abuser restricted her contact with outsiders. Not being able to easily leave the shelter to do errands or bring her children to school was emotionally challenging for Maria.

Luz, who does not speak English, was feeling ill at one point and decided to ask a staff member to take her to the store.

And so, I went downstairs and I asked if they would take me to Wal-Mart. They said, “For what?” And so, I told them “To buy something.” And so, she told me no. That we don’t go shopping until Saturday.

Having to explain her illness and having to ask for this “favor” (as she perceived it) was difficult for Luz, and she felt like her stay at Harbor was precarious, so she was reluctant to advocate for herself. If the staff member had known that she wanted to buy medicine, they could have made the accommodation, but Luz decided to walk to the drug store instead.

And so, I went on foot. I went to buy the pills, and then I went to lay down and I said—it really made me want to cry, because if she had asked, “Do you feel bad? Do you feel bad or something? We don’t usually do trips like that, but if you feel bad, we can take you. It’s okay!” But no, it turned out that they just said no.

Managing Chores, Cleaning, and Shopping

The most common and persistent complaint from residents during the 12 months of interviews was the cleanliness of the shelter and, in particular, the large shared kitchen. Harbor does not have the budget to pay for cleaning services and instead relies on the residents themselves to keep the common areas of the house clean. Staff stated that they are purposeful about not assigning chores—they do not want to unintentionally trigger any trauma for survivors. “I know it’s a domestic shelter so they really don’t want to enforce rules because it brings back memories of domestic [abuse] or whatever. And it was a mess. I was like cleaning every day behind basically everybody.” In every interview, I heard a similar version of the same complaint. Staff had communicated to residents why they did not mandate chores, but residents were frustrated with the outcome.

Power struggles over chores took on a more serious tone for women who had health concerns or were mothers. Despite being against policy, residents told stories of staff postponing or withholding grocery trips to put pressure on residents to clean the house. Rhonda is diabetic and insulin dependent and had part of her leg amputated because of her condition. She understood why staff did not require residents to do chores, but she was critical of staff methods for trying to get residents to keep the house clean.

They don’t enforce cleaning because they say it’s a form of abuse. So, when the kitchen’s a mess, it stays like that and I feel like mainly I’m the only one that really cleans…But there was one time when I needed to go to the store and I’m diabetic and I had an episode and they were talking about they would not take me to the grocery store because the kitchen was not clean.

Another resident, Jen, told a similar story months later: “They hold it over our heads because the kitchen’s not clean…I had to clean the kitchen last time because my baby needed milk that bad. I cleaned the whole kitchen by myself.”

Several of the mothers at Harbor had open DCF cases or feared that they might end up entangled in the system. A few months into my visits, residents stated that staff members had threatened that the health department or DCF would be visiting the shelter and might shut it down if it were found to be filthy. It put the women who were mothers into a panic, and they enlisted nonmothers to help them clean the house. It had the effect that staff desired—the house got clean—but it caused at least one fight between two residents over whose children had made the mess and caused a great deal of stress in the shelter. One resident explained, “Everybody got along, basically, until that day because they [staff] scared the people by saying that. And then they were lying. And nobody never came. Nobody never came.” When the director of Harbor learned that staff were using threats and coercion to get residents to clean, she reminded staff that it was not shelter policy to make weekly shopping contingent on kitchen cleanliness, and they were not allowed to threaten impromptu visits from DCF or the board of health.

Later, when staff tried to implement a “carrot” approach instead, which would incentivize cleaning the kitchen by offering a gift card to those who did, residents were disappointed when budget constraints meant that staff members could not follow through. In a lively exchange during one focus group, residents had sharp criticism for these shelter practices. When Zinnia asked a staff member for the gift cards, she was told that they did not have the donations to provide them. Jen said, “Well then, we ain’t got the hands to clean the kitchen” and Stefanie replied, “Well then, they ain’t gonna have the food for y’all to eat.” Olivia pointed out, “And haven’t we already all been through enough punishment, to begin with, for them to be saying we can’t have food?” Kitchen cleanliness and food had become a tense power struggle between residents and staff. The policy of not mandating chores that was part of the shelter’s feminist, trauma-informed framework was not experienced as empowering by residents, and funding constraints meant that attempts to remedy the problem were not successful.

Relationships With Staff

Residents wanted to feel respected by staff and they appreciated when staff members opened up about themselves or offered nonjudgmental moral support. Conversely, when staff kept residents at arm’s length, residents felt judged. Survivors did not necessarily know whether staff had experiences with interpersonal violence themselves and therefore whether they could empathize. The most poignant example of connection with a staff member came from Rhonda, who was living at the shelter when she learned that her 28-year-old son, her only child, had been murdered along with his best friend. “I just think about the way he went and I wasn’t there and there was nobody there.” After his murder, Rhonda shared with staff that she was suicidal, and they called an ambulance to take her to the hospital. When she came back to Harbor, the other residents were very kind to her, and one staff member went out of her way to be a listening ear, which made a big difference. Rhonda said,

When my son passed away, she was there. To me she was the only one that was really there. She was like, “Come on, let’s go out and smoke a cigarette.” And we sat out here and talked for hours. And she was very comforting.

The staff offices are physically separate from the residents’ living areas at Harbor. Staff are on a lower floor, and residents discussed their perceptions of whether staff tried to “come upstairs” to get to know them. While daytime staff are pulled in multiple directions, answering the hotline, getting ready for new arrivals, doing paperwork, transporting residents, and so on, nighttime staff could spend more time with residents, playing games or talking informally. Thus, residents had more opportunities for these kinds of interactions with nighttime staff. Maria said, “there are those who are very sweet, or those that you feel you can share things with. There are others where no, it’s barely ‘Good morning,’ ‘Goodbye.’” Tanya says she finds staff to be friendly, but “surface friendly” or “superficial,” and Pilar agrees, saying that many staff seem “like somebody working in a supermarket. Like you have to be friendly to everyone, even if they treat you bad.” When staff members were friendly and accommodating, listening to them, and being proactive in helping with their needs, it left an impression. For example, one nighttime staff member was an appreciated “alarm clock” in the mornings while another was tenacious about helping to figure out the confusing bus schedule. Residents especially appreciated that the night staff opened the office door and had memorized their names.

Survivors also felt that they would be able to relate more easily to staff who had shared similar experiences of abuse. In my informal conversations with staff members, it was clear that many had had personal experiences with interpersonal violence, but in their professional capacity at the shelter, they did not tend to share this with residents. Many residents interpreted this professional boundary as a wall that separated them and made it difficult to relate. Brenda said,

They can never be equals to somebody that’s been beaten on, unless they were beaten on themselves. You can’t—there’s just no comparison…Would you feel equal to somebody that’s telling you, you know, you can’t do this or you can’t do that?

If staff were more open about their own experiences, residents said that it would build trust. Brenda continued,

I think if they went through that downstairs they would be more open about it and they would let us know that they’ve gone through that [abusive relationships]. You know, to be able to relate, and try to make us feel more comfortable and have a little more faith and trust in them that they’re going to do everything in their power to help us get where we need to go to be part of society in a positive, productive way.

For some survivors, the professional boundary created a barrier to trust.

On Timing Out or Being “Kicked Out”

In every interview, survivors expressed relief at having a roof over their heads. The most common anxiety expressed was how long they would be able to stay at the shelter and where they would be going next, since no one wanted to go back to an abuser. For women who had an income over a few hundred dollars a month, they could qualify for the transitional living residence nearby, but 90% of respondents were not working. Staff helped residents navigate the complicated housing bureaucracy and aided with job searches.

First, residents expressed anxiety about the limited amount of time they are told that they can stay at the shelter. A resident who attended three focus group interviews explained that they can stay for 60 days and may then apply for a 30-day extension. I met several women who had been at the shelter for 4 months or more, so staff had discretion, but this process was not transparent to residents. Applicants for government housing vouchers are often put on a long wait-list, even if someone qualifies for the program. Tanya says, “It’s a wait. And like, oh 60 days isn’t gonna cut it. 90 days isn’t gonna cut it.” Pilar continues, “Yeah and they don’t give you enough time and then you’re just there with ‘Now what? Do I just go on the street?’” Tanya worried, “The doors are closed…I’m like, okay so do I go back then?” After just a week at the shelter, Ellen (who is legally blind) expressed similar anxieties about whether the 90-day stay would be sufficient for her to find a job and her own housing. “I’m worried about that, you know. And I—I’m sorry, I don’t want to find myself dead. I don’t, you know? I’m afraid I’m gonna feel pushed to go back with this guy.” Aside from feeling forced to go back to an abuser, the specter of ending up in a homeless shelter loomed. Christine had two young children with her in the shelter, “It’s not encouraging, it’s discouraging. It’s very depressing.” Jackie replied, “I started crying when they told me that” and Christine confirmed, “She really started crying when they told her that, you know. You have a child. Why would you want to go from a shelter to, to another shelter?”

Harbor residents had all heard of someone who had been asked to leave the shelter and were anxious about this happening to them, even if they were following the rules (when someone is asked to leave, it is typically due to violence or drug use). Thus, Harbor residents sometimes put up with aspects of the living situation that they perceived to be less than ideal. The first time Luz attended a focus group, she did not participate beyond a few words. The following month, she opened up more and stated that her infant daughter and two older children had been taken from her and put in DCF custody. Her stay at the shelter so far had been difficult. Luz speaks no English and her roommate at the shelter speaks no Spanish. Her roommate was often high on pills and alcohol and stole $50 from Luz on Luz’s second day at the shelter. Luz worried about asking to switch rooms though because she was worried that staff would see her as a “troublemaker” and ask her to leave. Being an undocumented immigrant added another layer of fear and uncertainty about advocating for herself. Luz said,

It scares me to complain because they can make me leave, or put me out. And I want my children. I love my children and my concern right now is whether they [DCF] are going to give them to me.

She goes on to say, “I sleep downstairs [in the living room] because they can tell me to go if I’m complaining too much, you know? So, better that I shut up and wait. I just want my children back.”

Despite fears of being asked to leave the shelter, some residents did find ways to advocate for themselves when they felt that they were receiving especially poor treatment from staff. Residents occasionally threatened to call the umbrella organization that oversees all the domestic violence programs and services statewide, for example. As Christine said,

I got sick, my kids got sick. It’s a health hazard. Sometimes we don’t even have heat in our room. It gets so cold. I told them, I said I’m going to call [the umbrella organization] and say we don’t have no heat. I told them that. And ever since I said that the heat’s been bumping.

Other residents were not convinced that there would be accountability, even if they called to complain. Vera said, “One of the staff mentioned to me, ‘Oh, every time you girls have a row, you’re gonna call the [umbrella organization]. You know what [they] do?’ Like, basically telling me [they] don’t do [expletive].” Lack of power and accountability were challenging for residents.

Discussion

This study found that despite formal policies reflecting feminist ideals of empowerment, practices reported by survivors in the shelter were disempowering at times. This was particularly true for women who were not fluent in English or had immigration issues, for mothers with pending DCF cases, and for those with physical disabilities. Financial constraints and the emphasis on professionalization and bureaucracy exacerbated these challenges.

While monitoring practices may make some residents feel safer, they were perceived as problematic for others, since one hallmark of abusive relationships is the abuser’s insistence on monitoring and controlling the victim’s movement and whereabouts (Yllo, 1998). For example, a study of women in homeless and domestic violence shelters in Ohio found that when individuals’ private decisions are monitored and scrutinized by staff, it can lead to social and organizational control, which is disempowering for residents (Hartnett & Postmus, 2010). Similarly, limiting transportation (whether intentionally or due to lack of funds) can be triggering for residents, since isolation is another common abuse tactic (Yllo, 1998). While many shelters require residents to participate in household maintenance duties (Hartnett & Postmus, 2010), Harbor’s choice not to require residents to do chores is a constant source of tension in the shelter. Incentives did work to the extent that staff could fulfill their promise to provide gift cards. Given financial constraints, however, this was often not possible. Staff could consider allowing residents to sign up for chores on a rotating basis, framing it within a feminist narrative emphasizing that the house is a communal space that everyone is responsible for maintaining. Allowing choices and emphasizing communal care and respect for the space could provide an opportunity for residents to feel more ownership over the space, which might ultimately be experienced as empowering. Using the specter of DCF involvement is not consistent with feminist practices, and battered mothers often decline to seek services in the first place, out of fear that their children may be removed from their care (DeVoe & Smith, 2003).

The value of sharing one’s own history of violence was documented in the study by Rodriguez (1988):

In the bureaucratic type of social agency organization, the personal history, political views, and personal life of the professional are separate from, and irrelevant to, the relationship with the client. In contrast, the staff at the [shelter] tries to personalize their communication with the residents. (p. 219).

One staff member interviewed in that study argued, “We need to keep sharing our personal life with the residents so that they can understand where we are coming from” and so that staff can serve as role models of empowerment (p. 219). Researchers have found that when residents and staff have equal partnerships, when staff members are not authoritarian, and when there is a sense of closeness with staff, battered women’s shelter residents gain strength (Haj-Yahia & Cohen, 2009). Staff at Harbor, on the other hand, did not tend to cultivate this closeness or equal partnership.

This study finds that tensions between shelter workers and residents were caused by two main factors. The first were financial challenges that limited how staff could respond to residents’ needs. Staff consistently made do with shrinking and precarious budgets. Not being able to reliably provide bus passes, gas cards, accessible van transportation, private bedrooms for each resident, or to hire cleaning staff or provide gift cards to thank residents for cleaning made it difficult for shelter staff to meet residents’ needs. These are concerns that are not unique to the shelter in this study, but rather are being faced by shelters across the country (Mantel, 2013). This was compounded by the precarious economic situation of the majority of women who sought shelter services (Dichter & Rhodes, 2011). When direct service organizations need to focus on writing and managing public and private grants, cultivating private donor relationships, holding fundraisers, and training volunteers because there are not enough paid staff, they have fewer resources to spend on the critical work of providing direct services to survivors.

Second, this research suggests that the standards of professionalism and bureaucratization of the shelter system caused tension at times between staff and residents. The question becomes, can domestic violence organizations live out feminist ideals of equality and empowerment within a bureaucracy? Rodriguez’s (1988) critique of bureaucratically organized domestic violence shelters centers on the hierarchy of professional staff who treat survivors not as equals but as clients who must adhere to rules and regulations to remain at the shelter. In the present study, when staff were perceived as overly distant, survivors reported feeling judged and unhappy with their treatment at the shelter. Conversely, survivors who felt connections with staff members had a more positive evaluation of their experience at the shelter. These connections were described not so much as personal disclosure by staff, but as respectful dialogue with staff members who seemed invested in survivors’ well-being and personhood, which is consistent with the code of ethics in social work (National Association of Social Workers [NASW], 2017).

This study concurs with Koyama (2006), who acknowledges power imbalances that existed between women of color and white women, U.S.-born women and immigrants, and service providers and recipients in the preprofessionalization era and does not advocate returning to it. Certainly, survivors benefit from the skills and expertise that professionally trained social workers and other personnel possess (Messing, Ward-Lasher, Thaller, & Bagwell-Gray, 2015). Instead, Koyama (2006) argues for acknowledging these power differences and pushing to change policies that result in paternalistic treatment. Soliciting input on agency policies from survivors, as the present study does, is identified as key to reducing paternalism (Koyama, 2006). More accountability and independent grievance procedures within the shelter system would also be part of a survivor-centric approach. Harbor residents who wanted redress when they experienced what they perceived to be poor treatment had few viable options and many feared that advocating for themselves might have negative consequences for their stay at the shelter. Utilizing an independent advocate who works solely on behalf of survivors would be a way to provide this accountability.

While the intention of this research was to shed light on how survivors felt empowered or disempowered, participants were more likely to report feeling disempowered. This was particularly true for survivors who were undocumented immigrants or non-English speakers and for those with physical disabilities. Hiring more staff who are bilingual and upgrading the facilities and transportation to accommodate those with physical disabilities would require additional resources but would be an important priority. In addition, future research that follows survivors longitudinally could look at how shelter dynamics and policies identified in the current research might affect the potential for long-term success after leaving the shelter, taking into consideration intersections of gender, ethnicity, immigration status, language, and disability.

It is important to note that the findings presented here are not intended to be generalizable, but rather to offer a critical depth of information about the experiences of domestic violence survivors and whether shelter policies and practices are perceived to be empowering or restrictive. Placed in a broader context of financial challenges, as well as the dual pressures of professionalization and bureaucratization, these survivors’ stories may shed light on the challenges that social workers and other direct service providers at other shelters face, even as they strive to empower survivors.

The goal of this study was to illuminate whether survivors at an emergency domestic violence shelter perceived shelter policies and practices as empowering or disempowering. The challenges of inadequate funding, combined with ongoing gendered economic inequalities and patriarchal institutions more broadly, remain significant. For social workers and other direct service providers, listening to the concerns of survivors and striving to create collaborative relationships are goals that are consistent with the ethical principles of respecting the dignity and human worth of the person and the importance of human relationships (NASW, 2017).