Bringing a Network-Oriented Approach to Domestic Violence Services

Abstract

Despite powerful evidence that informal social support contributes to survivors’ safety and well-being, mainstream domestic violence (DV) programs have not developed comprehensive models for helping isolated survivors reengage with these networks. Although many advocates use network-oriented strategies informally, they often do so without resources, funding, or training. This qualitative focus group study explored advocates’ use and perceptions of network-oriented strategies. Advocates working in a range of DV programs across one state described the importance of network-oriented work and articulated its five dimensions, including helping survivors build their capacity to form healthy relationships, identify helpful and harmful network members, reengage with existing networks, develop new relationships, and respond more effectively to network members.

Keywords

domestic violence intimate partner violence social support programs informal networks trauma-informed approach

Introduction

Intimate partner violence (IPV), that is, physical, psychological, or sexual abuse and stalking perpetrated by a current or former intimate adult partner, causes devastating physical, psychological, and economic damage to millions of people, primarily women, in the United States each year (Black et al., 2011).¹ As public recognition of IPV’s prevalence and impact has grown, so too has the public response. Over the last 35 years, antidomestic violence activists in the United States have transformed the country’s social services system responses to IPV, developing hotlines, shelters, and community-based services for women across the country.

At the same time, the last several decades have seen a dramatic expansion of scholarship on IPV. One focus of this scholarship has been the contribution of informal social support (support from family, friends, neighbors, and coworkers, among others) to emotional well-being and ongoing safety among survivors. This now robust body of research demonstrates that most survivors turn to their informal network members before, during, or instead of formal domestic violence (DV) services (Goodman, Dutton, Weinfurt, & Cook, 2003; Rose, Campbell, & Kub, 2000); that these networks contribute powerfully to survivors’ ongoing safety and mental health (Adkins & Kamp Dush, 2010; Bybee & Sullivan, 2005; Goodman, Dutton, Vankos, & Weinfurt, 2005); but that survivors can become isolated from potentially helpful networks, and sometimes network members can make things worse.

Despite the co-occurring growth of DV services and social support scholarship, however, the development of the former has not built on findings from the latter. Even in the face of powerful evidence that informal social support contributes to survivors’ safety and well-being, mainstream DV service models have not yet articulated as a core part of their work strategies for helping isolated survivors reengage with these networks. These strategies comprise what we call a “network-oriented approach” to services (Goodman & Smyth, 2011).

Although many advocates in mainstream DV programs appear to engage in network-oriented practices—often using innovative and creative strategies to help survivors move out of isolation—they often do so in a way that is not systematically named, credited, evaluated, or resourced (Goodman & Smyth, 2011). Understanding and framing these strategies would seem a key step toward supporting advocates who use them, deepening the conceptual underpinnings of network-oriented work, and creating new avenues through which to engage community members. The purpose of the study was to understand how, when, why, and to what extent DV advocates use network-oriented practices, and how they perceive the costs, benefits, and challenges of those activities. Before turning to the study itself, we begin with short reviews of, respectively, survivors’ use of informal social support and the nature of current IPV services.

Survivors’ Use of Informal Social Support

Every act of DV occurs within a specific community. Both the partner who is abusive and the partner who is abused are embedded in relationships with a variety of people who can contribute to the maintenance or alleviation of the problem (Mancini, Nelson, Bowen, & Martin, 2006). It is therefore not surprising that research, in a variety of ethnically diverse samples in the United States, shows that two thirds to virtually all IPV survivors access informal social support (defined here as the availability of instrumental and emotional assistance through family, neighbors, coworkers, or friends) to address the violence, whether or not they also access formal support (i.e., formal services such as the police, crisis hotlines, shelters, or community-based services); and that it is their informal networks that provide the most helpful and enduring support in the form of both practical and emotional assistance (Goodman et al., 2003; Mancini et al., 2006; Rose et al., 2000). Supporters can provide a wide array of instrumental or practical assistance such as a place to stay, transportation to needed help sources, child care, financial assistance, or resources that support the survivor’s safety strategies or that enable the survivor to participate in formal services (Fleury-Steiner, Bybee, Sullivan, Belknap, & Melton, 2006; Riger, Raja, & Camacho, 2002). Network members can also provide a broad range of emotional supports, including a shoulder to cry on, ideas about how to stay safe and parent within the relationship, encouragement to take steps toward safety, and commitment to stick with the survivor no matter what (Goodman & Epstein, 2008). In each of these cases, network members can build on dimensions of a survivor’s identity (e.g., informal matriarch of her neighborhood block, compassionate and competent child care provider) that might be invisible to or discounted by the formal service system. Moreover, cross-sectional and longitudinal studies indicate fairly robustly that these informal social supports contribute dramatically to survivors’ emotional well-being (Adkins & Kamp Dush, 2010; Coker, Watkins, Smith, & Brandt, 2003) and physical safety (Bybee & Sullivan, 2005; Goodman et al., 2005). Among survivors in shelter and community samples, social support is related to lower levels of suicide risk, mental health difficulties, and general distress (Adkins & Kamp Dush, 2010; Kaslow, Thompson, Brooks, & Twomey, 2000; Thompson et al., 2000) as well as lower rates of reabuse over time (Bybee & Sullivan, 2005; Goodman et al., 2005).

Several complementary theoretical models illustrate how social support works to improve survivors’ health and well-being: The main effects model holds that social support contributes to survivors’ well-being regardless of the amount or severity of abuse experienced; the moderator or buffer model holds that social support functions as a protective factor, mitigating the impact of IPV on survivors’ well-being; and the mediator model holds that social support mediates or explains the relationship between abuse and various outcomes. Evidence exists for each of these models (Beeble, Bybee, Sullivan, & Adams, 2009), and they certainly are not incompatible.

Of course, social networks are imperfect conduits of assistance for IPV survivors, just as they are for those who do not have violent partners. Family and friends sometimes fail to understand or express sympathy; become too frightened for their own safety (often for good reason) to provide support; push survivors either to leave before they are ready, or to stay to preserve the family; blame the abuse on the survivors themselves; or deny the complexity of the situation and overestimate survivors’ power to change things (Goodman & Smyth, 2011). Converse to the positive benefits provided by helpful network member responses, negative responses may actually decrease survivors’ well-being and increase their risk for reabuse (Bybee & Sullivan, 2005).

Moreover, despite the clear benefits of informal social support, survivors in shelter and community samples report lower levels of both emotional and practical forms of assistance than nonsurvivors (Levendosky et al., 2004; Thompson, Kaslow, Kingree, et al., 2000). This is not surprising, as many abusers take explicit steps to isolate their partners. But IPV can also disrupt relationships in less direct ways: Survivors may “use up” friends’ and family members’ willingness to help, feel ashamed to admit the abuse to potentially critical network members, or feel reluctant to burden already-stressed network members (Dunham & Senn, 2000; Rose et al., 2000). Whatever the reasons, the result is often diminished access to social support, leaving a survivor with fewer resources to deal not only with the abuse itself but also with other needs as well. Given that, ultimately, women’s ongoing connections are to their family and friends, not to DV services, it seems critical to consider whether and how DV programs support these connections.

Current IPV Services

About half of DV survivors access formal services, particularly crisis hotlines, emergency DV shelters, and community-based programs that provide services ranging from safety planning, counseling, and peer support groups to job search support and housing assistance (Coker, Smith, Bethea, King, & McKeown, 2000; Du Mont, Forte, Cohen, Hyman, & Romans, 2005). Despite the extensive literature on the contributions of informal social support to women’s physical safety and emotional well-being, however, most services do not make the maintenance and development of informal support networks central to their models. Indeed, in a prior article calling for a network-oriented approach to DV services (Goodman & Smyth, 2011), we explored how mainstream service models may even further survivors’ isolation by, for example, requiring that survivors move away from their communities if they want shelter or avoid contact with each other outside the bounds of a community-based peer support group.

Of course, the absence of attention to network-oriented practices in the formal documentation of mainstream service models does not necessarily mean that such practices are absent in practice. Recognition of the key role of survivors’ informal support networks has triggered a number of innovative grassroots efforts to prevent DV through changing social norms or to equip network members to intervene more effectively (Fullwood, 2002; U.S. Department of Health and Human Services, 2004). These efforts—usually initiated outside the bounds of mainstream domestic services—are based on the idea that network members are not just supports until a space opens up in a shelter or community program; rather, they are critical partners who can fill some short- and long-term needs as well as, if not better than, professional services (Goodman & Smyth, 2011; Kim, 2002). Furthermore, our own work with DV programs has revealed the range of interesting practices advocates use to help survivors connect with others. But to date, very little research has documented these strategies empirically or described the way advocates perceive them. This study attempts to redress this gap by using a qualitative focus group methodology to understand whether and how one state’s DV service providers engage in network-oriented practices. Our aim was to describe the range of practices now being employed, whether or not they are officially documented, as well as the rewards and challenges encountered in implementing them.

Method

Focus groups are unique in that they yield data from the individual as well as from the individual as part of a larger group (Massey, 2010). Both are important. Although individual stories alone are critical to understanding the nature and contours of a phenomenon about which little is known, the dynamic quality of group interaction creates the possibility of generating fresh insights that may be new even to individual participants (Wilkinson, 2003). In addition, focus groups lend themselves well to learning from populations that are used to thinking and making decisions as teams.

Sample and Recruitment

In the spring of 2011, we conducted three focus groups involving DV advocates working in 1 of the 14 agencies belonging to one state’s Domestic and Sexual Violence Coalition. The Coalition meets on a monthly basis to give staff across member programs a chance to work on new initiatives and compare notes on ongoing issues such as how to do trauma-informed work, shelter rules, or the nature of professional boundaries. A focus group on the subject of network-oriented work therefore fit nicely into the mission of the Coalition.

All Coalition member programs adhere to one basic model in that they each provide community education and outreach, emergency shelter and/or transitional housing, support groups, and direct service to survivors living in the community. All build on an empowerment philosophy that involves informing survivors of the community-based programs and legal strategies available to them and then helping them access the resources they need. Given the small size of each program, all program administrators, with no exception, also provide direct services to survivors.

To recruit potential participants, the director of the Coalition sent out an email invitation to all directors, shelter managers, and direct service coordinators who worked in 1 of the 14 member agencies. The emailed invitation described the study as an opportunity to learn about how and to what extent DV service agencies help survivors reengage their social networks, develop new ones, or support network members directly. Focus groups were scheduled for the 1.5 hr following the Coalition’s regularly scheduled meetings. Those who did not want to participate simply left after the meeting.

Out of 33 executive directors, shelter managers, and direct service coordinators in the state, a total of 28 participated in one of the three focus groups: 10 executive directors, 10 shelter directors, and 8 direct service coordinators. We organized the groups to be homogeneous in terms of job title to avoid a situation where participants would be in a group with someone who had greater or lesser job-related power. All participants were women and all but one (a Native American woman) were White. The average number of years at their jobs was 12.9, with a range of less than 1-35 years.

Procedures

The first and second authors facilitated all three focus groups. We began with a simple definition of network-oriented work as practices that helped survivors strengthen existing connections or develop new ones, or that helped equip network members to support survivors. We went on to ask a series of open-ended questions such as “When and how does your agency help survivors identify and engage their existing network members?” “Do you find any differences in your network-oriented work with survivors across race/ethnicity, age, sexual orientation, or immigration status?” “How do you provide training and supervision for network-oriented work?” “How do these practices benefit and/or harm survivors?” “What are the challenges of this work for your organization?” All focus groups were audiotaped and transcribed.

Data Analysis

Transcripts were coded using qualitative content analysis, a commonly used approach to analyzing focus group data (Wilkinson, 2003). This process requires that researchers stay close to the data, eventually developing a coherent description of the phenomenon of interest that is relatively low-inference (Charmaz, 2004). We conducted three levels of coding. In vivo or open coding involved keeping as close to the data as possible to generate short and immediate codes that captured the language of participants; second-level coding involved combining first-level codes into higher order categories with shared content; and third-level coding involved integrating categories into clusters based on the relationships among them (Graneheim & Lundman, 2004).

We sought agreement among coresearchers through a process of peer review (Lincoln & Guba, 1985). Peer review does not necessarily mean that researchers independently arrive at the same coding decisions. Instead, it requires that a “dialogue among co-researchers” about coding decisions is present and active throughout analysis (Graneheim & Lundman, 2004). Following a long process of training and practicing with the first author, the third and fourth authors did the first round of coding under the first author’s supervision. We started with the first focus group and then added, altered, and condensed codes as we analyzed the second and third focus groups. Where there was disagreement, we discussed the code, category, or cluster at issue until we reached consensus. Throughout the process, we continuously shared our results with the other two authors, who served as auditors and whose contributions enabled further refinements of codes, categories, and clusters.

Results

It appears that the advocates we interviewed employed a variety of creative network-oriented practices that targeted not only survivors themselves but also their networks and the broader community. Although participants were unanimous in their enthusiasm for this kind of work, they were equally forceful in pointing out the challenges and tensions that arise when doing it.

As a prelude to the “Results” section, it is important to note at the outset that participants in each of the three focus groups described a similar set of practices and tensions. Given the strong agreement on major themes across participants and across focus groups, we present the themes without reference to how many people articulated each. To do otherwise would be misleading given that there was not time for every single group member to speak in depth about each of the issues discussed. Nodding assent was sometimes sufficient. Finally, most (though not all) of the tensions described in these pages were presented as tensions within people rather than across people. We make sure to represent divergent views, however, on those few issues about which there was disagreement.

Specifically, this section describes the six distinct clusters that emerged through the coding process. Figure 1 presents a visual model of the clusters, and Table 1 lists practices that fall within Clusters 1 through 5.

Figure 1.

A network-oriented approach to IPV services.

Table 1.

Practices Within Clusters.

Cluster	Practices
Prework	Helping them regain a sense of themselvesTeaching them to identify safe and unsafe relationshipsSupporting their early efforts to interact in new and safe waysHelping them establish a psychological connection to a larger community
Supporting survivors to build internal foundations as preparation for reconnections by:


Taking stock	Asking good questionsUsing visuals (e.g., “web of safety”)
Supporting survivors to assess how existing network members might help or harm by:	Asking good questionsUsing visuals (e.g., “web of safety”)
Reconnecting	CoachingRole-playingHelping to manage expectationsProviding emotional supportThinking creatively and bending rules to make it possible for survivors to rebuild relationships
Helping survivors reengage existing networks through:



Starting new	Supporting them to act on developing interestsProviding opportunities for them to participate in DV or other support groupsHelping shelter residents connect safely with each otherProviding ongoing emotional support as survivors build new ties
Helping survivors to develop new relationships by:


Moving outward	Helping them understand and respond compassionately to survivorsEducating them about services for survivorsProviding emotional support as they reengage with survivors or discover abuse in their own lives
Working directly with network members to ensure that they are sources of support and not harm by:

Note. DV = domestic violence.

Cluster 1: Prework—Building Internal Foundations

Most participants made clear that work to build survivors’ networks had to begin with helping them repair abuse-related damage to their sense of self and responses to others. Some called these practices “building internal foundations” or “trauma-informed practice.” Others called them simply “prework.” We adopt this latter phrase to distinguish it from the more direct practices participants used to support survivors’ engagement with others.

First, many participants spoke eloquently about the ways in which survivors had become unable to attend to their own needs and interests or been forced to give them up over the course of the abuse. As a result, these participants felt that one of their first tasks was to help survivors identify (or reidentify) their own needs, feelings, goals and interests, as a pathway toward engaging with others. As one participant described it,

Everything they have done has always been, you know, what the abuser wanted and before that what their parents wanted and they haven’t really developed a sense of who they are. And they need to be able to see themselves as their own person [before connecting with someone else].

A second core prework practice involved helping survivors sharpen their capacity to assess the physical or emotional safety of a relationship, a capacity that the abuse may have disrupted. For example, one participant said,

They come to us from a controlling relationship . . . and then they go into the arms of their friends or family who also are controlling? They don’t see that. They see it maybe as caring. Or whatever, or they just need that help so much. . . . We try to work with them to help them see maybe this relationship isn’t so healthy. . . . And that can be very difficult because they’re like, this is my mom.

Third, participants talked about prework as helping survivors relearn to communicate with others in a way that did not produce shame on the survivor’s part or withdrawal on the network member’s part. For example, one participant talked about how, in her opinion, survivors often share too much in a way that leads to feelings of shame and regret. This participant partly blamed the system, as well as prior abuse, for what she considered “oversharing”:

One thing that we ask of women who are seeking services, everywhere they go, they have to tell their story. . . . It starts to be like an expectation. So how do you rein that in? How do you set new boundaries so that you know how to have a different kind of conversation with, for example, other parents at the school? . . . It’s OK to have compartments for your lives and I think for a lot of the victims that we work with they’ve opened the floodgate and they opened the floodgate everywhere and that’s so unsafe in a lot of ways.

These two dimensions of safety—identifying “safe” people and having safe interactions with them—were often discussed in tandem, as in the following example:

[Many survivors] don’t even know who is a safe person. How do I tell if this is a good friend? What are the red flags I should be looking for? [Some have too few boundaries and] tell someone their life story and that backfires; and other ones, they say that they have like put up land mines and razor wire and they’re assuming that everybody is bad, and they don’t want to let anybody in there.

In these instances, prework involved helping survivors develop new templates for thinking about whether a relationship is safe.

Finally, some participants talked about prework as helping survivors regain a sense of the larger community as potentially safe and welcoming, a place to go not only for help with problems but for entertainment and social contact. Participants helped survivors achieve this sense by organizing a variety of activities in the local communities. This, in turn, meant developing partnerships with local community members, including artists, Karate teachers, or clergy, and with local community agencies, including churches, retail stores, or educational settings. Sometimes, participants simply took survivors for outings in the community as in the following example:

We’ll load women up who want to go to the free concerts downtown. We’ll give them rides to the library and that kind of thing and sort of connect them back to the community so they get a sense that they can rebuild their lives and that life can be good. You know what I mean? They can do things that are fun and that there is more to life than that abuse.

Cluster 2: Taking Stock—Assessing Survivors’ Existing Networks

Alongside “prework,” helping survivors “take stock” of existing network members was considered foundational to network-oriented work. Taking stock meant doing an initial assessment of participants’ existing support systems—to determine who could help, what kind of practical and emotional support they might provide, and who might not be helpful or safe.

Most central here was the simple process of asking survivors to explore who in their networks might be helpful at the moment. Which ones are connected to the abuser? What might each do to help? How have they responded to requests for help in the past? And how has the survivor generally responded to that assistance? As one participant summarized,

One of the things we use as a visual with women [is] the Web of Safety to really look at what are the strong threads of safety [depicted as solid lines] and which ones are dotted [to indicate] that they may be connected to the abuser. So there’s a visual for women to really look at. . . . Like [they might identify] a religious leader, like Father so-and-so or Sister Mary so-and-so, and then they realize that the abuser really is also connected to them and is this priest trustworthy or is this person trustworthy?

Irrespective of the approach used to helping survivors take stock, participants noted that the results of the process were often surprising for survivors who may “really have just forgotten or haven’t even thought to seek that out and just feel so isolated.” Indeed, participants said that sometimes this brainstorming process helped survivors identify people they would not have thought about on their own, as in the case of one woman who reconnected with an elementary school teacher who had been a support for her in the past and now again became a resource.

Cluster 3: Reconnecting—Helping Survivors Engage Existing Networks

A small number of participants believed that once they helped survivors take stock of their networks, it was up to the survivors to figure out how and when to access those networks. But most felt that it was part of their job to help survivors do the actual work of reengagement. This was especially important, some participants noted, if survivors were considering returning to (or continuing to live with) their abusive partners. Participants helped survivors engage with their existing network members in several different ways.

First, participants used coaching and role-playing mainly to help survivors anticipate potential bumps in communication, manage expectations, and deal with rejection. For example, one participant described her work to help survivors manage the ups and downs of rebuilding their relationships:

There are all these hurts . . . so the reconnections are very tenuous and they’re lots of times in fits and starts and you know there’s a connection and then a disconnection and then a connection again while everybody’s testing the waters to see if it’s safe. And I think . . . you have to be with them to kind of coach them through that disconnection so that when they reconnect again they’re not going to have too high of expectations.

Second, some participants saw their job here as simply providing emotional support as participants experimented with various types of reconnection. As one noted,

And I think that gives them a real, a real sense of a solid footing where they feel more empowered to go out and make friends or to go try something or to reconnect with a family member. Because even if it doesn’t go well they know they can come back and tell us, it really didn’t go well.

Finally, many participants emphasized the need to “bend the rules” to be effective in helping clients reconnect with network members. One example of this was when a shelter decided to hold someone’s bed for 7-10 days so that she could go to her daughter’s college graduation in another state. This went against protocol but seemed extremely important for that survivor. Another participant discussed adjusting rules to make room in the shelter so that a resident could have overnight visits with her children on weekends even though the children were not regular residents with her. Whether acting as coaches or bending rules, participants were clear about the need to help survivors connect back to social networks that existed before contact with the program.

Cluster 4: Starting New—Helping Survivors Develop New Networks

Sometimes, it became clear that survivors could no longer rely on existing or old networks and needed to develop new ones. This was sometimes the case for first generation immigrant survivors whose abusive partners were embedded in the very small immigrant community of which they were a part. Sometimes, these survivors felt that they simply could not go back to their community without risking further abuse. This dilemma is reflected in one participant’s story:

We had a woman come from [name of foreign country] and she couldn’t go back into the community. One, she had been isolated from it and two, we didn’t know who knew her husband. She was brought over as an immigrant and sponsored by her husband. Which made it impossible for her to access some of the benefits.

When this was the case, participants helped survivors develop new ties through a variety of practices, some of which parallel those used in the previous cluster.

First, building on the prework stage, many participants tried to help survivors lean on key aspects of their identities as a way to reach out to others. Participants spoke about linking survivors to religious settings, helping them find scholarships to attend school, and helping them access specific clubs, groups, or classes (e.g., Alanon, Karate, drawing class). They did so in service of the twin goals of deepening survivors’ sense of self (as described under the section “Cluster 1: Prework—Building Internal Foundations”) and connecting them with people with shared interests and identities, as in this example:

Maybe they like to knit. Maybe they like to scrapbook or maybe they like to sew. Or kinda start getting to get them more connected with those kinds of social networks so they can start making friends that have similar interests.

This strategy was especially important, participants noted, for new immigrants. Given the demographics of the state, it was sometimes challenging to help them find and connect to people with similar backgrounds, as the following story illustrates:

We had a woman at our shelter who wanted to go to a culturally specific grocery store that was a little further away than the grocery store staff usually take our guests to. And staff were having her walk the extra distance and then picking her up. But it’s not just an issue of whether she can get the food that she needs, it’s also allowing her to connect to a cultural practice and potentially connect with other people who share that.

Often, participants used their own connections in the community as a bridge so that survivors could enter a new situation with some support, sometimes in the company of an advocate. For example, one woman said,

We have some good relationships with two different area churches. Some of the [survivors] have said . . . that is what helped me start to really heal. [With one woman], we arranged for her to go out to lunch with one of the ministers and meet her there, one of the women ministers was interested in doing that.

Creating new connections was challenging, however, in rural parts of the state, which were predominantly White. In these situations, survivors of color often experienced racism within the local community, which made them wary of reaching out and reluctant to stay away from their home communities. As one participant noted,

If it’s a woman of color or who goes into a rural shelter program in some of the outliers, she’s going to feel very ostracized by that community and so it’s going to [be] difficult for her to develop the support there. And if she’s going to find that she’d rather be back in the community where the abuser might be, rather than deal with the underlying racism within that community.

Second, participants talked about either encouraging women to join existing DV-related support groups, or creating new support groups if none were available. Among those running support groups, some discouraged participants from socializing outside the group. They worried that the risk of interpersonal harm (and, therefore, damage to the group dynamic) that might follow were simply not worth the potential gain. Others, in contrast, felt that it was important to allow or even encourage such outside connections, though they did so carefully. As one participant noted,

We talk a lot about what it will be like to see someone in the group outside of the group. We do it again halfway through the support group and discuss how it changes things that we’re halfway through. We try to acknowledge that relationships are going to develop.

Third, some advocates also supported their shelter residents to engage with each other both inside and outside the bounds of the shelter walls:

They’ll pair up and share resources and figure out, you know, OK this pantry is open on this day, and they’re guiding each other. There have been entire groups where, they might be from all different faiths but they’ll go around and they’ll say, “OK this Friday we’re going to attend this mass at this church” and they’ll go around with a lot, a lot of churches and show up, churches, temples you name it. . . . They just want to kind of get a taste of each other’s practices, and then decide which one they like. (Participant 1)

And then they feel stronger and they grow stronger and more confident in their abilities to maneuver through those systems and then into strange communities or back into their own. (Participant 2)

But alongside their descriptions of positive connections among shelter residents, some participants also noted that these relationships could be fraught and difficult, as described in more detail in the “Cluster 6: Tensions and Challenges” section.

Perhaps most importantly, advocates understood their role as simply being there with their clients—as sounding boards and emotional supporters—as survivors tried, and sometimes failed, to make new connections. One participant said,

And it’s really about providing that consistent support. So when the person, you know, you give them the option and they take the one you wish they wouldn’t have taken and sure enough what you thought would happen happened. But they know they can come back. And [when they do, you can ask] “OK now what’s plan B [given that] this is what happened?” And I think that gives them a real sense of a solid footing where they feel more empowered to go out and make friends or to go try something or to reconnect with a family member.

As a few participants noted, sometimes the only new relationships survivors have or can develop are with advocates themselves. One shelter director summarized the situation this way:

Something we’ve noticed too is that it’s important to have people just in the shelter who are simply there to connect. So we’ve really tried to have an advocate in the shelter who’s just there be a human. She’ll just sit there, or have a cup of tea with someone.

Cluster 5: Moving Outward—Working With Network Members Directly

In addition to working with survivors, some participants talked about working directly with network members to ensure that they were sources of support and not harm.

Participants talked about making contact with already-identified network members through a variety of mechanisms: Sometimes network members called the hotline themselves, sometimes a client would ask her advocate to talk to a family member with her or even on her behalf, and some network members came together in the context of a support group established by a particular DV agency.

Irrespective of how they came to talk with network members, participants listed three main goals in doing so. First, and most important, they sought to help network members understand and respond compassionately to survivors. Second, they wanted to educate network members about available services for survivors. And third, they wanted to provide emotional support to network members, who themselves were struggling emotionally. Illustrating the goal of building understanding and compassion, one participant said,

With a recent shelter resident, her previous [nonabusive] husband or partner—they have children that they share together. . . . And he was having a very difficult time understanding why she was making a lot of the choices she was making or just many aspects of what she was doing. And they were having arguments over the phone about the children about visitations, and because she had relocated into a [shelter]. She asked that we talk with him and have a conversation around what a victim’s perspective is and what often their situations are like and why they may be making these choices.

Illustrating the goal of education about services, one participant noted,

A lot of times they don’t know the system. And so if they’re going to accompany, if they’re going to go to court, they need to know what court is too, because oftentimes the misinformation can cause a lot of problems.

And exemplifying the desire to support network members emotionally, one participant said,

It’s also letting them know that when it’s someone that’s close to them that has been victimized, that they are also a victim as well. . . . And letting them know that our services are there for them as well, to come and talk, and also if they just need support or help that we’re there for them too.

These examples illustrate the wide variety of strategies participants used to help network members gain the knowledge, skills, and internal capacity to become stronger supports for survivors. But beyond working with survivors’ network members, participants also talked about venturing into the community to talk directly with community members. The main vehicle for doing so was educational workshops designed to help participants think about how to support survivors in their communities.

Cluster 6: Tensions and Challenges

Although every participant in every focus group emphasized the importance of network-oriented work with survivors as an ongoing set of practices, most also noted a series of tensions and challenges in doing this work. Although some of the challenges fit within just one or perhaps two of the clusters presented here, some ran through a few or most of them. They are therefore presented here in a separate cluster.

Timing

One commonly cited tension was that between helping survivors manage crises and helping them identify and reengage with potentially helpful informal supporters. On one hand, participants felt that crises were key moments in which to reach out to friends and family who might be available to help out with immediate needs such as a bed, transportation to court, or child care when, for example, the survivor sensed danger at home. But at the same time, participants worried that survivors in crisis were not in the best position to give thoughtful consideration to which network members might truly be supportive versus harmful. Overall, it seemed that participants who worked with survivors for the shortest periods of time (perhaps just once through a hotline) were less likely to do network-oriented work, with one exception: The practice of taking stock was fairly universal from the beginning, even when a participant talked with a survivor only once.

Supporting helpful connection while avoiding harmful engagement

In the context of discussing helping survivors connect or reconnect with others, participants highlighted the tension between wanting to help survivors do so positively and recognizing that such engagement carried risks. The content of participants’ concerns about reconnection ranged from worry that network members could end up undermining survivors by encouraging them to stay with abusive partners; that in reconnecting, survivors would have to deal with rejection, disappointment, stigma, and shame; or that network members themselves could be abusive, as in this example:

We don’t know those specific people. How do you know that that’s a safe place to send them basically? For us to try and have them reach out to that person, what if it really isn’t safe for them to do that, but we don’t know.

Finally, some worried about losing a survivor altogether by pushing her too hard in the direction of connection:

I feel like it really depends on the client though too. Because I know I had a client who she said that she was a recluse. She . . . didn’t socialize, didn’t go out. Only went to counseling which was a good thing. But . . . I encouraged her too much to, you know, to draw out of that shell where I lost her. You know, like she stopped coming.

It is important to note here participants never spoke about pushing a survivor to interact with any individual person about whom the participant expressed worry or reluctance; instead, most encouraged survivors simply to try to connect with someone in some way.

In addition to tensions inherent in supporting survivors’ relationships within the community, participants discussed the challenges of facilitating shelter residents’ capacity to provide mutual support to each other. Specifically, they articulated the tension between encouraging shelter residents to form relationships involving emotional and practical support and urging caution to ensure that women did not get in over their heads or come to feel too much responsibility for another resident as they worked on repairing their own lives. Exemplifying this tension, three participants talked about their ambivalence as they observed residents offering each other rides, though each resolved the ambivalence differently:

It’s a fine line of finding that balance. . . . If a woman comes in and she doesn’t have a vehicle, and the other person does, we don’t encourage that relationship that becomes dependent. (Participant 1)

No. We would encourage [that relationship] . . . If one shelter resident wants to offer it to another resident, that’s their decision. (Participant 2)

Rather than set forth more rules . . . we spend a lot of time providing education around how do you do that. . . . I mean I don’t necessarily feel comfortable saying you can or cannot do x, y, or z. . . . Perhaps one does have a car and one doesn’t. But you know what? The one that . . . has the car might be low income, but then the other one might have a little bit more and not have children so they may come up with some sort of symbiotic relationship whereby she says, “ you know what I’ll give you five dollars for [a ride].” I mean sometimes there’s an exchange in some way, shape, or form. . . . Sometimes they can balance with each other what their needs are and figure that out. (Participant 3)

The challenge of negotiating boundaries

Negotiating boundaries also arose as a key challenge to doing network-centered work. Advocates engaged in such work often found themselves in unusual situations that diverged sharply from traditional notions of service delivery. This, in turn, challenged their thinking about the kinds of relationships they wanted to have with survivors. For example, participants asked questions such as, What does it mean to survivors and advocates when advocates accompany survivors to community events outside the program? Is this inherently a violation of professional boundaries? Or does it depend on the nature of the relationship and the purpose of the visit? And how might one survivor feel if she is not accompanied to a community event when another survivor is? What might it mean to survivors when, with permission, advocates speak to survivors’ network members directly? Even assuming that permission has been granted, does this run the risk of breaching a professional boundary? When is it appropriate to take advantage of a personal connection in the local community to help a survivor develop or deepen a particular interest?

Although these questions triggered exciting and passionate discussions, they also raised profound disagreements: Some advocates strongly preferred to stay within the confines of clearly defined and fairly traditional boundaries, while others wanted to completely rethink how to establish boundaries with survivors. Among the former group, some simply felt uncomfortable with any attempt to question or rethink the boundaries that they had been trained to use and that they equated with professionalism. Others worried that responding to survivors in such individually tailored ways ran the risk of privileging some survivors over others. The following quotation nicely summarizes both of these concerns:

And I was thinking about the woman who went to church, if she had no ride or if she felt uncomfortable going by herself the first time but once she got there and met a few people she would make some connections and maybe be able to get a ride to church every week or whatever. Then my inclination would be to drive her if she wanted a ride and go with her and sit with her if she really felt that was the only way she would go. But I think there’s a tendency among some staff to stick to, you know, we have to do everything the same with every person you can’t, because if we give this one a ride we have to do it with everyone and that’s not the way we’ve done it in the past and do you know what I’m saying?

But sometimes, advocates were willing to rethink traditional boundaries in an effort to meet the individual needs of survivors. As one participant noted, “[Sometimes] you really have to step out of what you’re used to doing and just try to work with everyone on such an individual basis.” This meant stepping out of traditional physical boundaries (where advocates and survivors are in physical space in relation to each other—often in offices or even separate parts of a building) and interpersonal boundaries (the line between being a professional and being a friend to a client; Wies, 2009). Advocates who wanted to challenge traditional notions of “appropriate” boundaries talked about the importance of leaving their program offices and entering the community with survivors—to help transport them where they needed to go to find community or to accompany them as they visited members of their informal network or tried new activities for the first time. Perhaps more dramatically, some participants also talked about the importance of challenging traditional ideas about interpersonal boundaries. These participants talked about the importance of sometimes dropping the “professional mask” and being a “friend” to a survivor by becoming more transparent than in traditional helping relationships. This could involve any kind of activity—sharing a cup of tea, taking a walk—that allowed for conversations that went beyond a focus on the survivor’s “target problem” and enabled the development of a trusting relationship.

Empowerment versus dependency

Strongly related to boundary-related tensions, participants described the tension between empowering survivors versus creating dependency or “doing with” survivors versus encouraging them to do for themselves. Although this tension clearly extends beyond network-oriented work, it is particularly salient with practices that involve accompanying survivors into the community, which, some argued, required an extra level of support. As one participant put it,

I know that we have attended like first AA meetings or along those lines with clients. But we . . . we’re also working with them so they’re, they can do it themselves. That’s our goal, like they’re going to do it themselves. . . . We don’t have the time ourselves to go to all these meetings but the first one we’ll go with you but then we’ll prepare you for you to go next time by yourself.

And, in discussing how she encouraged advocates to see their work as relatively short term, another noted,

We’re a stopgap. We’re not a solution. . . . You know, as people get further away from their crisis often social resources open up for them. The friends that they didn’t know they could contact found out that they were out of this relationship and it might be safe to go back into.

In contrast, many participants felt that in the absence of any other type of connection, advocates had to do more: They had to understand that they themselves were perhaps the closest tie a survivor had and to respond to them accordingly. In either case, participants thought it was important to engage in ongoing staff conversations about empowering survivors versus “creating new dependencies.” As one participant put it, “How do you do this but not forever?”

The need for preparation and training

Another overarching challenge participants cited frequently was the dearth of training and other structural support to do network-oriented work. Most participants acknowledged that such work was extremely difficult, as the boundary discussion illustrates, requiring a high degree of self-knowledge on the part of advocates, as well as agency policies and attitudes that were conducive to this work.

Regarding self-knowledge, some participants discussed the importance of training in terms of what kinds of questions advocates should be prepared to ask themselves before engaging in network-oriented work. These included the following questions: Do I know people in the local community with whom I could connect survivors with shared interests? Do I have the social wherewithal to accompany a survivor to a new church, or a community event that involves crossing boundaries of race, class, geography, culture, sexual orientation, or religion? Do I have the internal flexibility to have an authentic relationship with a survivor as she makes tentative steps toward rekindling attenuated relationships?

Regarding policy, given that network-oriented work is so highly individualized and requires so much creative thinking and innovation, most participants expressed a strong desire for more flexible policies governing the legitimate scope of their activity. Many felt that program rules were often too restrictive for them to do the flexible and creative work required to help survivors develop relationships. They wanted policies that allowed for more “gray areas” in terms of what kind of work is possible within agency guidelines.

Given these challenges, participants were unanimous in their conclusion that network-oriented work required a great deal of care and sophistication. As noted earlier, not everyone can or should do it, and all who want to should have the resources, training, and flexibility to do so. Unfortunately, most noted a dearth of resources and training in this area, as well as overly rigid policies that did not allow for enough flexibility.

The challenge of short-term versus ongoing relationships

Finally, participants talked about the difficult challenge of how to do network-oriented work in agencies that did not encourage ongoing relationships between survivors and specific advocates. Sometimes, this was because of the nature and mission of an agency and the service provided, as is the case with a hotline that usually works with survivors on a one-time or sporadic basis. And sometimes, agencies more actively discouraged ongoing relationships given that survivors could not count on a particular advocate being available during a crisis. Although practices such as taking stock could be done in one encounter, this would be much more difficult with others. One participant summed up this dilemma this way:

I think one of the challenges we’ve been having in the main office is that when someone comes in it’s just whatever advocate happens to be there that works with them. So they may be working with a lot of different advocates. So one person may know that maybe their aunt and uncle are very supportive and have in the past, but when they come in and talk to the next person about something, they’re not going to know anything about that. So getting everybody to have that whole big picture, you know we may be missing some things on connecting people.

This tension was one that felt especially insurmountable for participants, most of whom expressed a desire to be able to build more ongoing relationships with survivors whenever possible.

Discussion

The purpose of this study was to understand how, when, why, and to what extent DV advocates use network-oriented practices and how they perceive the costs, benefits, and challenges of these practices. The results indicate that those working with survivors are highly attuned to survivors’ isolation. They understand that helping survivors move into a network of support is critical to maintaining their safety and emotional well-being; and they believe that supporting survivors to build community is a proper and indeed central responsibility of DV programs, one that reflects a core idea of the DV movement—that stopping IPV is everyone’s responsibility. Indeed, most feel that sustained change emerges only from the combined effort of survivors themselves, responsive formal networks, and activated informal networks. It is important to note that participants were not interested in asking network members to do for free what DV and other social services generally do. Instead, they recognized that peoples’ informal network members can support each other in ways that formal services simply cannot.

Specifically, participants articulated five dimensions of a network-oriented approach. First, it helps isolated survivors prepare to reenter relationships by building their internal capacities, that is, helping them repair abuse-related damage to their sense of self and responses to others. This might involve working with survivors to identify (or reidentify) their own needs, feelings, goals, and interests, as a pathway toward engaging with others, teaching them to identify safe and unsafe relationships as well as safe and unsafe interactions, and helping them regain a connection to some larger community.

Second, a network-oriented approach takes survivors through a process of taking stock to identify potentially helpful and harmful network members and to determine what kinds of support different network members might provide at different times.

Third, a network-oriented approach assists survivors to repair relationships that may have been disrupted by the abuse through strategies such as coaching, role-playing, providing emotional support, managing expectations, and creatively bending rules to enable contact between survivors and network members.

Fourth, it enables survivors to develop new ties, particularly when their own connections have become too frayed or unhelpful. Supporting the establishment of new ties involves helping participants to develop new interests, providing opportunities for them to participate in support groups that are IPV focused or focused on other key dimensions of their lives (e.g., parenting), helping them build mutual support systems, and providing ongoing emotional support as survivors try and sometimes fail to forge new friendships. These strategies seem particularly important for two specific marginalized groups at least in the state in which this study was conducted: new immigrants whose abusive partners remained embedded in the often small communities upon which they depend, and survivors of color seeking shelter in rural white communities, whose encounters with racism in these communities make them wary of reaching out.

Finally, a network-oriented approach assists network members to support the survivors in their lives by helping them understand the dynamics of IPV, educating them about services for survivors, and providing support as they reengage with survivors or discover abuse in their own lives. All participants engaged in at least some of these practices; most engaged in all of them at least some of the time.

At the same time, all participants also acknowledged that network-oriented practices brought with them a series of challenges. Perhaps the most frequently cited challenge to network-oriented work was the dearth of training and supervision to guide them. This, participants noted, was particularly problematic given the many tensions that arose in doing the work, including how to take a network-oriented approach in the context of crisis management, encourage survivors to forge helpful relationships but not harmful ones, tailor responses to meet the needs of individual survivors without crossing boundaries in potentially harmful ways, accompany survivors into the community without making them feel unable to act alone, and take a network-oriented approach in the context of short-term or even one-off relationships.

It is worth noting that the tensions that arose in focus group discussions about network-oriented practice are quite similar to those that arise repeatedly in discussions about survivor-centered advocacy more broadly (Goodman & Epstein, 2008). Indeed, network-oriented work simply extends a survivor-centered perspective into the domain of relationship development. Stated simply, survivor-centered advocacy is based on the idea that survivors—and their needs—differ on the basis of a range of factors that advocates need to understand and respond to in tailored and creative ways (Davies, Lyon, & Monti-Catania, 1998; Goodman & Epstein, 2008). There is no-one-size-fits-all approach when it comes to a truly survivor-centered practice.

The same is true for network-oriented work. Advocates using this approach often have to step out of their usual ways of doing things to be maximally helpful. This in turn requires a rethinking of traditional rules about “appropriate” physical and interpersonal boundaries, a great deal of self-knowledge, and a high level of cultural sensitivity concerning how survivors might differ in their approach to relationships and network by virtue of race, ethnicity, class, sexual orientation, ability status, and other social locations (Goodman & Epstein, 2008).

Of course, the relative paucity of training and supervision on network-oriented work must be understood within the broader context of resource limitations with which DV programs are struggling. Over the past several years, DV programs across the country have reported a loss of revenue while the demand for services has increased (Mary Kay Truth About Abuse Survey, 2011). Many participants noted that network-oriented work felt difficult in the face of limits—on their time, their role descriptions, and their ability to work with the same person over an extended period. Interestingly, however, although most participants in the current study described the terrible pressure they felt to do more and more with fewer and fewer resources, many also noted that network-oriented work is more important than ever in an era of scarce resources. Indeed, as some mentioned, network-oriented work offers one potential avenue for extending the reach of DV programs even when resources are not expanding apace. As one program participant said,

We need to reach out to survivors’ communities more than ever because as agencies are trying to do more for people with less money, it’s really survivors’ networks that are going to be supporting them—not us.

Study Limitations

Although this study provides an important first step in understanding how DV advocates practice network-oriented work, it also has several limitations. First, some of the practices described may be specific to the state in which the study was conducted. Over the last few years, the state coalition that participated in this study, like many state coalitions, has initiated a series of conversations among practitioners about how to tailor their services to the individual needs of survivors. Much of this exploration has centered on the need to modify existing shelter rules, which are difficult for many shelter residents. Ongoing conversations about how and why rules were developed, whom they serve, and how to change them to support survivors better has moved programs toward recognizing the need for more flexible approaches that respond more directly to survivors’ needs. This, in turn, has set the stage for network-oriented practices.

Second, the sample was comprised almost entirely of White women. This is a reflection of the race/ethnicity of advocates across the state and the population of the state itself. Practices developed by practitioners of color and in communities of color may look quite different. Indeed, as noted earlier, many of the grassroots efforts to engage communities on the topic of IPV have emerged from communities of color. For example, Shimtuh, a California-based effort to organize the Korean community in the Bay Area, worked to build a conversation about IPV in that community by organizing block parties, developing web campaigns linked to local churches and temples, and holding forums sponsored by Korean-language newspapers. To build a still broader conversation, Communities Against Rape and Abuse (CARA), based in Seattle, Washington, worked with the African American community, the disabled community, and young people to organize a series of popular education workshops, discussion groups, and community events where connections between IPV and other community concerns were explored (see Fullwood, 2002; U.S. Department of Health and Human Services, 2004, for excellent descriptions of these and other efforts).

Third, this study focused exclusively on the perspective of practitioners rather than the experiences of survivors themselves. Although this is appropriate given our initial aim, it is critical that future research explore survivors’ experience of each of these practices and their long-term impact and efficacy.

Implications for Research

Perhaps the chief utility of this study is that it provides a frame for continuing exploration of network-oriented practices. For example, beginning with the practices described here, researchers might develop a way to identify and measure network-oriented practices quantitatively. Armed with such a measurement tool, they could then examine the extent to which these practices are used in other states and with varied types of DV programs (e.g., hotlines, community-based organizations, shelters) and varied types of communities. Such research could in turn assist organizations to assess whether and how they want to incorporate new practices or expand existing ones. Given that survivors of color and members of other marginalized communities (e.g., immigrants, refugees, lesbian, gay, bisexual, transgender and queer communities, women with disabilities) are especially reluctant to use mainstream services (Goodman & Epstein, 2008), advocates who work mainly with these populations may have developed more creative and interesting network-oriented practices and made more headway in dealing with the tensions and challenges they evoke. It would therefore be particularly important to extend our research into these communities.

Ultimately, researchers could explore the impact of network-oriented practices on survivors and network members themselves. Which of these practices do survivors and network members find most helpful and at what stage of their process (i.e., earlier in crisis, during shelter stay, while accessing community-based support)? Such investigation would also help illuminate some of the tensions described here. For example, practitioners raised the very important question of when network-oriented practices increase empowerment and when they might unintentionally make survivors feel less able to strike out on their own. This seems a good question for survivors. Of course, as was true for this study, any of these endeavors would require ongoing collaboration between researchers and practitioners as both have critical knowledge to contribute.

Implications for Practice

As this study’s findings demonstrate, network-oriented work is complex. Programs interested in a more explicit part of their models would need to help practitioners acquire new skills, knowledge, and tools (although many might find that they are already using some of the methods made explicit here, without having framed them quite this way). Some advocates would need support to learn about survivors’ informal networks and how they might differ across cultures and social locations. Important questions also remain regarding boundaries and confidentiality.

Practitioners could contribute enormously to these issues by working to develop best practice in the area of network-oriented work and articulating a set of principles that help resolve some of its inherent tensions. They could also develop training that would enable advocates to learn about network-oriented work and strategies that address some of its key tensions.

Ultimately, for DV program staff to truly embrace a network-oriented approach, they would need to enlarge their vision of the role of the advocate to include two key forms of collaboration: partnerships with community members, with each member of the partnership bringing his or her best skills, whether professional training, experiential expertise, or both, to the task of supporting survivors (Goodman & Smyth, 2011); and collaboration with community organizations that might also be able to support survivors, including libraries, schools and universities, arts programs, churches, and other human service organizations to which survivors might turn to find people with shared interests or identities. Far from diminishing services’ role, this approach would allow for the application of relevant expertise and resources more precisely, based on the particular constellation and availability of support and knowledge within a given survivor’s formal and informal network.

Ultimately, this study makes clear that advocates find network-oriented work relevant and useful. As scholarship indicates and participants observed, no matter how we design our programs and practices, survivors seek help from those they know long before and long after their involvement in formal systems. Leveraging the potential of informal social networks could therefore strengthen our advocacy models, build more informed and safer communities, and expand survivors’ opportunities for safety and well-being, perhaps especially among marginalized survivors. As noted in a recent report of the Ms. Foundation (Das Gupta, 2003), “A victim and her abuser live in the community and this is where the security of women inevitably rests” (p. 19).

Footnotes

Acknowledgements

We are enormously grateful to the advocates who participated in this project. Their thoughtful and creative work with intimate partner violence (IPV) survivors is a source of ongoing inspiration. We also thank Linda Douglas and Karin Ashton for organizing the focus groups at the heart of this study.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Lisa A. Goodman is a professor in the Department of Counseling and Applied Developmental Psychology at Boston College. She has authored over 100 publications on the interaction between intimate partner violence survivors and the practices, systems and policies that aim to help them and consults extensively with local and national domestic violence organizations. She is the co-founder of the Domestic Violence Program Evaluation and Research Collaborative and co-author of the book Listening to Battered Women: A Survivor-Centered Approach to Advocacy, Mental Health, and Justice.

Victoria Banyard, PhD, is a professor in the Department of Psychology at the University of New Hampshire and a research and evaluation consultant with Prevention Innovations. She conducts research on the long-term mental health consequences of interpersonal violence including resilience in survivors. She also conducts research on community approaches to prevention of interpersonal violence. She is part of a research team that has received funding from the NIJ, DOJ and CDC.

Julie Woulfe is a sixth year doctoral candidate at Boston College. Her research interests include health disparities of lesbian, gay, bisexual, transgender and queer (LGBTQ) individuals and the intersection of interpersonal and structural violence.

Sarah Ash is a fifth year doctoral candidate at Boston College. Her current research focuses on understanding the well-being of youth with disabilities and their families.

Grace Mattern has been involved in the movement to end violence against women for over 35 years, serving as executive director of the NH Coalition Against Domestic and Sexual Violence and participating on statewide and national boards. She has published two volumes of poetry and her poetry and prose have been published widely in journals and anthologies. She blogs at .

References

Adkins

K. S.

Kamp Dush

C. M.

(2010). The mental health of mothers in and after violent and controlling unions. Social Science Research, 39, 925-937.

Beeble

M. L.

Bybee

Sullivan

C. M.

Adams

A. E.

(2009). Main, mediating, and moderating effects of social support on the well-being of survivors of intimate partner violence across two years. Journal of Consulting and Clinical Psychology, 77, 718-729.

Black

M. C.

Basile

K. C.

Breiding

M. J.

Smith

S. G.

Walters

M. L.

Merrick

M. T.

. (2011). The National Intimate Partner and Sexual Violence Survey (NISVS): 2010 summary report. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention.

Bybee

Sullivan

C. M.

(2005). Predicting re-victimization of battered women 3 years after exiting a shelter program [Special Issue: Theoretical and Methodological Innovations in Research on Intimate Partner Violence]. American Journal of Community Psychology, 36, 85-96.

Charmaz

(2004). Premises, principles, and practices in qualitative research. Qualitative Health Research, 14, 976-993.

Coker

A. L.

Smith

P. H.

Bethea

King

M. R.

McKeown

R. E.

(2000). Physical health consequences of physical and psychological intimate partner violence. Archives of Family Medicine, 9, 451-457.

Coker

A. L.

Watkins

K. W.

Smith

P. H.

Brandt

H. M.

(2003). Social support reduces the impact of partner violence on health: Application of structural equation models. Prevention Medicine, 37, 259-267.

Das Gupta

(2003). Safety and justice for all: Examining the relationship between the women’s anti-violence movement and the criminal justice system. New York: Ms. Foundation. Retrieved from http://files.praxisinternational.org/safety_justice.pdf

Davies

J. M.

Lyon

Monti-Catania

(1998). Safety planning with battered women: Complex lives/difficult choices. Thousand Oaks, CA: Sage.

10.

Du Mont

Forte

Cohen

M. M.

Hyman

Romans

(2005). Changing help-seeking rates for IPV in Canada. Women & Health, 41, 1-19.

11.

Dunham

Senn

C. Y.

(2000). Minimizing negative experiences: Women’s disclosure of partner abuse. Journal of Interpersonal Violence, 15, 251-261.

12.

Fleury-Steiner

R. E.

Bybee

D. I.

Sullivan

C. M.

Belknap

Melton

H. C.

(2006). Contextual factors impacting battered women’s intentions to re-use the criminal legal system. Journal of Community Psychology, 34, 327-342.

13.

Fullwood

C. P.

(2002). Preventing family violence: Community engagement makes the difference. San Francisco, CA: Family Violence Prevention Fund.

14.

Goodman

L. A.

Dutton

M. A.

Vankos

Weinfurt

(2005). Women’s resources and use of strategies as risk and protective factors for reabuse over time. Violence Against Women, 11, 311-336.

15.

Goodman

L. A.

Dutton

M. A.

Weinfurt

Cook

(2003). The need for continued reform: The broad scope and deep impact of intimate partner violence. Violence Against Women, 9, 163-186.

16.

Goodman

L. A.

Epstein

(2008). Listening to battered women: A survivor-centered approach to advocacy, mental health, and justice. Washington, DC: American Psychological Association.

17.

Goodman

L. A.

Smyth

K. F.

(2011). A call for a social network-oriented approach to services for survivors of intimate partner violence. Psychology of Violence, 1, 79-92.

18.

Graneheim

U. H.

Lundman

(2004). Qualitative content analysis in nursing research: Concepts, procedures and measures to achieve trustworthiness. Nurse Education Today, 24, 105-112.

19.

Kaslow

N. J.

Thompson

M. P.

Brooks

A. E.

Twomey

H. B.

(2000). Ratings of family functioning of suicidal and non-suicidal African American women. Journal of Family Psychology, 14, 585-599.

20.

Kim

(2002). Innovative strategies to address domestic violence in Asian and Pacific Islander communities: Examining themes, models, and interventions. San Francisco, CA: Asian and Pacific Islander Institute on Domestic Violence. Retrieved from http://www.apiidv.org/files/Innovative.Strategies.Rpt-APIIDV-2002(Rev.2010).pdf

21.

Levendosky

A. A.

Bogat

G. A.

Theran

S. A.

Trotter

J. S.

Von Eye

Davidson

W. S.

(2004). The social networks of women experiencing domestic violence. American Journal of Community Psychology, 34, 95-109.

22.

Lincoln

Y. S.

Guba

E. G.

(1985). Naturalistic inquiry. Newbury Park, CA: Sage.

23.

Mancini

Nelson

Bowen

Martin

(2006). Preventing intimate partner violence: A community capacity approach. Journal of Aggression, Maltreatment & Trauma, 13, 203-227.

24.

Mary Kay Truth About Abuse Survey. (2011). Retrieved from http://www.ncdsv.org/images/marykay_truthaboutabusesurvey_2011.pdf

25.

Massey

O. T.

(2010). A proposed model for the analysis and interpretation of focus groups in evaluation research. Evaluation Program Planning, 34, 21-28.

26.

Riger

Raja

Camacho

(2002). The radiating impact of intimate partner violence in women’s lives. Journal of Interpersonal Violence, 17, 184-205.

27.

Rose

L. E.

Campbell

Kub

(2000). The role of social support and family relationships in women’s responses to battering. Health Care for Women International, 21, 27-39.

28.

Thompson

M. P.

Kaslow

N. J.

Kingree

J. B.

Rashid

Puett

Jacobs

Matthews

(2000). Partner violence, social support, and distress among inner-city African American women. American Journal of Community Psychology, 28, 127-143.

29.

U.S. Department of Health and Human Services. (2004). National consensus guidelines on identifying and responding to domestic violence victimization in health care settings. San Francisco, CA: Family Violence Prevention Fund.

30.

Wies

J. R.

(2009). Boundaries in carework: A case study of domestic violence shelter advocates in the USA. Global Public Health, 4, 464-476.

31.

Wilkinson

(2003). Focus groups. In Smith

J. A.

(Ed.), Qualitative psychology: A practical guide to research methods (pp. 184-204). London, England: Sage.