Content uploaded by Jason Whiting
Author content
All content in this area was uploaded by Jason Whiting on May 20, 2015
Content may be subject to copyright.
ORIGINAL ARTICLE
Challenges and Retention of Domestic Violence Shelter
Advocates: a Grounded Theory
Lisa Vallie Merchant &Jason B. Whiting
Published online: 11 March 2015
#Springer Science+Business Media New York 2015
Abstract Using grounded theory methods, this study exam-
ines the experience of shelter advocates and the relationship
between the challenges of advocacy, shelter culture, and re-
tention. Challenges fell into three categories: managing shelter
shock, letting go of being the hero, and balancing advocate
roles. Sub-challenges included hearing client stories, manag-
ing crisis, accessing resources, accepting clients going back to
abusive situations, facilitating empowerment, and enforcing
rules. Shelter culture strongly influenced advocates’adjust-
ment. Advocates with supportive cultures expressed less frus-
tration and were more likely to continue employment, while
those with less-supportive cultures expressed more frustration
and were more likely to leave the domestic violence field or
promote within the field to create macro-level change. Impli-
cations for shelters and future directions for research are
included.
Keywords Advocacy .Domestic violence .Employee
retention .Organizational culture .Shelters .Turnover
The National Intimate Partner and Sexual Violence Survey
(Black et al. 2011) estimates that 9 million women and men
are slapped, pushed, or shoved by an intimate partner annual-
ly. Intimate partner violence, also known as domestic violence
(DV), may affect a victim’s physical and mental health, fi-
nances, employment, housing, and relationships with family
and friends (Barnett et al. 2011). It may also result in death
(Catalano et al. 2009). Because DV shelters address these
consequences and provide safety, they are an important
avenue for victims seeking relief. According to the National
Network to End Domestic Violence (NNEDV 2011), 37,519
victims of abuse were residing in DV shelters in the United
States in a single day in 2010, with another 5686 victims being
turned away due to staff and funding shortages. An additional
33,129 adults and children received non-residential assistance,
such as counseling, legal services, and support groups
(NNEDV 2011). Because the services provided by shelters
are critical for many victims, it is essential that these programs
provide effective assistance.
Although there is abundant research on DV interventions,
very little of that research has found its way into the hands of
front line advocates (Guterman 2004; Murray et al. 2010).
When research does make it to these advocates, many have
complained about its lack of practical applications (Murray
and Welch 2010). Moreover, participants in the National Vio-
lence Against Women Prevention Research Center (2001)sur-
vey noted that some topics have been Bdone to death^(p. 5)
and that is it Btime to break new ground^(p. 5). As such,
several researchers have argued that research must be more
collaborative, with scholars spending more time Bin the
trenches^(Murray and Welch 2010, p. 2290), working direct-
ly with victims or shadowing advocates to better understand
the needs of practitioners. It is this proposition that is at the
center of this study. By exploring the experiences of front line
workers, others can better appreciate the day-to-day experi-
ence of advocates. In that sense, this study provides a voice to
DV shelter workers and serves an advocacy role for these
advocates.
Background
As is the nature of qualitative research, this project developed
as it was being carried out. It began with a general question
about advocates, but as interviews were conducted and
L. V. Merchant (*):J. B. Whiting
Department of Community, Family, and Addiction Services, Texas
Tech University, Human Sciences Building, Room 271, 1301 Akron
Avenue, Lubbock, TX 79409-1210, USA
e-mail: lisa.merchant@ttu.edu
J Fam Viol (2015) 30:467–478
DOI 10.1007/s10896-015-9685-y
analyzed, the focus shifted as concepts were defined and de-
veloped (Corbin and Strauss 2008). Although the heart of the
study—to understand the experience of DV advocates—
remained unchanged, we began to wonder about the chal-
lenges of advocacy and how advocates resolve those chal-
lenges. Because several participants had left or were contem-
plating leaving their advocacy positions, we were also curious
as to how challenges and resolutions to those challenges af-
fected the advocates’leave/stay decision.
Therefore, this study was further designed to shed light on
the experiences of DV advocates so those with the power to
effect change, such as supervisors and directors, can design
programs that improve advocate retention and client outcomes.
Although data specific to DV shelters was unavailable, turnover
rates as high as 50 % annually have been reported in human
service organizations (Massachusetts Council of Human Ser-
vice Providers [MCHSP 2007). Anecdotally, DV shelter direc-
tors have reported the average life of a shelter advocate to be
two to three years (Behounek 2011;Consulting 2012).
Recruiting and training new employees and paying overtime
for employees who must pick up the slack left by the vacancy
have been estimated to cost between 30–150 % of the vacancy’s
salary, which is a substantial loss for DV shelters, who are
funded primarily through grants and donations (Selden 2010).
Vacancies also heighten employee stress and frustration though
increased caseloads, which lead to dissatisfaction, burnout, and
more vacancies (MCHSP 2007). Understanding and rectifying
turnover in DV advocates is essential for providing quality ser-
vices, as heavy caseloads and disruptions in care caused by
turnover can result in unmet client needs (MCHSP 2007;
NNEDV 2011; USGAO 2003).
Despite the pervasiveness of turnover and the importance
of retention, investigations into these topics have been few.
However, a survey by the National Association of Social
Workers (N ASW 2006), found that 26 % of social workers
were planning to leave their current position within the next
two years, and 4.6 % were planning to leave the social work
profession altogether. Social workers planning to leave per-
manently were more likely to have less education, earn a
smaller salary, work with uninsured clients, complete tasks
below their training, have access to fewer resources, and have
less on-the-job training. These workers also reported less re-
spect for social work services within the organization, less
guidance from supervisors, and less satisfaction with time to
complete their work. Further, vacancies and safety issues were
more common in their workplaces. These workers said they
would changepositions for a higher salary, increased mobility,
improved supervision, and less stress (NASW 2006). Al-
though not all DV advocates are licensed social workers or
have degrees in social work, being that they often fill the role
of a social worker, it would seem likely that advocates expe-
rience similar frustrations with comparable results. This is
especially probable given that many of the challenges listed
above correspond to the challenges described by DV
advocates.
Although the adversities of DV shelter advocacy have not
been researched directly, studies on related topics, like burn-
out or boundaries, provide illumination. Based on these stud-
ies, difficulties fall into three categories: societal, workplace,
and client challenges. Societal challenges include lack of
funding, prevailing social attitudes that shame or minimize
victimization, and working with outside agencies (Bemiller
and Williams 2011;UllmanandTownsend2007). Workplace
challenges include the professionalization of services, racism,
rigid hierarchies with insufficient supervision, reduced com-
pensation and fringe benefits, insufficient time to complete
tasks, co-worker stress, restrictive policies and procedures,
on-call requirements, inadequate number of staff, division be-
tween staff and supervisors, poor communication, indistinct
goals, and lack of safety (Baker et al. 2007;Behounek2011;
Bemiller and Williams 2011;BrownandO’Brien 1998;
Haley-Locke 2007;Schow2006; Ullman and Townsend
2007). Client challenges include the difficulty of working with
child victims, maintaining professional boundaries without
sacrificing physical and interpersonal connections, enduring
the crisis-orientation of shelter work, accepting the victim’s
decision to return to his or her abuser, and hearing client
stories (Behounek 2011; Dunn and Powell-Williams 2007;
Wies 2009; Ullman and Townsend 2007). While the descrip-
tion in many of the previous studies was quite rich and
insightful, only Behounek (2011) connected advocacy chal-
lenges to advocate retention. Advocates in her study cited on-
call requirements and lack of supervisor support as reasons for
turnover. Thus, there were no studies that focused broadly on
understanding the day-to-day experiences of domestic vio-
lence advocates and how those experiences shape outcomes.
By examining the daily life of DV shelter advocates, we can
better understand the factors influencing turnover and design
programs to improve satisfaction, retention, and, ultimately,
client outcomes.
The purpose ofthis study, then, was to develop a model that
represented the experiences of domestic violence shelter ad-
vocates in their own words. Specifically, we wanted to under-
stand the process that advocates worked through as they faced
challenges in their roles with clients. We also wondered what
factors mitigated those challenges and how these challenges
influenced retention.
Method
We used constructivist grounded theory methods, as elucidat-
ed by Charmaz (2006), to develop a model of the challenges
faced by advocates. In grounded theory, a model is derived
from the data to explain a process, action, or interaction
(Crewell 2007); in a constructivist grounded theory, the data
468 J Fam Viol (2015) 30:467–478
and analysis are both assumed to be social constructions as
opposed to objective facts. As such, constructivist grounded
theory encourages researchers to be reflexive about their in-
fluence over the data and analysis and to recognize that all
research is interpretive. This does not mean, however, that the
process is not rigorous or representative.
There were many steps taken in the research journey that
were designed to ensure trustworthiness of the process (Lincoln
and Guba 1985). The data were gathered and analyzed
according to established grounded theory methods as
described by Corbin and Strauss (2008)andCharmaz(2006),
which delineate a deep engagement with the data while devel-
oping and fleshing out the primary categories. The first author
of this study conducted the interviews and primary analysis,
while the second author, in the role of a Bpeer debriefer^
(Crewell 2007, p. 208), reviewed the process and category de-
velopment. Additionally, extensive memos were written during
the research. This was not only key to the analytic process, but
was important for the reflexivity of the researchers. The integ-
rity of a grounded theory depends, in part, on the ability of the
researchers to recognize how their own experiences shape the
research process (Allen 2000). For the first author, this was
especially important, as she is a former DV shelter advocate.
To minimize the influence of her experience, she used open-
ended questions, wrote memos about her reaction to the data,
conferred with her co-author, and bracketed her experience as
much as possible. During interviews, she also took a Bnot
knowing^stance, meaning she asked questions as if she were
naïve so as not to make assumptions about participants’expe-
riences. This is not to say, though, that she interviewed with an
Bempty head^(Charmaz 2006, p.48). Rather, she drew upon
her experience to follow leads and clarify meaning. Likewise,
she utilized her experience during analysis to rework emerging
categories and relationships until they Bfelt right^(Corbin and
Strauss 2008,p.47).
Participants
Participants were recruited via several avenues: (a) emails sent
to shelter directors, (b) emails forwarded from listserv an-
nouncements, and (c) snowballing techniques. Altogether,
nineteen current and former shelter workers participated from
nine shelters in the southwestern United States. Seventeen
participants were female and two were male. Seven reported
their ethnicity as Caucasian, seven as Hispanic, and five as
African-American. Participants ranged in age from 23 to 61,
though most were in their late 30s or early 40s. Four had
master’s degrees, two had taken some graduate courses, seven
had bachelor’sdegrees,twohadassociate’s degrees, two had
attended some college, one had a high school diploma, and
one had less than a high school diploma. Six participants had
professional licenses (e.g., Licensed Clinical Social Worker,
Licensed Marriage and Family Therapist). Length of time
working in a shelter ranged from 1 to 12 years, with an aver-
age of 5.4 years of experience. Fifteen of those who responded
were currently working in shelters, while three were previous
employees who had left shelter employment five to six years
prior to this study after three, six, and 10years of employment,
respectively. Three current advocates had previously worked
in another shelter. Positions of participants included two child
care workers, one legal advocate, eight case mangers/advo-
cates, four advocate supervisors, one hotline advocate, one
therapist, and one mixed-position advocate.
The nine shelters ranged in size from 18 to 120 beds, with
six shelters falling between 30 and 70beds. Operating budgets
and number of employees were only available for five shel-
ters. The smallest budget reported was $400,000, while the
largest was $7 million, with the number of employees ranging
from 7 to 120. Four shelters described their location as urban,
three as rural, and two as mixed. The diverse range of partic-
ipants and shelters included in this study was helpful, as the
interviews reflected a variety of experiences that were impor-
tant in getting the final model to a place that felt saturated and
representative of the advocacy process.
Procedure
The first author conducted individual, face-to-face, semi-
structured interviews with participants. The two grand-tour
interview questions were: (a) BTell me about a typical day,^
and (b) BWhat is it like to work in a shelter?^Although
follow-up questions were based on participant responses, the
researcher asked the same follow-up questions across many
interviews, including: What are you doing at work when you
feel at your best? What are you doing at work when you feel at
your worst? How successful is the shelter in meeting its pur-
pose statement? On a scale of 1–10, how satisfied are you at
the shelter? What makes it a [number] instead of [a lower
number]? What would need to be different to make it a [higher
number]? What would help you do your job better or help the
shelter to be more successful? Are there things you wish you
knew or understood that would help you in your work? Ques-
tions were primarily open-ended, and part of the not-knowing
stance included asking questions about taken-for-granted
meanings. For example, when one advocate mentioned em-
powerment, the interviewer asked, BWhat do you mean by
empowerment? How do you empower people?^Interviews
were gathered before the main analysis could take place,
which limited our ability for constant comparative analysis.
Analysis
Trained, paid, undergraduate assistants transcribed the inter-
views. Priorto coding, interviews were re-listened to and read.
The comment function in Microsoft Word was used for initial,
line-by-line coding. Consistent with recommendations by
J Fam Viol (2015) 30:467–478 469
Charmaz (2006), coding was initially done with gerunds in
order to focus on what was happening in the line. During
coding, mini-memos were written in the margins, and longer
memos were written or typed. As themes became apparent in
the data, focused coding commenced. Using QSR
International's NVivo 8 qualitative data analysis software,
segments of data were placed into categories. To specify the
dimensions of categories and link categories together, which
Charmaz (2006) calls axial and theoretical coding, categories
and their data were printed, highlighted, and hand-sorted.
Throughout this process, memos and diagrams were kept in
ajournal.
Results
Figure 1displays the challenges of domestic violence shelter
advocacy as described by participants, how advocates man-
aged those challenges, and how management of challenges
affected the advocate’s decision to leave shelter employment.
Challenges fell into three broad categories: managing shelter
shock, letting go of being the hero, and balancing advocate
roles. Each category included sub-challenges, such as hearing
client stories, managing crisis, accessing resources, accepting
clients going back to abusive situations, facilitating empower-
ment, and enforcing rules. Strongly influencing advocates’
adjustment to these adversities was the shelter culture, which
was either supportive or demoralizing. Advocates with a sup-
portive shelter culture expressed less frustration with the chal-
lenges and were more likely to continue their shelter employ-
ment. Those in less-supportive shelters expressed more frus-
tration and were more likely to leave the domestic violence
field or seek to promote within the field in order to create
macro-level change.
Getting Hired
There seemed tobe two primary types of people applying for a
position in a shelter: people with and people without a pre-
established passion for domestic violence victims. Those with
a passion cited Bwanting to be a hero^or Bwanting to help
others as I had been helped^as instrumental in the initial
application. These individuals either had prior experience with
domestic violence or wanted specifically to help victims of
violence. Those without a passion for working with victims
of DV cited practical reasons for applying, such as Ba
paycheck^or experience. This is not to say that those without
a passion were cold or indifferent, as many had degrees in
helping professions (e.g., social work or counseling). Regard-
less of how advocates came into the field, they faced similar
challenges.
Advocacy Challenges
Managing Shelter Shock Participants described the first few
months of shelter life as Beye opening,^both personally and
professionally. Said one advocate about employees who quit:
It’s a hard transition, because maybe they’re used to
doing something specific, or they think it’s going to be
easy, that, ‘I’m going to work my hours, 7:00 to 3:00,
and I’m going to have hour sessions, and I’m going to
do this for clients, and it’s going be formatted, and the
victim is going to say thank you very much’and in
reality it’s not that.
Two of the earliest challenges for advocates seemed to be
coping with the emotional intensity of client’s stories and
managing the day-to-day chaos of shelter life.
Getting Hired
Supportive
Cultures
Demoralizing
Cultures
Settle In
Promote
Quit
Intensity of DV
Crisis & chaos
Managing Shelter Shock
Scarcity of resources
Not ready for change
Letting go of Being the Hero
Enforcing rules
Enabling vs. empowering
Bootstrap, enabling, & balanced
shelters
Balancing Advocate Roles
Advocacy Challenges Shelter Culture and Retention
Fig. 1. Model of challenges of domestic violence (DV) advocacy and their relationship to shelter culture and retention. Solid lines indicate the most
common path, while dashed lines indicate less common paths
470 J Fam Viol (2015) 30:467–478
Coping with the Emotional Intensity of DV Advocates, even
those with training or a history of DV, were often shocked by
the severity of violence. As one said, BI was flabbergasted. I’m
like, ‘Oh my god. I didn’t know people did this; that men did
this to ladies.’^Another recounted her first client: BI’ll never
forget... she was a white lady, but I always refer to her as
colored because she was covered in bruises. I had never seen
that before.^Others were shocked by how victims
downplayed the abuse. One advocate related the story of a
victim with a broken nose who had been in the shelter previ-
ously with a broken jaw: BI’ll never forget the comment she
made, ‘You know how it is. We’re Mexican and we just take
it.’^For some, the shock of abuse was compounded by rec-
ognizing DV in their own histories or by re-opening wounds
from past abusive relationships. Professionally, advocates
sometimes realized how unprepared they were for dealing
with trauma and crisis work.
Regardless of training or experience, one participant noted
that Badvocates with a heart are going to feel something,^and,
therefore, must learn how to cope with those feelings. In the
presence of clients, advocates reported aiming for profession-
alism, which they defined as listening and validating without
showing emotions. Afterward, though, advocates turned to
co-workers or supervisors for catharsis, though some coped
privately. Receiving support from co-workers helped advo-
cates feel less alone, while those who coped privately felt
more isolated. Even though the hardship of hearing these
stories never fully abated, most said that, over time, they ac-
climated. However, as one seasoned advocate noted, BThose
stories never leave your head, and even though you aren’t
supposed to, you take them home with you.^
On a more positive note, witnessing the severity of DV
endowed some advocates with passion, compassion, and grat-
itude. As one stated, BOnce my eyes were opened to it, I
couldn’t walk away. There’s no leaving. There’s no forgetting
about it.^Increased understanding and compassion for vic-
tims, family, and friends who stayed in violent relationships
was common. Others spoke about learning to be grateful, see-
ing struggles from a much broader perspective, and educating
family and friends about red flags of abuse.
Managing Crisis and Chaos As communal crisis centers, sev-
eral advocates lamented coming to work and never knowing
what to expect. Although the advocates’official duties gener-
ally included mundane activities, such as transporting clients
and providing referrals, advocates also often found themselves
snaking toilets, cleaning up vomit, mediating residents’quar-
rels, calming fussy babies, and otherwise ensuring the smooth
operation of a communal living facility. This was in addition
to their duties as a crisis responder: providing a compassioned
response to victims with stories of horrific abuse, retrieving
victims from their homes with the police in tow, and being on-
call to admit families to the shelter 24-h a day. Very few of
these duties are predictable, meaning that shelter life can be
chaotic, stressful, and frustrating, and that advocates must
wear many hats. Referring to her time in a rural shelter, one
advocate recalled, BI did everything. I was the kitchen staff. I
was security. I was residential advocate. I was police advo-
cate.... I was everything because there was no one else.^An-
other reported,
You’re overwhelmed because you have the crisis hotline
calling, you got a client that needs your help right then,
you got a client that needs meds, you got a client that
needs transportation because ‘I’ve got a job I gotta get
to.’... It does get chaotic.
A third advocate elaborated on the stresses of communal
living:
[S]haring different personalities, different mental ill-
nesses, homelessness... You’re having complaints about
snoring, complaints about the kids of another family...
you’re having to put singles who are not used to children
with the children... [it is] real difficult.
Surviving this sometimes chaotic, crisis-laden environment
requires flexibility and clear thinking. Established protocols
and colleagues willing to step into the fray aided advocates’
ability to withstand the pressures of a crisis. These advocates
were better able to step back from the chaos and move from
crisis mode to problem-solving mode, whereas advocates
reporting less accommodating supervisors and co-workers
seemed more exhausted. Some advocates with clear protocols
and involved colleagues grew from being terrified of crises
when hired to embracing crises as opportunities to be the hero.
Over time, these advocates became confident in their ability to
manage stressful situations and assured that fellow employees
would step in as needed. Advocates in less supportive envi-
ronments, however, continued to feel overwhelmed by crises,
even after years of employment. These advocates did not
doubt their ability to manage the chaos, but resented having
to manage it alone.
Letting go of Being the Hero Although not all advocates had
expectations of being a hero, many came with the desire to
help victims escape abuse. As such, advocates were surprised
to find: (a) how few resources there were and how difficult it
was to connect people to the resources needed to make their
escape; and (b) not all victims were ready to change, and many
returned to their abusers. This led some advocates to feel in-
effective. As one participant stated: BIt’s a difficult population
to see a lot of success in, so it’s kinda like, ‘Oh my goodness.
Am I good at this or not?’It weighs on you.^Another added,
BWhen your job is to help clients and I don’t have any clients
that want my help, well, I must suck.^Part of the adjustment
for advocates was to give up expectations of rescuing victims.
J Fam Viol (2015) 30:467–478 471
Scarcity of Resources BWe work with victims of intimate part-
ner violence,^said one advocate, Bbut you know what? The
problem is not just the violence. The problem is a lack of
education, poor social skills, poor self-image, and no
finances.^Therefore, many advocates saw connecting clients
to resources as a key function of their advocacy. Unfortunate-
ly, resources were often inaccessible because they did not
exist, were limited in availability, or Bred tape^prevented
access. This led to feelings of frustration and powerlessness
as expressed in the following statement:
I feel at my worst when there just isn’t a resource there
for a client. I can’t make it work in the way that it needs
to. The system is failing you, but I feel like I am failing
you.
Limited shelter stays and concerns that victims will go back
to their abuser amplified the pressure advocates felt to connect
clients to resources. Additionally, clients also put pressure on
advocates to secure resources. One advocate shared, BThey
would come in and want you to rescue them immediately.
They’dsay,‘Tell me what to do’and we couldn’t do that.
And we’d get blamed when things didn’t work out right.^
Another added, BWhen something happens and you’re not
able to get them housing, they are like, ‘Why did you lead
me on? Why did you make me think I was getting into a
housing program?’^As such, advocates tried to temper their
referrals with cautious optimism, which could be a balancing
act. On one hand, they did not want to offer assurances when
the situation was uncertain, while on the other hand, they did
not want to quash hope.
In most shelters, advocates coped with the frustration of
scarce resources by brainstorming with colleagues. On a prac-
tical level, these sessions generated new ideas, solutions, and
referrals. Just as important, however, brainstorming helped
advocates feel connected to and supported by colleagues and
more effective in their work. Unfortunately, some advocates
reported their supervisors Bmade you feel bad for not
knowing^or were unavailable for brainstorming. These advo-
cates also described increased feelings of isolation, failure,
and burnout.
Not Ready for Change One goal of advocacy is helping cli-
ents to be successfully independent. However, as mentioned,
this may not be the victim’sgoal—he or she may not be ready
to be Brescued.^Instead the victim may return to the abusive
situation or enter quickly into another relationship. Advocates
described these two events as Bheartbreaking^and
Bfrustrating,^and supervisors attested that these feelings led
to exhaustion, apathy, and cynicism. This assertion was con-
firmed in this sample, as the more exhausted and cynical ad-
vocates expressed increased frustration and less understanding
of victims who go back to an abusive relationship; they were
also more likely to hold a hero mentality and to view a victim
returning to an abusive relationship as a their own personal
failure.
However, some advocates expressed that they were at
peace when victims returned to abusive situations. These ad-
vocates recognized that going back was not a reflection of
their advocacy skills but rather the victim’s readiness to leave
the relationship. This endowed these advocates with compas-
sion for the victim’s plight and hope that the victim will even-
tually escape the abusive relationship. These advocates fo-
cused on educating victims about abuse and hoped for a Blight
bulb moment.^When clients announced plans to return, these
advocates would express their concerns, plan for safety, ex-
plore pros and cons, and emphasize that the shelter is an on-
going resource.
Balancing Advocate Roles If an advocate’s role is not the
hero, then what is it? While some advocates described them-
selves as encouragers, cheerleaders, or listeners, almost every
advocate identified his or herself as an empowerer. However,
adopting the latter identity sometimes proved difficult. Some
advocates struggled with the boundary between empower-
ment and enabling. Others suffered when their shelter’svision
of empowerment clashed with their own. Moreover, when
advocates stepped into the role of rule enforcer, residents did
not always view advocates as empowerers, but instead, as
abusers. In all these cases, advocates had to balance the de-
mands of competing roles—roles dictated by clients, their
shelter, and their own convictions. Naturally, advocates strug-
gled less when the demands of these roles harmonized.
Enforcing Rules All shelters had rules, which were necessary
for safety and order. However, advocates were often tasked
with enforcing rules, which created role confusion. On one
hand, advocates recognized the importance of rules, but en-
forcement sometimes felt awkward and abusive. BIt is funny
how [enforcing rules] can feel eerily like the situation they just
came out of,^said one advocate. This was especially true
when rules were rigidly enforced. In some cases, strict adher-
ence to rules led to an Bus versus them^attitude in which,
according to one advocate, clients spread rumors like Bthey’re
kicking everyone out^or Bthey’re out to get us.^Another
advocate noted that sometimes it felt like other workers would
Bpull together against a victim^to have him or her evicted
from the shelter for rule violations. As such, strict rule en-
forcement led to division between advocates and residents
and between advocates and other workers, including supervi-
sors. BThe worst is seeing people leave when you wish they
had been given another chance,^shared one advocate, adding,
BI understand you have to follow rules and that there are
consequences, but I felt like we were just pushing them back
out there. Advocates in these shelters were notably less satis-
fied and more frustrated.
472 J Fam Viol (2015) 30:467–478
Conversely, when rules were flexible, hierarchy was flat-
tened, and the client-advocate relationship was emphasized,
advocates struggled less with enforcing rules. For example,
one supervisor reported, BWe have a really good relationship
with the clients.... I’m in the shelter with them all day. So, if
you’re having a bad day, maybe you need to come run errands
with me.^When it came to enforcing rules, she further stated:
I think it’s a lot of how I talk to them and they know
where they stand with me at all times. I’massertive,but
I’m nice about it. I think it’s just the way you treat
them.... Clients tell me that they’ve been to shelters that
are just as bad if not worse than [their abuse]. I don’t
ever want clients to feel that way here, so we do a lot of
mediation and talking it out. But, I also tell them, ‘If you
aren’t going to be able to live cooperatively, then this
may not be the place for you.’
In this shelter, and others like it, rules were bendable and
based on individual group needs. In other words, chores, cur-
fews, wake-up times, and the like were somewhat dependent
on residents’preferences, schedules, and abilities. In one shel-
ter, residents assigned chores and decided bedtimes as a group
with shelter staff overseeing the process. However, having
flexible rules does not equate to lax enforcement or few con-
sequences. Advocates still enforced rules, though they seemed
to struggle with it less. As another supervisor related,
I’m really flexible. It depends on the [residents] as to
how tight our rules are... I tell them, ‘Yo u’re an adult.
I’m going to treat you like one. If you can’thandleit,
then you’ll have to find somewhere else...’
Most shelters functioned somewhere between ridged ad-
herence to rules and complete flexibility. As one supervisor
stated, shelters must grapple to create a space Btough enough
to create safety, but sensitive enough to understand the client’s
perspective.^Mid-sized shelters seemed to struggle most in
creating this balance, often erring by either being too rigid or
too lax in enforcing rules. Some larger shelters excused them-
selves from governing residents’schedules by building single-
family cottages instead of communal dorms. Families in cot-
tages decided their own curfews, chores, meals times, and the
like. Rural shelters with fewer beds were most able to adopt
flexible rule policies.
Enabling Versus Empowering Advocates defined empower-
ment as facilitating access to external and internal resources.
External empowerment was described as helping with tangi-
ble goals, like getting a job, finding a house, or getting a GED,
while internal empowerment was focused on building confi-
dence, identifying strengths, and reducing isolation. Many
advocates struggled with their role in empowerment. As one
noted, BYou have to balance between empowering and
enabling. There is a fine line. You can’t cookie cut it.^It
may be helpful to conceptualize empowerment on a continu-
um. On one end, there is doing too much for clients and on the
other, doing too little. Advocates struggled with empower-
ment on both a personal and organizational level.
On a personal level, no advocates believed they did too
little for their clients, though some did believe they did too
much, which they referred to as enabling. The challenge of
balancing empowerment and enabling stemmed from two
concerns. Some advocates equated enabling with client ma-
nipulation, as seen here: BThere was always some excuse why
they couldn’t do something. They were just kind of wanting
you to do for them and if you did, then they would take
advantage of that and then you would be exhausted.^An
advocate supervisor called this Bworking harder than the cli-
ent,^and noted that it hastens burnout. Of course, not all
advocates agreed that making excuses was a sign of manipu-
lation, but instead saw slow progress as a consequence of
trauma. Additionally, advocates expressed concern about the
disabling effects of enabling, noting that it discourages inde-
pendence and internal empowerment. As one advocate stated,
BWe’re not here to enable them. We like for them to do the
work. It helps with their self-esteem so much... and I don’t
ever want them to think I did [the work] for them.^
Bootstrap, Enabling, and Balanced Shelters On an organiza-
tional level, advocates struggled with empowerment when
their shelters fell on either end of the empowerment continu-
um. Some struggled when their shelters adopted a Bpull your-
self up by your own bootstraps^paradigm that commanded
residents to Bget a job, get on it, and get over it,^as put by one
advocate. In this approach, residents who were not meeting
the demands of their empowerment plan or plan of service
were often asked to leave the shelter. This contradicted advo-
cate attitudes that encourage flexibility, accommodation, and
understanding. Shelters with a Bbootstrap mentality^also
tended to neglect internal empowerment, as noted in the quote,
BI wish there was a constructive piece besides, ‘Okay, you’re
out of danger. Get a job, get a house, get your feet back under
you.’I wish we could build strengths into people.^
Other shelters had more of an enabling attitude, where ad-
vocates often ended up doing things for the clients, such as
completing housing applications for clients they have not seen
in months. Advocates in these shelters were concerned that
clients were not learning how to make their own decisions,
how to advocate for themselves, or how to be independent.
These advocates wanted more resident accountability and a
move away from Bjust warehousing people,^as described by
one advocate. When asked what would help the shelter be
more effective, another worker wanted a Bphilosophy change^
that encouraged residents’internal motivation through exter-
nal motivation (e.g. requiring class attendance). Advocates
cited funding requirements and not wanting to control clients
J Fam Viol (2015) 30:467–478 473
as reasons they could not require more of their clients. In both
bootstrap and enabling shelters, advocates reportedattempting
to change the culture of the shelter by modeling balanced
empowerment, suggesting policy changes, and proposing in-
novative programs. Unfortunately, these efforts often resulted
in little change, leading advocates to feel frustrated and
powerless.
When shelters had a more balanced approach to empower-
ment, advocates struggled less. Unlike bootstrap and enabling
shelters, balanced shelters encouraged internal and external
empowerment. One advocate referred to this as helping vic-
tims Bbloom into the women they want to be.^They did this in
a plethora of ways, such as providing support groups and court
accompaniment to help victims feel supported, adopting a
flattened hierarchy that views residents as people rather than
clients, and providing classes and outings to improve resi-
dents’physical, emotional, social, and spiritual health. These
shelters strongly discouraged enabling, as it robbed clients of
the opportunity to gain self-confidence. Finally, because em-
powerment was seen as both an internal and external process,
goals did not necessarily have to be externally measurable. As
one advocate noted, BOne client’s goals was to get closer to
God. That’s not something I can say when you made it, but
that’sokayifthat’syourgoal.^
Shelter Culture and Retention
Although the challenges of advocacy were fairly consistent
across shelters, advocates’job satisfaction varied widely, with
workers often either very satisfied or very dissatisfied. Satis-
faction did not seem to be related to shelter size, location,
budget, or access to resources. Instead, satisfaction was highly
related to the work environment, which several advocates re-
ferred to as the Bshelter culture.^
Supportive Cultures When advocates were asked to rate their
satisfaction on a scale of one to 10, those who rated it above an
eight all worked in shelters with a supportive and encouraging
culture. In these shelters, there was a spirit of connection,
collaboration, and unity. These shelters were also more likely
to encourage both internal and external empowerment and
have a flattened hierarchy between residents and staff. When
asked why she rated her satisfaction at a 10, one advocate
replied, BBecause we have an excellent group of people that
work extremely well together. We work in fluid movement. A
big crew comes in and someone’s grabbing paperwork, some-
one’s taking the kids, someone’s taking the parent.^BWe ’re
not just helping clients,^said another, Bwe are helping each
other help clients.^
Having a supportive shelter culture helped advocates man-
age the challenges of shelter life. When advocates were
shocked by the inhumanity of violence or overwhelmed by
the chaos of shelter life, they turned to each other. When
advocates were heartbroken by victims returning to abusive
relationships or felt ineffective, lost, or unsure, it was in each
other that they found comfort and confidence.
Advocates also repeatedly emphasized the importance of
management and executive staff in shaping their experience.
BIt starts from the top down,^said one supervisor:
[The executive director] sets the examples for all of us.
The way she treats me transfers to how I treat my clients.
So, if I love my job, and love the work atmosphere, and
love the staff, then it just continues into what I do with
clients.
Supervisors and executive directors with satisfied advo-
cates shared two common characteristics. First, these supervi-
sors were involved. They spent time daily in the shelter and
were able to greet clients by name. They were willing to do
any task, including scrubbing toilets, fielding the crisis line, or
transporting clients, which enabled them to better understand
the experience of their advocates. Involved supervisors also
had Bopen door policies^that made them available to listen to,
process, and trouble shoot with advocates. In addition to help-
ing advocates feel supported, and valued, being involved
allowed supervisors to spot subtle signs of burnout. Satisfied
advocates appreciated their supervisors’emphasis on self-care
and reported flexible schedules that encouraged time off.
Second, satisfied advocates had supervisors and executive
staff with a vision that unified and provided direction. Stated
examples of shelter visions included: Bemphasizing wellness,^
Btaking a client-first approach^or Baddressing the whole
client.^Shelter visions promoted teamwork by providing a
common goal while also allowing advocates to identify their
niche in the vision, such as legal advocacy or mental wellness.
Supervisors with a vision also seemed more apt to communi-
cate their expectations for advocates and provide formal and
informal trainings to assist advocates in meeting those expecta-
tions. Moreover, these supervisors hired with their vision in
mind and selected staff that could enhance the team.
Demoralizing Cultures Dissatisfied advocates described shel-
ters with demoralizing cultures in which they felt de-valued
and isolated. As one advocate related:
When I started, I would go home and talk about my
clients and go, ‘Oh, my gosh!’and now when I go home
and go ‘Oh, my gosh!’it is about a co-worker. I just
can’t stand it and every person that has left or is leaving
it’s because of that. You can’t stay and have a good heart
and I hate that. I think it’sonlybecauseIamkindof
isolated that I stay, which stinks because I hate being
isolated. But, at the same time, it lets me just love my
clients and not have to deal with the people that are
making other people leave... My clients have never been
474 J Fam Viol (2015) 30:467–478
the reason I thought about leaving. It is about support
and wishing I was valued.
Dissatisfied advocates reported having supervisors with
poor leadership abilities. In addition to being uninvolved and
lacking vision, these supervisors had difficulty resolving con-
flict. Even satisfied advocates reported occasional difficulty
with co-workers, however, those problems were addressed
and resolved. In unsupportive shelters, advocates complained
that policies and procedures were not followed, that problems
between employees were not addressed, and that favoritism
prevented Bproblem employees^from being fired. As de-
scribed by one worker, this caused a Bwhy should I [care]^
attitude that Blower[ed] the bar^for everyone. Moreover, ad-
vocates reported little awareness of burnout and noted that
while taking time off for self-care was permitted, supervisors
subtly discouraged it. Finally, these shelters were also more
likely to adopt enabling or bootstrap empowerment strategies
and an authoritarian hierarchy.
Leaving, Promoting, and Settling in Both advocate and shel-
ter factors contributed to the leave/stay decision, but a sup-
portive shelter culture was most strongly related to advocates
who settled in to the shelter life. Settled advocates had worked
for their agency for several years and reported their level of
satisfaction at eight or above. These advocates stayed with
their shelter because of their passion for the work and the
supportive shelter culture. In most of these shelters, turnover
still occurred, which settled advocates attributed to advocate
characteristics. BSome people just can’t handle the crisis,^
said one. Another advocate added, BWhen people are hired,
we tell them you have to wear many different hats... some
people don’t like that.^However, two shelters denied that
turnover was a problem. Both of these shelters were smaller,
rural, and had flattened hierarchies that emphasized empow-
erment, which may contribute to low levels of turnover. Set-
tled advocates also attributed turnover to a lack of passion. As
one advocate stated, BYouhavetobeinlovewiththiswork.
It’s one of those things you either have or you don’t. There are
no in-betweens. If you are just here for a paycheck, it’sgoing
to be a long day.^
While having passion is undoubtedly a protective factor,
the advocates who had left or were leaving demonstrated pas-
sion on par with the settled advocates. When asked what had
kept them in their positions, the advocates who had decided to
leave cited their love for the mission and their love of clients.
When asked why they left, they blamed the shelter culture.
One former supervisor commented, BI was miserable because
of the personnel issues. It was being the boss of that many
people and having such non-support.^Advocates who had
decided to leave their shelters were also more likely to work
in bootstrap or enabling environments that did not fit their
views of advocacy. These advocates further cited low
wages and few opportunities for promotion as reasons
for leaving.
Instead of leaving or settling in, some advocates—both
satisfied and dissatisfied—promoted, though they did so for
different reasons. When satisfied advocates promoted, it was
by furthering their education to obtain a higher position, to
learn a specialized skill, or to further macro-level changes in
client services. Dissatisfied advocates, on the other hand, pro-
moted to survive. Changing positions breathed life into
exhausted employees and provided an opportunity to imple-
ment new procedures to improve advocate self-care, training,
and connectedness. Interestingly, these advocates often found
that their efforts to create change were resisted by both higher
and lower level employees. Over time, emotional exhaustion
would set-in again.
Discussion
This study explored the relationship between the challenges of
DV shelter advocacy, advocate satisfaction, and advocate reten-
tion. The adversities described herein were consistent with diffi-
culties described elsewhere, including accepting chaos (Ullman
and Townsend 2007), hearing clients’stories (Behounek 2011),
and victims returning to abusive situations (Dunn and Powell-
Williams 2007). This article furthers our understanding of advo-
cates’experiences by identifying workplace challenges related to
empowerment and rule enforcement, as well as elucidating how
advocates cope with these struggles, which has important impli-
cations for supervisors and directors.
While challenges were consistent across DV shelters, ad-
vocate satisfaction and retention varied widely depending on
shelter culture. When the shelter culture enabled workers to
successfully manage advocacy challenges, workers were more
satisfied and more likely to settle in. Conversely, when poli-
cies and procedures led advocates to feel isolated, unsupport-
ed, abusive, or ineffective, advocates were more dissatisfied
and more likely to leave. This suggests that supervisors and
directors can improve retention by prioritizing connection,
cooperation, flexibility, innovation, and self-care, thus en-
abling workers to better meet the challenges of advocacy.
Although turnover is likely a complicated phenomenon
resulting from both advocate and shelter factors, this research
proposes that supervisors and directors can impact turnover.
While several advocates posited that turnover stemmed from
advocate characteristics (e.g., a lack of passion, inability to
manage crisis), which may be partially true, this study sug-
gests that passionate, able advocates are being lost due to other
factors. In other words, capable workers are leaving because
they do not have the tools to manage the challenges, not be-
cause of the challenges themselves. Additionally, this research
indicates that supportive environments may build strengths
into workers who are not already equipped to thrive in shelter
J Fam Viol (2015) 30:467–478 475
life. Recall that several advocates began working at their shel-
ter for financial reasons and only became zealous after
witnessing the severity of domestic violence. Early support
can enable such workers to overcome these heart-wrenching
experiences, while ongoing encouragement will help them to
eventually settle in.
The idea that organizational culture influences turnover is not
a new finding, though this does appear to be the first study to
examine this phenomenon in a domestic violence shelter. Several
studies have documented that organizational culture affects em-
ployee retention (Aarons and Sawitzky 2006a; Glisson and
James 2002; Williams and Glisson 2013). These studies defined
organizational culture as the normative beliefs and shared expec-
tations about how things are done in a workplace (Glisson and
James 2002). Culture reflects an organization’s deep, unstated
values and assumptions (e.g., prioritizing the conservation of
resources or believing many people abuse the system), though
it is also may stem from the entities’environment (Glisson and
James 2002). For example, an organization that values employee
mental health may adopt rigid rules in a chaotic, high-stakes
environment to prevent employee exhaustion. This proposition
has important implications for supervisors and directors in
demoralizing shelters, as it suggests that these cultures developed
in reaction to the challenges of advocacy in an effort to protect
workers, residents, and resources. Unfortunately, these responses
generally lead to increased emotional exhaustion and turnover
(Glisson and James 2002).
One reason that turnover may result from negative cultures
is the dissonance between employees’values and the values of
the shelter culture. For example, when one advocate
complained,
We make it more difficult for them with ‘now you have
to leave’... People are going to break the rules and we
have to work with those rules rather than putting them
back in [an abusive situation],
she seemed to be commenting on a difference in values. In
other words, while she valued flexibility and individual
wellbeing,administration possibly prioritized fairness, confor-
mity, and group safety. Though these values are not terribly
different on paper, in practice they may feel far apart, espe-
cially when the result is a victim returning a dangerous situa-
tion. Taken together, all this indicates that shelters should
evaluate the purpose of their policies, the values that inform
them, and the effect they have on workers and residents.
Moreover, shelters must be certain that the values informing
their policies are the ones they wish to emphasize.
Changing Shelter Culture
Unfortunately, there is no guaranteed method for improving
organizational culture. As a matter of fact, some researchers
have found that employees in unsupportive, demoralized cul-
tures are resistant to change (Aarons and Sawitzky (2006b).
As such, improving culture may be an arduous task. Regard-
less, as this study suggests it is necessary for the wellbeing of
advocates, possible means of improving culture are included
here.
It seems prudent for supervisors and directors to begin by
evaluating the present state of their shelter’sclimate.Thismay
include asking questions similar to those used in this study,
asking directly about shelter strengths and weaknesses, or
using a formal measurement, such as the Organizational So-
cial Context (OSC; Glisson et al. 2008). The OSC assesses
Boccupational climate,^culture, and work attitudes by mea-
suring rigidity, proficiency, resistance, stress, engagement,
functionality, and moral. Norm-based results allow shelters
to compare their scores on these measures to other
organizations.
As noted previously, directors and supervisors may also
want to examine the values and assumptions that underlie
current policies and procedures to ensure a fit with their vision
for serving clients and employees. For example, shelter staff
could explore their understanding of empowerment and how it
fits with what they currently do. As seen in this study, many
advocates wanted procedures that encouraged empowerment
both externally (e.g., gaining employment, renting an apart-
ment) and internally (e.g., feeling connected and capable).
Similarly, if the shelter lacks a unifying vision, one should
be developed. Involving workers and residents in creating a
vision for the shelter may foster investment in the mission.
Then, directors and managers can develop programs and pro-
tocols to meet the shelter’s vision, build connections between
management and advocates, and address the challenges of
shelter life; to increase investment, this step should be done
with the input of workers and residents. This is also an oppor-
tune time to explore adopting more flexible rules, such as
allowing residents to decide chores or only requiring a Bplan
of service^for long-term clients. Though shelter size may
limit flexibility, when possible, bendable rules reduce the
angst associated with enforcement.
Additionally, increasing supervision and training, mandat-
ing workers take a paid monthly Bmental health day,^and
providing regular opportunities to process difficult cases and
brainstorm solutions can facilitate connections and confi-
dence. These activities may be especially important during
the first yearof employment when advocates are facing shelter
shock and would likely aid them in Bletting go of being the
hero.^Further, helping employees identify strengths and find
their niche could increase feelings of effectiveness.
Research Implications
Before we can understand how to create change in
demoralized shelters, however, the nuances of both supportive
476 J Fam Viol (2015) 30:467–478
and demoralizing cultures must be elucidated. Expanding in-
terviews to include executive staff, former advocates, and cli-
ents may help researchers answer questions such as: How are
cultures created and maintained? What is the advocate’srole
in sustaining the culture? What values and assumptions un-
derlie the shelter culture, and into what behaviors do these
translate? What other factors contribute to advocate longevi-
ty? What is the effect of shelter culture on clients’experiences
and outcomes? And, more specifically, what is the impact of
empowerment style and hierarchy on client outcomes?
Our results also provide more practical directions for future
research. First, client interventions must fit the context of the
shelter. In other words, researching and developing interven-
tions that can be implemented quickly and informally by pro-
fessionals, para-professionals, and volunteers that address cri-
sis and trauma would be most helpful. Second, advocates
reported that substance abuse among residents was common,
but they lacked the tools and training necessary to address it.
Specifically, they wondered if violence and substance abuse
could be treated concurrently or if it is necessary for one to be
treated before the other. Moreover, they worried about the risk
to other residents by admitting victims with known addictions.
Future research should address these questions and concerns.
Limitations and Conclusion
Unfortunately, data collection methods may limit the findings
of this study. Interviews were gathered before the main anal-
ysis could take place, thus limiting our ability for constant
comparative analysis and the opportunity to ask more pointed
questions. Additionally, time constraints prevented member-
checking prior to publication. Also, the sample was obtained
from one state, and it may be that other states have policies
that affect advocates’experiences in a way that differs from
the experiences of this sample.
Demographics may also limit this study. For example, only
two of the 19 participants were male. However, this may re-
flect the population of domestic violence advocates and of
social workers as a whole. The NASW (2006) survey, for
instance, reported 81 % of nearly 5000 responders were fe-
male. Likewise, Murray and Welch’s(2010) survey of shelter
advocate found that over 90 % were female. Additionally, the
racial composition of this study may be more diverse than
found in a typical shelter. The NASW (2006)reportedthat
86 % of their survey respondents were White/Non-Hispanic,
7 % were Black/African American, and 4 % were Hispanic/
Latino. In this study, however, 37 % were Caucasian, 37 %
were Hispanic, and 26 % were African American. This study
may be further limited by the number of participants, though
small sample sizes are often common with qualitative
research.
Despite these limitations, this study furthers our under-
standing of advocates’daily experience by providing insight
into the challenges of shelter life. More importantly, it iden-
tifies shelter culture as a moderator of these challenges, which,
thankfully, is a variable over which shelter staff have control.
Hopefully, this encourages advocates, supervisors, and direc-
tors to take steps to better their environment, leading to im-
proved outcomes for workers and clients alike.
References
Aarons G. A., & Sawitzky A. C. (2006a). Organizational climate partially
mediates the effect of culture on work attitudes and staff turnover in
mental health services. Administration and Policy in Mental Health,
33(3), 289–301.
Aarons G. A., & Sawitzky A. C. (2006b). Organizational culture and
climate and mental health provider attitudes toward evidence-
based practice. Psychological Services,3(1), 61–72.
Allen K. R. (2000). A conscious and inclusive family studies. Journal of
Marriage and the Family,62(1), 4–17.
Baker L., O'Brien K., & Salahuddin N. (2007). Are shelter workers
burned out?: An examination of stress, social support, and coping.
JournalofFamilyViolence,22,465–474. doi:10.1007/s10896-007-
9103-1.
Barnett O. W., Miller-Perrin C. L., & Perrin R. D. (2011). Family violence
across the lifespan (3rd ed., ). Thousand Oaks: Sage.
Behounek E. K. (2011). No Help for the Weary. An ethnographic exam-
ination of factors impacting burnout among domestic violence and
sexual assault advocates Retrieved from: http://opensiuc.lib.siu.edu/
gs_rp/50.
Bemiller M., & Williams L. S. (2011). The role of adaptation in advocate
burnout: A case of good soldiering. Violence Against Women,17(1),
89–110.
Black M. C., Basile K. C., Breiding M. J., Smith S. G., Walters M. L.,
Merrick M. T., Chen J., & Stevens M. R. (2011). The national
intimate partner and sexual violence survey (nisvs): 2010 summary
report. Atlanta: National Center for Injury Prevention and Control,
Centers for Disease Control and Prevention.
Brown C., & O’Brien K. (1998). Understanding stress and burnout in
shelter workers. Professional Psychology: Research and Practice,
29(4), 383–385.
Catalano, S., Smith, E., Synder, H. & Rand, M. (2009). Bureau of Justice
Statistics, Selected Findings, Female Victims of Violence. Retrieved
from: http://bjs.ojp.usdoj.gov/content/pub/pdf/fvv.pdf.
Charmaz K. (2006). Constructing grounded theory: a practical
guidebook. Thousand Oaks: Sage.
Corbin J., & Strauss A. (2008). Basics of qualitative research: techniques
and procedures for developing grounded theory (3rd ed., ).
Thousand Oaks: Sage.
Crewell J. W. (2007). Qualitative Inquiry and research design: choosing
among five approaches (2nd ed., ). Thousand Oaks: Sage.
ConsultingL. P. (2012). The power of partnership: strategic restructuring
among domestic violence organizations Retrieved from: http://
www.blueshieldcafoundation.org.
Dunn J., & Powell-Williams M. (2007). BEverybody makes choices^:
Victim advocates and the social construction of battered women's
victimization and agency. Violence Against Women,13(10), 977–
1001.
Glisson C., Dukes D., & Green P. (2006). The effects of the RC organi-
zational intervention on caseworker turnover, climate, and culture in
children’s service systems. Child Abuse & Neglect,30,855–880.
J Fam Viol (2015) 30:467–478 477
Glisson C., & Hemmelgarn A. (1998). The effects of organizational cli-
mate and interorganizational coordination on the quality and out-
comes of children’s service systems. Child Abuse & Neglect,
22(5), 401–421.
Glisson C., & James L. R. (2002). The cross-level effects of culture and
climate in human service teams. Journal of Organizational
Behavior,23,767–794. doi:10.1002/job.l62.
Glisson C., Landsverk J., Schoenwald S., Kelleher K., Hoagwood K. E.,
Mayberg S., & Green P. (2008). Assessing the Organizational Social
Context (OSC) of mental health services: implications for research
and practice. Administration and policy in mental health,35,98–
113. doi:10.1007/s10488-007-0148-5.
Guterman N. B. (2004). Advancing prevention research on child abuse,
youth violence, and domestic violence: emerging strategies and is-
sues. Journal of Interpersonal Violence,19(3), 299–321.
Haley-Locke A. (2007). A workforce or workplace crisis?–applying an
organizational perspective to the study of human services employ-
ment. Administration in Social Work,31(3), 41–61.
Lincoln Y. S., & Guba E. G. (1985). Naturalistic inquiry. Beverly Hills,
CA: Sage.
Massachusetts Council of Human Service Providers. (2007). Recruiting
and retaining the next generation of human services workers in
massachusetts. Retrieved from: http://www.donahue.umassp.edu/
docs/help-wanted-2
Murray C. E., Smith P. H., & Avent J. R. (2010). Solutions to the
research-practice gap in domestic violence: a modified Delphi study
with domestic violence coalition leaders. Journal of Aggression,
Maltreatment & Trauma,19,424–449. doi:10.1080/
10926771003781354.
Murray C., & Welch M. (2010). Preliminary construction of a service
provider-informed domestic violence research agenda. Journal of
Interpersonal Violence,25(12), 2279–2296.
National Association of Social Work. (2006). Assuring the sufficiency of
a frontline workforce: a national study of licensed social workers.
Retrieved from: http://workforce.socialworkers.org/studies/nasw_
06_execsummary.pdf
National Network to End Domestic Violence (2011). Domestic violence
counts 2010: a 24-hour census of domestic violence shelters and
services. Retrieved from: http://nnedv.org/docs/Census/
DVCounts2010/DVCounts10_Report_Color.pdf
National Violence Against Women Prevention Research Center (2001).
Fostering collaborations to prevent violence against women: inte-
grating findings from practitioner and researcher focus groups.
Charleston: Author.
Schow D. (2006). The culture of domestic violence advocacy: values of
equality/behaviors of control. Wo me n & H ea lt h,43,49–68. doi:10.
1300/J013v43n04_04.
Selden D. R. (2010). The effects of staff turnover on psychiatric rehabil-
itation programs. Psychiatric Rehabilitation Journal,34(1), 71–73.
Ullman S., & Townsend S. (2007). Barriers to working with sexual as-
sault survivors: a qualitative study of rape crisis center workers.
Violence Against Women,13(4), 412–443.
United States General Accounting Office. (2003). HHS could play a
greater role in helping child welfare agencies recruit and retain staff
(Report to Congressional Requesters, GAO-03–357). Retrieved
from: http://www.gao.gov/new.items/d03357.pdf
Wies J. (2009). Boundaries in carework: a case study of domestic vio-
lence shelter advocates in the USA. Global Public Health,4(5),
464–476.
Williams N. J., & Glisson C. (2013). Reducing turnover is not enough:
the need for proficient organizational cultures to support positive
youth outcomes in child welfare. Children and Youth Service
Review,35,1871–1877. doi:10.1016/j.avb.2005.02.002.
478 J Fam Viol (2015) 30:467–478
