Content uploaded by Patricia A Resick
Author content
All content in this area was uploaded by Patricia A Resick on Jan 05, 2015
Content may be subject to copyright.
Journal of Family Violence, Vol. 19, No. 5, October 2004 (
C
2004)
Coping Among Adult Female Victims of Domestic Violence
Angela E. Waldrop
1,3
and Patricia A. Resick
2
This paper examines the current literature regarding coping among battered women. It considers a
number of contextual factors that are related to women’s choices in coping with partner abuse, includ-
ing factors related to the relationship (e.g., frequency and severity of abuse, length of relationship)
and women’s resources (e.g., social support, financial resources). Relationships between different
forms of coping and psychological outcomes are also examined. As the research in domestic violence
coping is somewhat lacking in coherence, theories, and models from the broader coping literature
are used to organize the findings from the domestic violence field. Methodological concerns, such as
research methods, measurement issues, and sampling, are raised throughout the paper. Suggestions
for future research are made.
KEY WORDS: battered women; coping; environmental factors; psychological difficulties.
The negative physical and psychological sequelae of
intimate partner abuse have been documented by many
researchers in the social sciences (e.g., Follingstad et al.,
1991; Gelles & Harrop, 1989; Goodman et al., 1993a,b).
Abuse inflicted by an intimate partner is the source of a
great deal of psychological distress for many women. Yet,
some manage to survive and emerge from abusive rela-
tionships with fewer negative outcomes than others. This
points to the need for investigations addressing coping
strategies utilized by battered women that preserve their
psychological functioning and their physical well-being
during and after battering relationships.
Despite the tremendous amount of research on cop-
ing over the past two decades, studies of coping strategies
in samples of battered women are few. Very little integra-
tive work has been done with this population; much of the
coping research that does exist in populations of abused
women has been qualitative or descriptive in nature. Re-
searchers in this area also tend to overlook contextual
issues when evaluating the adaptive/maladaptive nature
1
Medical University of South Carolina, Department of Psychiatry,
Charlesron, South Carolina.
2
National Center for PTSD, Women’s Health Sciences Division, VA
Boston Healthcare System.
3
To whom correspondence should be addressed at Medical University
of South Carolina, Dept. of Psychiatry, 67 President St. Charleston,
South Carolina 29425; e-mail: waldrop@musc.edu
of victims’ coping strategies. Abused women’s strategies
for survival have sometimes been compared to the cop-
ing strategies used by community samples in response to
ordinary life stressors with the eventual conclusion that
battered women are lacking in problem-solving. This pa-
per will argue that a battering relationship creates a special
set of circumstances under which a woman decides how
to react and that those circumstances cannot be ignored in
understanding her ways of coping with the violence.
The current review will attempt to address the com-
plexities of the relationships between context, stress, and
coping among battered women. The contextual predictors
of coping strategies will be examined, including such fac-
tors as abuse severity and frequency, available resources,
and skills deficits. Given the paucity of theory in the part-
ner abuse literature, the current review will apply theo-
ries and models from the more general coping literature
to organize the existing research on coping in battered
women. Following the review of contextual factors, the
psychological outcomes associated with different coping
strategies will be addressed. The paper will conclude with
implications for treatment and for future research.
CONCEPTUAL OVERVIEW OF COPING
Coping strategies include a broad diversity of
thoughts and behaviors used to manage the demands of
291
0885-7482/04/1000-0291/0
C
2004 Plenum Publishing Corporation
292 Waldrop and Resick
a taxing situation (Lazarus & Folkman, 1984). Multi-
ple research groups have attempted to organize coping
strategies (and styles) into different categories. There has
been a great deal of variability in the number of cop-
ing dimensions described; however, two primary descrip-
tive factors emerge repeatedly across studies. The first
is the distinction between approach and avoidance cop-
ing (Moos, 1995), also described as active and avoidant
coping (Holahan & Moos, 1987; Mitchell & Hodson,
1983, 1986) and engagement/disengagement (Tobin et al.,
1989). The approach/avoidance construct, as the label im-
plies, indicates whether the individual makes attempts to
change the situation or distance him- or herself from the
stressor as a way to reduce negative outcomes. Examples
of approach coping are “Talked with a friend about the
problem” and “Made a plan of action and followed it”
(Holahan & Moos, 1987, p. 949). In contrast, “Refused
to believe that it happened” and “Kept my feelings to
myself” (Holahan & Moos, 1987, p. 949) are examples of
avoidance coping.
A second factor that has appeared in several stud-
ies is a distinction between cognitive versus behavioral
strategies (De Ridder, 1997; Holahan & Moos, 1987).
Behavioral coping may include observable actions taken
to reduce the impact of stress, for example, “Got away
from things for a while” (Holahan & Moos, 1987, p. 949).
Cognitive coping strategies include attempts at changing
one’s way of thinking about a situation (e.g., “Tried to
see the positive side of the situation”) (Holahan & Moos,
1987, p. 949).
The approach/avoidance and cognitive/behavioral
dimensions have been used to organize microlevel cop-
ing strategies in studies that used rational derivation of
categories (Holahan & Moos, 1987; Moos, 1995), as well
as factor analytic methods (Tobin et al., 1989). Tobin et al.
(1989) provide an elegant demonstration of the hierarchi-
cal factor structure of coping measured by the Coping
Strategies Inventory (CSI). Eight primary factors yielded
problem-focused and emotion-focused secondary factors
and approach and avoidance tertiary factors. Moos (1995)
has described the approach/avoidance construct as the “fo-
cus” of coping and the behavioral/cognitive distinction as
the “method” of coping.
There are two primary methodological perspectives
that commonly appear in the literature on coping. The
inter-individual approach describes responses to stress in
terms of coping “styles” that a person is expected to use
across different types of stressors (Cohen, 1987). Typi-
cally, measures of coping styles simply ask the respondent
how he or she usually reacts to a variety of stressors. A
second approach to measuring coping, the intra-individual
approach, suggests that people change their coping strate-
gies according to the stressor (Cohen, 1987; Holahan &
Moos, 1987; Swindle & Moos, 1992). In this approach,
situation-specific measures assess the coping strategies
used by individuals in response to a particular situation.
Empirical evidence in the coping literature has sup-
ported the intra-individual approach, in that although in-
dividuals may have preferred ways of coping with stres-
sors, their actual responses to stressors tend to vary with
the characteristics of the stressful situation (De Ridder,
1997; Holahan et al., 1996). Therefore, a thorough under-
standing of coping strategies requires taking situational
variability into account. In the context of a battering rela-
tionship, a woman may have a preference for a particular
coping strategy, but find it necessary to adjust that strategy
to fit particular situations. For example, she may prefer
to use approach strategies in the work domain to solve
problems, while choosing to use avoidance strategies in
coping with her severely violent husband.
METHODOLOGICAL ISSUES IN RESEARCH
ON COPING IN RESPONSE TO BATTERING
A potential weakness in the coping literature on
domestic violence is the issue of sampling. As is typical
in clinically-oriented research, participants may be drawn
from help-seeking populations. In the case of domestic
violence survivors, samples are often drawn from shelters
or from agencies that serve battered women. Help-seeking
participants may not reflect the population of domestic
violence survivors as a whole. For example, these women
could differ on level of violence experienced or resource
accessibility.
The current body of literature on domestic violence
is also limited by reliance on cross-sectional designs. Al-
though there is some evidence to suggest a feedback mech-
anism in coping (e.g., Mitchell & Hodson, 1983; Moos,
1995) coping research is typically conducted with assess-
ments made at single time points or with retrospective
designs (Wheaton et al., 1997) that may span relatively
long amounts of time. To gain a more thorough, accu-
rate understanding of the process and causal relationships
in stress and coping, more longitudinal data need to be
collected at multiple time points.
The traumatic stress literature has provided some
guidelines for assessment of stressors. Wheaton et al.
(1997) suggested controlling for prior cumulative stress
and trauma exposure. They contended that particular types
of stressors, such as abuse, may create a unique context
that must be accounted for when considering the effects
of other stressors. Wheaton et al. (1997) described a “ma-
trix of disadvantage” (Wheaton et al., 1997, p. 70) in
abusive environments that contrasts with the experience
Coping in Battered Women 293
of other stressors. Abuse may occur as part of a set of
related problems and may stem from that set of problems
(Wheaton et al., 1997). In this case, other life circum-
stances and stressors may take on importance in under-
standing coping responses to the stressor in question. For
example, a stressor, such as domestic abuse, may occur
in the context of social isolation and a partner’s alco-
hol abuse. Further, there is a body of empirical evidence
that has examined prior trauma history among battered
women and found that childhood sexual and physical
abuse, adult sexual abuse history, and other abusive inti-
mate relationships are associated with later victimization
(e.g., Hilberman & Munson, 1977–78; Koss et al., 1994).
This finding underscores the need to control for prior his-
tory in research on coping among battered women. Each
new trauma may be more stressful due to accumulated
stress. Alternatively, each new stressor may be less stress-
ful due to improved coping or desensitization.
There is a great deal of variability in the use of
coping measures across different studies, making it more
difficult to meaningfully compare results (Follingstad
et al., 1988). As mentioned earlier, due to the variability
in coping constructs, this discussion will focus on
the two more commonly occurring constructs across
studies: approach vs. avoidance coping and cognitive vs.
behavioral coping (Cohen, 1987; De Ridder, 1997). At
this point, Tobin et al. (1989) appear to have created the
only empirically validated measure which incorporates
both of these coping constructs.
COPING IN DOMESTIC VIOLENCE
The literature examining coping in women who have
suffered physical abuse by a spouse or dating partner is
relatively limited. As a result, the previous review draws
most of its theoretical structure and organization from the
general coping research. This larger body of research is
based on coping in response to life stressors and everyday
disturbances that are found rather commonly in the gen-
eral population. The problem of domestic violence is by no
means uncommon. However, as argued earlier, it presents
a special set of circumstances within which coping takes
place (Gondolf & Fisher, 1988).
Predictors of Coping Strategies
Given the importance of context in the selection of
coping strategies (Mattlin et al., 1990; Moos & Swindle,
1990), it is important to consider what factors might be
associated with how women choose to deal with abuse
and its deleterious effects (Dutton, 1993). Lazarus and
Folkman (1984) described constraints that could affect the
choice and usefulness of coping strategies. They organized
these limitations within three categories: environmental
constraints, personal constraints, and extreme threat. In
the case of battered women, multiple examples of cop-
ing constraints could be cited. For example, lack of social
contacts might be considered an environmental constraint.
A personal constraint could be high dependency needs.
Intensity (severity and frequency) of emotional or physi-
cal abuse can be considered indicators of extreme threat.
Given the multiple limitations on coping that an abusive
environment creates, coping behaviors among battered
women cannot be meaningfully understood outside the
context of chronic physical and emotional abuse (Dutton,
1993; Gondolf & Fisher, 1988).
Frequency of Physical Abuse
When examining domestic violence as a research
problem, two of the most salient aspects of battering re-
lationships are frequency and severity of violence. There
is some evidence to support the relationship between fre-
quency of violence and women’s choice of coping strate-
gies. Herbert et al. (1991) examined the cognitive coping
strategies used by women who stayed with and those who
left their abusive partners. They included cognitive strate-
gies, such as amount of focus on positive aspects of the re-
lationship, negative changes in the relationship, extent of
downward social comparisons, and personal attributions
for abuse (Herbert et al., 1991). These cognitive variables
discriminated between the women in this sample who had
left their abusive relationships and those who had not.
The results demonstrated that women who remained
in their abusive relationships (34.4% of the original sam-
ple of 130 women) did not exhibit a relationship between
frequency of physical abuse and types of cognitive strate-
gies used. However, there was a significant negative cor-
relation between frequency of verbal abuse and use of
downward social comparisons. The authors reported that
this pattern of results was similar among the entire sam-
ple. However, they focused on the women who had stayed
with their abusive partners and gave specific correlation
coefficients for this group only.
Frequency of abuse has also been examined in con-
junction with active coping. Gondolf and Fisher (1988)
compared shelter residents and nonresidents, and found
that women who had decided to use the shelter had expe-
rienced more frequent abuse than women who did not use
the shelter. This would suggest that some active forms of
coping are more likely when frequency of abuse increases.
Mitchell and Hodson (1983) examined the relationship
294 Waldrop and Resick
between frequency of abuse and several different forms
of active behavioral coping, active cognitive coping, and
avoidance coping. In their sample of 60 women, they
found that women’s use of avoidance coping increased
with the frequency of physical violence they experienced.
No significant correlations were found between frequency
of battering and amount of active cognitive or behavioral
coping. These results seem to contradict the results of
Gondolf and Fisher (1988). However, it is possible that
active strategies to escape the abuse (e.g., shelter use,
leaving, prosecution) may be associated with different
contexts and predictors than active strategies used while
still in the relationship (e.g., arguing with the abuser, call-
ing police, talking to a friend).
It is interesting to note that these studies (Gondolf
& Fisher, 1988; Herbert et al., 1991) found that the ac-
tive forms of coping significantly related to frequency of
abuse were those that removed a woman from the dan-
gerous relationship (e.g., leaving the relationship, staying
in a shelter). When other active forms of coping were
examined (i.e., talked with a friend, tried to find out more
about the situation) there was not a significant relation-
ship to frequency of violence (Mitchell & Hodson, 1983).
Perhaps frequency of abuse has its largest effect on coping
strategies when it clarifies the need to leave the abusive
situation. However, this interpretation of the pattern of
results should be viewed with caution, as none of them
have tested this hypothesis directly.
Severity of Physical Abuse
Severity of physical abuse is another index of the in-
tensity of abuse that can predict coping strategies. Mitchell
and Hodson (1983) found that higher levels of violence
were associated with more avoidance coping. Women ex-
periencing more severe abuse were also less likely to re-
ceive social support from friends when they attempted to
talk about the abuse. These results point to the potential in-
terrelationships among different types of coping. Perhaps
women who received negative reactions to their active
coping strategies (e.g., seeking social support) become
more likely to engage in avoidant responding.
Mitchell and Hodson (1986) have also used an eco-
logical perspective to examine a model of interrelated
factors in battered women’s coping efforts. Consistent
with their previous research, they found relationships be-
tween severity of abuse and social support variables, and
suggested that more severe abuse was related to more en-
forced social isolation for women. Severity of violence
was negatively associated with both number of social
support network members and number of social contacts
made independent of the abusive partner. Similar to their
previous results, severity of abuse was again related to an
increased likelihood that a woman’s friends would react
to her with avoidance when she sought support. These
results demonstrate the direct and indirect ways in which
more severe violence creates more difficult circumstances
for a woman to take advantage of social support as a
coping resource. Severely abused women may be either
deliberately isolated from social contacts, or their existing
friends become less willing or feel less able to help them.
Mitchell and Hodson (1986) also described the re-
lationship of severity of violence to three types of cop-
ing behaviors: active-cognitive, active-behavioral, and
avoidance. Both childhood exposure to violence and at-
titudes toward women’s social roles were examined as
moderators of the relationship between severity of vio-
lence and types of coping strategies employed. Results
suggested that women without a history of violence in
their family of origin were more likely to use more ac-
tive coping strategies as violence increased in severity;
there was not a significant relationship between severity
and avoidant coping in this subsample. Women with a
history of violence in their families used more avoidant
coping and less active coping efforts in response to in-
creasingly severe physical violence. Mitchell and Hod-
son (1986) found a similar pattern when attitudes toward
women’s roles was examined as a moderator. They found
that battered women with less traditional attitudes toward
women’s roles in society used more active-behavioral cop-
ing strategies in response to more severe violence. Women
with more traditional attitudes used less active-behavioral
coping when the violence was more severe. Therefore,
childhood exposure to violence in the home and degree
of traditional attitudes toward women both moderated the
relationship between severity of violence and the types of
coping strategies women chose.
The authors concluded that childhood exposure to a
violent environment and more traditional attitudes toward
women’s roles contribute to less effective ways of coping
with increasing levels of violence. This study presents
a strong design for studying the role of severity of vio-
lence in determining coping strategies because it exam-
ined important moderating factors. If the authors had not
examined the interactive effects of violence severity and
other predictors, they might have concluded that severity
of violence did not play a role in what coping strategies
women chose to deal with the violence. The investiga-
tion of personal and contextual factors together (Moos &
Swindle, 1990) probably provides the most informative
data regarding domestic violence coping.
Also investigating the factor of abuse severity,
Herbert et al.’s research (Herbert et al., 1991) focused on
Coping in Battered Women 295
what cognitive strategies women used when they stayed
in abusive relationships and what factors were related
to their decisions to leave. The authors found that nei-
ther moderate nor severe physical abuse was significantly
related to cognitive coping strategies. However, in that
study, severity of violence was predictive of the active be-
havioral strategy of leaving the relationship. The results
of their investigation are in contrast to the results of a
longitudinal study of battering couples in which sever-
ity of husband-to-wife physical abuse at Time 1 was not
predictive of whether couples stayed together or were sep-
arated 2 years later (Jacobson et al., 1996). Unfortunately,
there was only an initial assessment of violence severity;
therefore, patterns of increasing or decreasing levels of
violence and their association with relationship status at
the 2-year follow-up could not be examined.
It appears that changes in abuse severity are related
to shifts in women’s approaches to coping with the vio-
lence. Miller and Porter (1983) cited evidence that bat-
tered wives place increasing blame on their husbands for
the abuse and are less likely to blame themselves as the
severity of the violence increases. In a small qualitative
study of shelter residents (n = 15), Ferraro and Johnson
(1983) found that a sudden shift in the severity of abuse
was related to women moving from cognitive strategies
of coping to more active behavioral strategies, such as
leaving the relationship, in reaction to the violence. For
these women, the shift to using active behavioral strate-
gies seemed to occur when women suddenly realized
the potential lethality of the abuse (Ferraro & Johnson,
1983).
These results have been supported by a study using
a much larger sample (n = 100) and quantitative analy-
ses (Rusbult & Martz, 1995). This larger study also sug-
gests that when abuse severity increased, women were less
likely to return to their abusive partners. Follingstad et al.
(1991) also found that relative stability in the severity of
abuse predicted coping behaviors. Women who experi-
enced stable levels of abuse were more likely to seek psy-
chological services specifically related to the abuse than
were women for whom the violence had an increasing or
decreasing trend. However, women who had experienced
less severe abuse over time accessed psychological ser-
vices more for other problems than did their counterparts
who suffered abuse of increasing or fluctuating severity.
These results present a very mixed picture of how sever-
ity of abuse may affect active help seeking, but changes
in abuse level do seem to precipitate changes in coping
strategies.
Both severity of violence and changes in severity
of violence appear to be associated differentially with
various forms of coping behavior. Perhaps an increase
in severity is associated with more active forms of be-
havioral coping, but only those that seem most likely
to end the abuse, such as leaving or utilizing the crim-
inal justice system. Other “active” coping efforts, such as
talking to friends or seeking counseling services, may be
viewed as less likely to be helpful in increasingly dan-
gerous abusive situations. Increasing violence may have
the effect of discouraging women from coping approaches
that help them to stay and deal with violence rather than
removing themselves or their abusive partners from the
situation.
Length of Relationship
Severity and frequency of violence are understand-
ably two of the most common relationship factors ex-
amined in the literature on battered women. The length
of the abusive relationship is an important longitudinal
factor that has been considered less often in the study
of women’s coping efforts. The length of an abusive re-
lationship can change a woman’s coping responses over
time. Some researchers have examined relationship length
as an index of investment (Rusbult & Martz, 1995). This
conceptualization predicts that the more time spent in a
relationship, the more a woman may feel dedicated to
making it work, particularly if the abuse did not appear
until later in the relationship.
In their view of relationship length as an index of
investment level, Rusbult and Martz (1995) were inter-
ested in predicting commitment level and whether women
would stay with or leave their abusive partners. The results
of this longitudinal study indicated that a longer relation-
ship was predictive of a higher commitment level (i.e.,
less intention to separate, more intention to return to the
partner). There was a nonsignificant trend for longer rela-
tionship duration to be associated with a quicker return to
the abusive partner.
Available Resources
As mentioned earlier, a crucially important outcome
of many abusive intimate relationships is to diminish a
woman’s contact with resources that might have helped
her cope with the violence. For example, a battering
partner may prevent a woman from working outside the
home, potentially restricting social contacts and personal
financial resources. Social support, financial resources,
and other tangible sources of help may all contribute to
a battered woman’s choice of coping strategies (Moos,
1995).
296 Waldrop and Resick
One of the more obvious forms of tangible resources
is money. The amount of money a woman has available
to her is relevant to the options that she has in dealing
with an abusive relationship. In a study mentioned earlier
(Rusbult & Martz, 1995), having an independent income,
a higher income, and more available money was asso-
ciated with women either not returning to their partners
within 1 year of the first assessment or taking longer to
return to them. Strube and Barbour (1983) also found
that women’s economic dependence on their batterers,
as measured by employment status and employment his-
tory, was related to their decision to leave the relation-
ship or not. Jacobson et al. (1996), however, did not find
significant differences in wife income between couples
who remained together versus those who separated or di-
vorced at a 2-year follow-up assessment. However, those
authors may not have had sufficient statistical power to
detect any differences in their data. Gondolf and Fisher
(1988), in a comparison of shelter vs. nonshelter resi-
dents, found that women who were not staying in a bat-
tered women’s shelter had higher incomes on average.
Perhaps women who had greater access to funds were able
to choose other living arrangements when they left their
partners.
Mitchell and Hodson’s ecological model (Mitchell
and Hodson, 1983, 1986) of coping conceptualized per-
sonal resources (including individual income) as one de-
terminant of support from institutional help sources (e.g.,
police) and intimate others. They measured personal re-
sources with a composite score of income, education,
and occupational status that is conceptually equivalent
to an indicator of socioeconomic status. Personal re-
sources, social support, and institutional responsiveness
were all expected to determine directly which coping re-
sponses an abused woman would employ. As predicted, a
woman’s greater personal resources were related to hav-
ing a larger number of social support contacts and hav-
ing more frequent outside social contacts unaccompanied
by her abusive partner. These data suggest that women
who have higher occupational status and/or greater in-
come are more likely to have social contacts outside the
home. Having more social contacts increases the amount
of potential available support resources for dealing with
the abuse. A strength of this study is that it examined the
quality of the social support received by abused women
when they chose to utilize their sources of potential
support. Again, women with greater personal resources
were at an advantage. They received more empathic re-
sponses and fewer avoidance responses from friends when
they sought support from them. As will be discussed
in the next section, the responsiveness of friends is re-
lated to whether women engage in active or avoidant
coping in response to violence (Mitchell & Hodson,
1983).
Social Support
Social support is a variable that has been considered
as both a means of coping and a resource contributing
to the availability of other forms of coping (Aspinwall
& Taylor, 1997). The different ways of conceptualizing
the role of social support in coping, and the relative lack
of studies that examine reciprocal relationships between
coping factors, has created some confusion about what
roles social support may play in helping women deal
with domestic violence. In the coping literature on general
stressors and negative life events, social support is often
described as a means for individuals to gain access to
emotional support, but also as a way to gain increased in-
formation and accurate appraisals of stressors (Aspinwall
& Taylor, 1997).
Mitchell and Hodson (1983) considered the quality
and quantity of perceived social support and its relation-
ship to women’s coping responses. They asked women
to report the typical responses they received from friends
when they attempted to talk about the abuse itself. They
found that the fewer avoidant responses women received
from friends, the more likely they were to engage in active-
behavioral and active-cognitive coping. There was no sig-
nificant relationship between friends’ avoidant responses
and avoidance coping in the respondents themselves. It
seems likely that women who engaged in more avoidant
coping did not choose to talk to friends about the abuse.
However, using more avoidance coping strategies was
negatively associated with number of supporters and with
number of contacts with friends and family members inde-
pendent of the abusive partners. Interestingly, the amount
of empathic responses from friends was not significantly
related to the forms of coping examined.
Because the data were correlational, it is unclear
whether women’s coping styles affected their social net-
works and their friends’ responses or the reverse. It is pos-
sible that when a battered woman’s friends see her using
active coping responses they are more likely to respond
supportively; friends who see a woman rely on avoidance
coping may distance themselves, leaving her with fewer
supporters. If, in contrast, social support increases active
coping, then women whose friends will discuss the abuse
with them may be encouraged to cope more actively with
the violence. A third possibility is that friends’ responses
and women’s coping strategies are both affected by a third
variable, such as severity of abuse or the level of control
exerted by the partner.
Coping in Battered Women 297
Dutton et al. (1994) examined the circumstances
of women who attempted to murder their abusive part-
ners. They hypothesized that these women felt more iso-
lated than other abused women. In their study of forensic
and clinical samples of battered women, they found that
women charged with murder or attempted murder of their
abusive partners did report a lower level of perceived so-
cial support than that reported by the clinical sample.
Women in the clinical sample felt more sense of belong-
ing to a social network and more perceived availability
of others who were emotionally supportive. However, the
two groups did not significantly differ in the degree of
tangible social support they reported (e.g., assistance with
transportation or money). The authors interpreted the re-
sults as indications of the greater social isolation and cor-
responding lack of perceived alternatives in women who
eventually turned to extreme violence as a way to cope
with the abuse that they were suffering.
In a more qualitative study, Mills (1985) examined
the role of social support in women’s changing views of
the abuse. In a small shelter sample, she found that some
women reportedly began to re-evaluate the abuse in light
of messages from others that the abuse was wrong. Women
who had resorted to minimization or rationalization of
the violence began to change their ways of thinking as
they heard the opinions of supportive others. This is a
form of social support often termed “appraisal support”
(e.g., Aspinwall & Taylor, 1997). In this sample, this form
of support allowed them to move on to engage in the
active choice to leave the abusive relationship. Ferraro and
Johnson (1983) reported similar results in their qualitative
study of women in a battered women’s shelter. Sullivan
et al. (1992) reported that over 75% of the women in
their study said that they needed more social support, the
highest rated need after material goods. These women
clearly believed social support to be a valuable resource
in their efforts at coping with the violence they suffered.
As mentioned earlier, a causal relationship between
social support and coping has not been empirically es-
tablished. Despite the lack of causal data, the availability
and quality of social support are typically interpreted as
influences on the coping strategies that a woman chooses
in reaction to domestic violence. More available perceived
social supports and more empathic responses (vs. avoid-
ance) from those supports may facilitate more active cop-
ing and greater perceived options for battered women.
Social support may be especially important for increas-
ing the feasibility of a battered woman’s ability to leave
her abuser. The support that others can provide, such as
reappraisal of the violence and tangible support, could
contribute to whether a woman feels trapped in the violent
relationship (Dutton et al., 1994). These are all feasible
explanations of the data. However, despite the tendency
of many researchers to interpret the data in this man-
ner, it is also possible that women’s ways of coping with
violence affect their potential support relationships. For
example, many supporters may turn away if they feel that
a woman is not trying to remove herself from an abusive
relationship, or a woman who is using avoidant coping
may have distanced herself from others. In light of these
possible alternative explanations, causal conclusions must
be established by future research.
Prior Effectiveness of Coping Strategies
Women’s use of particular coping strategies is re-
flective of context and personal factors as well as the out-
comes that they expect from their strategies. Therefore,
one determinant of a woman’s choice of coping strategies
will be the results that she achieved after using that strat-
egy in the past (Bowker, 1983). The research in the area
of prior coping effectiveness has stemmed in part from
the literature on the responsiveness of formal and infor-
mal help sources. The responses of support sources (e.g.,
police, family members) to a woman’s request for help
have sometimes been problematic. For example, women
have cited cases in which their own families and friends
have refused to aid them in a violent situation (Ferraro
& Johnson, 1983), or in which the police have refused to
arrest an abusing husband (Dutton, 1993).
Some researchers present a learning model to ex-
plain how the outcomes of prior coping strategies affect
future coping efforts. In her classic book on the topic,
Walker (1979) suggested that women who make many ul-
timately unsuccessful attempts to end the violence in their
relationships eventually develop learned helplessness and
stop their attempts to engage in the strategies they may
have tried before. Dutton (1993) also presented a type of
learning model, citing unsuccessful strategies for coping
(e.g., calling the police) as experiences in which women
learn that a particular strategy is not a helpful one, and,
therefore, decrease their use of it.
In their ecological model of coping responses,
Mitchell and Hodson (1986) considered institutional re-
sponse and responses of social supports as important
correlates of battered women’s coping strategies. Based
on their prior experiences, participants rated the helpful-
ness of police, lawyers, and therapists. Participants who
reported greater helpfulness from institutional sources
used more active cognitive and active behavioral cop-
ing. Women who received more avoidance responses from
friends used less active cognitive and behavioral coping
responses. This evidence could be interpreted to mean that
women’s coping responses affected the ways in which
298 Waldrop and Resick
others around them responded to the abuse and to the
women themselves, or that women changed their ap-
proaches to coping following less helpful responses from
outside sources.
Some authors (Bowker, 1983; Pagelow, 1984) have
examined institutional responsiveness in the context of
describing women’s help-seeking efforts. Pagelow (1984)
cited several studies in which researchers found that it
was relatively common for women to call the police and
receive very little or no assistance from them. Bowker
(1983) asked women to rate the level of success of differ-
ent helping sources in assisting them to deal with violent
partners. He found that women rated the police as very or
fairly successful in 34% of incidents, and fairly or very
unsuccessful in 33% of incidents. Social service agencies
were also rated for helpfulness in this study. They were
rated as at least fairly successful in 59% of cases, and fairly
or very unsuccessful in 19% of cases. Among lawyers and
district attorneys, women rated their successfulness as at
least fairly successful in 59% of cases, and fairly or very
unsuccessful in 20% of cases. The clergy were rated by
women as at least fairly successful in 55% of cases, and
fairly or very unsuccessful in 24% of cases. Among infor-
mal help sources, women rated family highly, with ratings
of at least fairly successful in 80% of cases. In-laws were
rated as at least fairly successful in being helpful in 52%
of cases. Neighbors and friends were rated very highly,
reportedly successful in 75% and 84% of instances, re-
spectively. When women sought temporary shelter with
friends, family, or women’s centers, they reported that
96% of these help-seeking incidents were at least fairly
successful.
In the cases of all of these help sources, it is possible
that abused women approach only those people whom
they expect to be receptive (Bowker, 1983). Their ex-
pectancies probably differ across different help sources
even before making any personal attempts to use them.
For example, abused women may expect more assistance
from the police than from clergy. They are then likely
to choose help sources in the future based on the degree
to which their expectancies were met. There were gen-
eral trends for women to access formal help sources in
increasing numbers as the battering continued (Bowker,
1983). Unfortunately, the data do not provide informa-
tion on different patterns of outside resource use by indi-
vidual women. However, data were collected on the fre-
quency with which women accessed different help sources
and the data are consistent with helpfulness ratings. The
most commonly used informal sources were friends and
family members, and the most common formal sources
were social services, legal assistance, and police (Bowker,
1983).
ASSOCIATION BETWEEN COPING
STRATEGIES AND OUTCOMES
Coping research has extensively examined the rela-
tionship between coping strategies and psychological out-
comes. In research on the population of battered women,
the research has typically been much more descriptive in
nature, focusing on the types of strategies used and the
psychological effects of abuse, but infrequently examin-
ing the relationship between the two. As discussed earlier,
women who suffer abuse in intimate relationships are at
risk for many negative psychological outcomes (Gleason,
1993). They are also likely to engage in a multitude of cop-
ing strategies to survive or end the abuse in their relation-
ships or to leave the relationships (Bowker, 1983). Some
limited empirical evidence exists that examines the spe-
cific associations between psychological outcomes and
particular ways of coping among battered women. Three
of the psychological constructs that have been most com-
monly investigated in relation to domestic violence are
depression, PTSD, and self-esteem.
Depression
Depression occurs at high rates among battered
women as evidenced by one review of the research litera-
ture (Holzworth-Munroe et al., 1997), which revealed that
the prevalence rate of depression among battered women
is routinely found to be above 50%, and sometimes re-
ported to be as high as 83%. Using a structured diagnos-
tic interview, Gleason (1993) found major depression in
63% of the women in a shelter sample and 81% of bat-
tered women living in the community. In the same study,
17% of the shelter women and 34% of community bat-
tered women met criteria for dysthymia. The author noted
that a participant could be diagnosed with both major
depression and dysthymia, but did not report comorbidity
rates. Cascardi and O’Leary (1992) found that 52% of the
treatment-seeking women they surveyed scored in the se-
vere range of depressive symptomatology on a self-report
measure. In a larger sample of women who responded
to an advertisement for a study on serious relationship
problems, 28% of the battered women were moderately
to severely depressed, and 11% were severely depressed
on a self-report measure (Campbell et al., 1997).
A limited number of studies have examined the re-
lationships between different coping strategies and de-
pression. Campbell et al. (1997) found that “self-care
agency” negatively predicted depression. Self-care agency
was measured by a questionnaire that yielded a composite
score reflecting a woman’s ability to take care of her own
Coping in Battered Women 299
mental and physical health. It included items not only rel-
evant to problem-solving and decision-making, but also
included items related to self-perceptions and physical
energy levels. These latter items could also be considered
indicators of depression. Therefore, there appears to be
some overlap in the conceptualizations of coping and de-
pression in this study. This overlap in constructs makes
interpretation of the correlation between coping (self-care
agency) and depression difficult. From a more cognitive
perspective, Sato and Heiby (1992) found that “realis-
tic” assessment of responsibility for the abuse and a low
frequency of positive self-reinforcement were associated
with higher depression scores. Contrary to their predic-
tions, low assertiveness, low social support, and negative
self-blame were not significant predictors of depression.
They speculated that these results may have been due to
the overlap among predictor variables.
One longitudinal study compared battering couples 3
years after an initial interview (Quigley & Leonard, 1996).
Results of the investigation demonstrated that women
whose husbands were no longer violent at the follow-
up had lower mean depression scores. Women whose
husbands continued to be violent had higher depression
scores. These data suggest that a woman may be able
to recover somewhat from the negative psychological ef-
fects of abuse if her partner makes genuine changes in his
behavior.
The general coping literature suggests that greater
depression is associated with more avoidance coping
(Billings & Moos, 1981). It is unclear to what extent
depression and avoidance coping have a reciprocal re-
lationship. It seems likely that greater depression could
result from avoidance coping, and that having more de-
pressive symptoms could lead one to increase use of avoid-
ance coping. Among battered women, the picture is fur-
ther complicated by the presence of violence or threat of
violence, which is associated with withdrawal behavior
(Walker, 1991). There is also the possibility of comorbid-
ity with posttraumatic stress disorder, which may have an
avoidance component.
Posttraumatic Stress Disorder
Posttraumatic stress disorder, like depression, is a
common diagnosis among battered women (Holzworth-
Munroe et al., 1997; Saunders, 1994). The trauma of a
single incident of violence or chronic abuse from a partner
can lead women to have any of the cluster of symptoms
of re-experiencing, hyperarousal, and avoidance. The es-
timates of PTSD prevalence among battered women vary.
Gleason (1993) found that 40% of his shelter sample and
31% of his community-dwelling sample met diagnostic
criteria for PTSD. Saunders (1994) diagnosed PTSD in
about 60% of his sample of women seeking assistance
from domestic violence agencies and other agencies. In a
study with more heterogeneous sources of participants, the
rate of PTSD was 81% among physically abused women,
and 63% among women who had been emotionally abused
(Kemp et al., 1995). A significant portion of the sample in
this study were treatment seekers, accounting in part for
the particularly high rate of PTSD in the sample.
As mentioned earlier, the presence of symptoms from
the avoidance cluster of PTSD criteria may lead women
to engage in coping behaviors that are typically consid-
ered maladaptive in the clinical and research literature
(Walker, 1991). Disengagement coping has been associ-
ated with a greater degree of PTSD symptomatology in
battered women (Kemp et al., 1995) and among other
trauma groups (Jones & Barlow, 1990). Symptoms from
the avoidance symptom cluster in particular are associated
with disengagement coping. However, with a correlation
of .39 in one study, PTSD avoidance and disengagement
coping appear to be distinct constructs (Kemp et al., 1995).
There is some limited research examining the rela-
tionship of women’s coping to PTSD symptoms. Among
battered women seeking assistance from public agencies,
conceptually a more active form of coping, Saunders
(1994) found more PTSD among those who sought help
at agencies with services specific to battered women.
However, this relationship disappeared after controlling
for degree of violence experienced. Less perceived so-
cial support has also been associated with higher PTSD
symptoms (Kemp et al., 1995). Unfortunately, this study
assessed only the quantity of perceived social support, not
the quality of that support.
Self-Esteem
Another area of psychological health affected by
battering is self-esteem. Depression and self-esteem are
typically associated with one another, but Cascardi and
O’Leary (1992) found that self-esteem was much more
highly related to abuse than was depressive symptoma-
tology. Each variable shared some unique variance with
physical abuse. The authors suggested that self-esteem
may be a more stable construct and depressive symp-
toms may vary relatively frequently over time (Cascardi
& O’Leary, 1992).
A study comparing chronically battered women and
formerly abused women found that women who were
still involved in a battering relationship had lower self-
esteem on average than formerly abused women (Frisch
300 Waldrop and Resick
& MacKenzie, 1991). The authors went on to describe the
possibility that lowered self-esteem could be associated
with learned helplessness, and therefore, less active forms
of coping (Frisch & MacKenzie, 1991). This formulation
suggests that self-esteem can contribute to a cycle in which
women are less able to engage in active coping, then feel
worse about themselves, and consequently are even less
able to act assertively on their own behalf. Somewhat con-
trary to this suggestion is evidence from Campbell et al.
(1994) in which battered women’s self-esteem was not re-
lated to their relationship status at follow-up over 2 years
later. The authors argue that these data do not support the
learned helplessness model in battered women.
CONCLUSIONS
At this point, the data regarding coping strategies
utilized by battered women permit several conclusions.
Women appear to use more avoidance strategies when they
are still in abusive relationships and trying to cope with
ongoing violence. However, the more severe and frequent
the abuse becomes, the more likely battered women are
to try to take active steps to leave the relationship. Some
factors that might place constraints on these efforts are the
amount of commitment to preserving the relationship and
the resources available to help the woman cope with the
abuse in an active and direct manner. Another important
factor is the responsiveness of potential help sources, such
as friends, family, police, and the courts. Women who
receive a more positive response from help sources are
likely to have greater confidence in their abilities to change
their situations, and may be more likely to access support
services in the future. It is, of course, important to keep in
mind that the data may have looked somewhat different
if the participants in this research were not so heavily
weighted toward those who had already left the battering
relationship.
It is also important to remember that most of the
data in this area are correlational. Despite this fact, many
researchers interpret their data with causal conclusions.
However, the direction of the relationships between pro-
posed predictors and coping strategies can only be hy-
pothesized at this point; they remain to be tested with
experimental and longitudinal research. A related issue is
the difficulty of examining coping as a process. Earlier it
was asserted that coping is probably flexible, constantly
changing with feedback and the influence of other contex-
tual variables. This point is especially salient in research
on domestic violence, where the dynamics of chronic bat-
tering relationships contribute to the complexity of the
coping process. Until researchers are able to formulate
a way to observe the coping process, we must continue
to rely on self-reports about how and why coping has
changed over time. This limits the conclusions that we
may draw about the process of coping. Other limitations
of the current literature include biased sampling from
help-seeking agencies, a predominance of cross-sectional
research, and a tremendous amount of variability in the
measures used across studies. All of these issues interfere
with the development of coherent theories and explana-
tory models in the area of coping in domestic violence
research.
Despite the weaknesses in the research, some mean-
ingful information can be drawn from existing findings
to help inform treatment planning. The evidence demon-
strates that contextual forces are important to women’s
selection of strategies in coping with intimate violence.
Women in battering relationships typically try a vari-
ety of coping strategies to deal with the abuse, some of
which are more effective than others (Bowker, 1983). As
Carlson (1997) has suggested, treatment providers should
be willing to view women’s unsuccessful or maladaptive
attempts at coping as efforts to deal with a threatening and
frightening situation. She suggests emphasizing a collab-
orative relationship with a battered woman in which she
can begin to develop her own sense of self-efficacy in
coping with the effects of abuse.
Moos (1995) suggested that interventions aimed at
decreasing negative effects from life stressors should fo-
cus on both the person and his or her life context. Ac-
cordingly, research on the contextual factors in coping
suggests that providing training to women in areas such
as problem-solving skills may not be particularly effective
in the absence of other supports to help them decrease the
violence or leave those relationships. It is important to
help women build the supports that they need to broaden
their options when they decide how to handle the abuse.
Other more fruitful areas of intervention include helping
women to increase their social supports and job skills
and their access to social services, such as legal assis-
tance and orders of protection (Sullivan et al., 1992). It
is also frequently necessary for service providers to assist
women in decreasing the psychological symptoms that
result from victimization. As part of that process, Carlson
(1997) has suggested that the assessment phase should
include questions that help the therapist to understand the
amount of threat that the client experienced in the abusive
relationship, as well as any cognitive distortions that might
interfere with treatment.
There are areas in which the research on coping
among battered women could be improved to provide
better information for understanding the interrelation-
ships of coping and contextual variables, as well as for
Coping in Battered Women 301
designing interventions. The research on coping among
battered women has been largely descriptive in nature. In
the future, researchers in this area should draw on exist-
ing models of coping or create new ones as a means of
understanding and integrating their findings. A productive
starting point may be to continue to gather more data using
the constructs of approach vs. avoidance coping and be-
havioral vs. cognitive coping strategies. Tobin et al.’s work
(Tobin et al., 1989) in this area has helped to integrate the
findings of many researchers and theorists in the coping
field. Their work could be continued through application
of their measure, the Coping Strategies Inventory, in more
research projects. The need for more longitudinal work is
clear, as the direction of relationships between coping and
contextual variables cannot be firmly established at this
point. More research should be conducted to test empiri-
cally the proposed feedback mechanisms in coping (e.g.,
Aspinwall & Taylor, 1997; Mitchell & Hodson, 1986;
Moos, 1995). By examining women at various stages in
abusive relationships, we may begin to understand how
coping may change over the course of a relationship.
REFERENCES
Aspinwall, L. G., and Taylor, S. E. (1997). A stitch in time: Self-
regulation and proactive coping. Psychol. Bull. 121: 417–436.
Billings, A. G., and Moos, R. H. (1981). The role of coping responses
and social resources in attenuating the stress of life events. J. Behav.
Med. 4: 139–157.
Bowker, L. H. (1983). Beating Wife-Beating, Lexington Press,
Lexington, MA.
Campbell, J. C., Kub, J., Belknap, R. A., and Templin, T. N. (1997). Pre-
dictors of depression in battered women, Violence Against Women,
3(3): 271–293.
Campbell, J. C., Miller, P., Cardwell, M. M., and Belknap, R. A. (1994).
Relationship status of battered women over time. Journal of Family
Violence 9(2): 99–111.
Carlson, B. E. (1997). A stress and coping approach to intervention with
abused women. Fa m. Relat. 46: 291–298.
Cascardi, M., and O’Leary, K. D. (1992). Depressive symptomatology,
self-esteem, and self-blame in battered women. J. Fam. Violence 7:
249–259.
Cohen, F. C. (1987). Measurement of coping. In Kasl, S. V., and Cooper,
C. L. (eds.), Stress and Health: Issues in Research Methodology,
Wiley, New York, pp. 283–305.
De Ridder, D. (1997). What is wrong with coping assessment? A review
of conceptual and methodological issues. Psychol. Health 12: 417–
431.
Dutton, M. A. (1993). Understanding women’s responses to domestic
violence: A redefinition of battered woman syndrome. Hofstra Law
Rev. 21: 1191–1242.
Dutton, M. A., Hohnecker, L. C., Halle, P. M., and Burghardt, K. J.
(1994). Traumatic responses among battered women who kill. J.
Trauma. Stress 7: 549–564.
Ferraro, K. J., and Johnson, J. A. (1983). How women experience bat-
tering: The process of victimization. Soc. Probl. 30: 325–339.
Follingstad, D. R., Brennan, A. F., Hause, E. S., Polek, D. S., and
Rutledge, L. L. (1991). Factors moderating physical and psycho-
logical symptoms of battered women. J. Fam. Violence 6: 81–95.
Follingstad, D. R., Neckerman, A. P., and Vormbrock, J. (1988). Reac-
tions to victimization and coping strategies of battered women: The
ties that bind. Clin. Psychol. Rev. 8: 373–390.
Frisch, M. B., and MacKenzie, C. J. (1991). A comparison of formerly
and chronically battered women on cognitive and situational di-
mensions. Psychotherapy 28: 339–344.
Gelles, R. J., and Harrop, J. W. (1989). Violence, battering, and psycho-
logical distress among women. J. Interpers. Violence 4: 400–420.
Gleason, W. J. (1993). Mental disorders in battered women: An empirical
study. Violence Vict. 8: 53–68.
Gondolf, E. W., and Fisher, E. R. (1988). Battered Women as Survivors:
An Alternative to Treating Learned Helplessness, Lexington Books,
Lexington, MA.
Goodman, L. A., Koss, A. P., and Russo, N. F. (1993a). Violence against
women: Mental health effects. Part II. Conceptualizations of post-
traumatic stress. Appl. Prev. Psychol. 2: 123–130.
Goodman, L. A., Koss, A. P., and Russo, N. F. (1993b). Violence against
women: Physical and mental health effects. Part I. Research find-
ings. Appl. Prev. Psychol. 2: 79–89.
Herbert, T. B., Silver, R. C., and Ellard, J. H. (1991). Coping with an
abusive relationship: I. How and why do women stay? J. Marriage
Fam. 53: 311–325.
Hilberman, E., and Munson, M. (1977–78). Sixty battered women. Vic-
timol.: Int. J. 2: 460–471.
Holahan, C. J., and Moos, R. H. (1987). Personal and contextual deter-
minants of coping strategies. J. Pers. Soc. Psychol. 52: 946–955.
Holahan, C. J., Moos, R. H., and Schaefer, J. A. (1996). Coping, stress
resistance, and growth: Conceptualizing adaptive functioning. In
M. Zeidner & N. S. Endler (eds.), Handbook of coping: Theory,
research, applications New York: Wiley, pp. 24–43.
Holzworth-Munroe, A., Smutzler, N., and Sandin, E. (1997). A brief
review of the research on husband violence: Part II: The psycho-
logical effects of husband violence on battered women and their
children. Aggression and Violent Behavior 2(2): 179–213.
Jacobson, N. S., Gottman, J. M., Gortner, E., Berns, S., and Shortt,
J. W. (1996). Psychological factors in the longitudinal course of
battering: When do the couples split up? When does the abuse
decrease? Violence Vict. 11: 371–392.
Jones, J. C., and Barlow, D. H. (1990). The etiology of posttraumatic
stress disorder. Clin.Psychol.Rev.10: 299–328.
Kemp, A., Green, B. L., Hovanitz, C., and Rawlings, E. I. (1995). In-
cidence and correlates of posttraumatic stress disorder in battered
women. J. Interpers. Violence 10: 43–55.
Koss, M. P., Goodman, L. A., Browne, A., Fitzgerald, L. F., Keita, G. P.,
and Russo, N. F. (1994). No Safe Haven: Male Violence Against
Women at Home, at Work, and in the Community, American Psy-
chological Association, Washington, DC.
Lazarus, R. S., and Folkman, S. (1984). Stress, Appraisal, and Coping,
Springer, New York.
Mattlin, J. A., Wethington, E., and Kessler, R. C. (1990). Situational
determinants of coping and coping effectiveness. J. Health Soc.
Behav. 31: 103–122.
Miller, D. T., and Porter, C. A. (1983). Self-blame in victims of violence.
J. Soc. Issues 39: 139–152.
Mills, T. (1985). The assault on the self. Stages in coping with battering
husbands. Qual. Sociol. 8: 103–123.
Mitchell, R. E., and Hodson, C. A. (1983). Coping with domestic vi-
olence: Social support and psychological health among battered
women. Am. J. Commun. Psychol. 11: 629–654.
Mitchell, R. E., and Hodson, C. A. (1986). Coping and social support
among battered women: An ecological perspective. In Hobfoll, S. E.
(ed.), Stress, Social Support, and Women, The Series in Clinical and
Community Psychology, Hemisphere, New York, pp. 153–169.
Moos, R. H. (1995). Development and applications of new measures
of life stressors, social resources, and coping responses. Eur. J.
Psychol. Assess. 11: 1–13.
Moos, R. H., and Swindle, R. W., Jr. (1990). Person–environment trans-
actions and the stressor-appraisal-coping process. Psychol. Inq. 1:
30–32.
302 Waldrop and Resick
Pagelow, M. D. (1984). Family Violence, Praeger, New York.
Quigley, B. M., and Leonard, K. E. (1996). Desistance of husband ag-
gression in the early years of marriage. Violence Vict. 11: 355–370.
Rusbult, C. E., and Martz, J. M. (1995). Remaining in an abusive relation-
ship: An investment model analysis of nonvoluntary dependence.
Pers. Soc. Psychol. Bull. 21: 558–571.
Sato, R. A., and Heiby, E. M. (1992). Correlates of depressive symptoms
among battered women. J. Fam. Violence 7: 229–245.
Saunders, D. G. (1994). Posttraumatic stress symptom profiles of bat-
tered women: A comparison of survivors in two settings. Vict.
Violence 9: 31–43.
Strube, M. J., and Barbour, L. S. (1983). The decision to leave an abu-
sive relationship: Economic dependence and psychological com-
mitment. J. Marriage Fam. 45: 837–844.
Sullivan, C. M., Basta, J., Tan, C., and Davidson, W. S. (1992). After the
crisis: A needs assessment of women leaving a domestic violence
shelter. Violence Vict. 7: 269–275.
Swindle, R. W., Jr., and Moos, R. H. (1992). Life domains in stressors,
coping, and adjustment. In Walsh, W. B., Craik, K. H., and Price,
R. H. (eds.), Person–Environment Psychology, Erlbaum, Hillsdale,
NJ, pp. 1–33.
Tobin, D. L., Holroyd, K. A., Reynolds, R. V., and Wigal, J. K. (1989).
The hierarchical factor structure of the Coping Strategies Inventory.
Cognitive Ther. Res. 13: 343–361.
Walker, L. E. (1979). The Battered Woman, Harper and Row, New York.
Walker, L. E. (1991). Post-traumatic stress disorder in women: Diagnosis
and treatment of battered woman syndrome. Psychotherapy 28: 21–
29.
Wheaton, B., Roszell, P., and Hall, K. (1997). The impact of twenty
childhood and adult traumatic stressors on the risk of psychi-
atric disorder. In Wheaton, B., and Gotlib, I. H. (eds.), Stress
and Adversity Over the Life Course: Trajectories and Turning
Points, Cambridge University Press, Cambridge, UK, pp. 50–
72.
