A Brief Review of The Research On Husband Violence: Amy Holtzworth-Munroe, Natalie Smutzler, and Elizabeth Sandin

Aggression and Violent Behavior. Vol. 2, No. 2, pp.
179-213, 1997
Copyright © 1997 Elsevier Science Ltd
Pergamon Printed in the USA. All rights reserved
1359-1789/97 $17.00 + .00
PII S1359-1789(96)00016-X
A BRIEF REVIEW OF THE RESEARCH

ON HUSBAND VIOLENCE
Part I1" The Psychological Effects of Husband Violence
on Battered Women and Their Children
Amy Holtzworth-Munroe, Natalie Smutzler,

and Elizabeth Sandin
Indiana University
A B S T R A C T , This is the second article, in a series of three, reviewing currently available

empirical data on the problem of husband violence. As discussed in the introduction to this
series of articles (see Holtzworth-Munroe et al., 1995), marital violence is a serious problem
in this country, affecting millions of couples and their children each year. While advocates for
battered women have been actively helping women fiw over twenty years, marital violence
has only received widespread attention from researchers and clinicians in the past 10 to 15
years. Thus, many psychologists did not receive formal training regarding marital violence
and are not well informed about this problem. For example, Browne (1993) is concerned that
therapists often misdiagnose and, thus, mistreat battered women (e.g., medicating depressive
symptoms) because they do not consider that a female client's symptoms may be a conse-
quence of abuse rather than a traditional psychiatric s.vndrome. Similarly, many clinicians do
not adequately assess the possibility that the child problems they are treating may result from
experiencing interparental violence.
This paper was written to provide others, particularly clinicians, with an overview of the
research data on battered wives and their children. By reviewing the available findings, we
hope to provide information regarding the psychological effects of marital violence. © 1997
Elsevier Science Lid
BATTERED VERSUS NONBATTERED W O M E N

The Impact of Battering on Women: Psychological Problems and Psychiatric
Disorders
FOR SEVERAL REASONS, STUDY of battered women' s psychological symptoms and the
application of diagnostic labels to battered women is quite controversial (Sato & Heiby,
1991). First, historically, application of diagnostic labels to battered women often focused on
Correspondence should be addressed to A. Holtzworth-Munroe, Indiana University, Department of

Psychology, Psychology Building, Bloomington, IN 47405-13001.
179
180 A. Holtzworth-Munroe, N. Smutzler, and E. Sandin
the psychological "shortcomings" of these women and/or implied that they were responsible
for their own victimization. For example, early accounts described the masochistic tendencies
of abused women (Snell, Rosenwald, & Robey, 1964; Reynolds & Siegle, 1959), and more
recently, Walker (1991) noted that the DSM-III-R diagnosis of Self-Defeating Personality
Disorder reifies the notion that battered women are somehow responsible for their partners'
aggression.
Second, feminists "argue that diagnosing victims deflects energy from alleviating societal
deficits" (Sato & Heiby, 1991; p. 151). According to feminist theory, marital violence is the
result of an overarching patriarchal system in which violence is simply one of many means
used by men to ensure that women remain subordinate to them (Smith, 1990). Thus, feminists
argue that to stop wife abuse, emphasis must be placed on "leveling" male privileges,
building a more egalitarian society, and holding men (and society) accountable for male
violence (Pence, 1989). Given this view, focusing on the psychological characteristics of the
victim seems not only irrelevant to understanding marital violence but also potentially
counterproductive, as it may allow men to continue to ignore their role in the problem.
In contrast, clinicians and researchers often value diagnostic labeling. First, it can facilitate
treatment planning (Sato & Heiby, 1991). For example, therapists should be made aware of
the symptoms suffered by battered women so that they will consider the possibility of marital
violence in their female clients' lives, rather than assuming that the psychological symptoms
alone should be the focus of treatment (e.g., Browne, 1993). In addition, understanding the
negative impact of physical aggression on women can help demonstrate (e.g., to the public,
the criminal justice system, and social service and funding agencies) that marital violence is
a serious problem, highly deserving of attention and scarce resources.
Despite disagreements about the potential usefulness of studying psychological symptoms
of battered women, researchers from divergent backgrounds have increasingly begun to agree
that symptoms of battered women should be viewed as the result or consequence of abuse
rather than as stable personality characteristics predating abuse or as causes of abuse. Indeed,
Browne (1993) warns that "psychologists continue to risk pathologizing women victims and
[fail] to identify potential danger by not integrating what is known about assaults by
relational partners with the rich literature on human responses to trauma" (p. 1077). Thus,
theories which take traumatic impact into account may help bridge the gap between feminists
and clinicians because it allows for psychological diagnoses that do not "blame the victim."
Before beginning our review, we should note that (with the exception of Post Traumatic
Stress Disorder), researchers studying battered women have rarely examined only one
psychological phenomenon (e.g., depression, self-esteem) in each study. Instead, most
researchers have simultaneously measured several psychological indicators of battered wom-
en's psychological health. However, for ease of presentation, we discuss prevalence rates of
individual psychological problems. We also should note that, unlike in the other sections of
our review, many of the studies reviewed in this section do not include a nonbattered
comparison group. We chose to include these studies for several reasons. First, we wish to
alert clinicians to the prevalence of certain psychological problems among battered women,
irrespective of their prevalence among nonbattered women. Second, some of the studies
report important correlates of psychological symptoms among battered women. Third,
studies reviewed represent the current state of the empirical literature on battered women.
Posttraumatic Stress Disorder (PTSD). For battered women, PTSD involves symptom
development following exposure to an extreme traumatic stressor (i.e., battering) that
includes actual or threatened death or serious injury. PTSD symptoms can include increased
arousal, avoidance responses, numbing of general responsiveness, reexperiencing of the
trauma (Spitzer & Williams, 1985) as well as "perception of life threat, threat of physical
Women and Children 181
violence, physical injury, extreme fear or terror, and a sense of helplessness at the time of the
incident" (Browne, 1993; p. 1081). Several of these symptoms overlap with Walker's (1979)
description of the Battered Woman Syndrome (e.g., numbed affect, flashbacks, increased
startle responses, sleep and eating disorders, and recurrent nightmares). Recently, researchers
have examined prevalence rates of PTSD among samples of battered women, documenting
that PTSD is a frequent problem for battered women. They also have examined correlates of
PTSD symptoms among battered women.
Houskamp and Foy (1991) asked 26 battered women from domestic violence clinics to
complete several measures of P T S D - - the Structured Clinical Interview (SCID; Spitzer &
Williams, 1985), a Symptom Checklist that assessed PTSD symptoms (Foy, Sipprelle,
Rueger, & Carroll, 1984), and the Impact of Event Scale (Horowitz, Wilner, & Alavarez,
1979), which assesses intrusion and avoidance behaviors related to a traumatic event. A
substantial number of women met the criteria for PTSD (i.e., 43% according to the Symptom
Checklist and 45% according to the SCID). In addition, when the women were categorized
as having been exposed to either high or low degrees of perceived life threat, 60% of the
women in the high exposure group met criteria for PTSD compared to 14% in the low
exposure group. Finally, the extent and severity of the abuse (as measured by the Conflict
Tactics Scale), length of relationship, and the perception of life-threat were all positively
related to the severity of PTSD symptoms.
Astin, Lawrence, and Foy (1993) used the Impact of Event Scale and the PTSD Symptom
Checklist to assess PTSD symptomatology among 53 women who were clients at battered
women shelters or a counseling center for abused women. The women had to meet criteria
on both measures to be diagnosed as PTSD-positive; 33% of the sample was classified as
such. Severity of exposure to violence, as well as the recency of the abuse, were positively
related to PTSD symptomatology. Using multiple regression analysis, they found that social
support and positive life events occurring since the battering were negatively correlated with
PTSD symptoms, whereas negative life events and developmental family stressors (i.e.,
poverty, parental discord, mother battered, divorce, death) were positively related to the level
of PTSD severity.
Gleason (1993) conducted a study of 30 battered women who sought emergency help at a
shelter and 32 battered women who sought help from the same agency but lived at home
(with or without the batterer). The lifetime prevalence rate of PTSD was high for both groups
of battered women (i.e., 40% of the shelter population and 31% of the community women).
Saunders (1994) similarly compared PTSD symptoms in two samples of battered women--
159 who had sought help at domestic violence programs and 33 who had obtained help at
other types of programs. A substantial number of women in both groups met criteria for
PTSD (i.e., 60% in domestic violence programs and 62% of those who sought help
elsewhere). Intrusive memories of the abuse, avoidance of reminders of the abuse, hyper-
arousal, and nightmares were relevant symptoms for at least 75% of the women.
Kemp, Green, Hovanitz, and Rawlings (1995) conducted a study examining prevalence of
PTSD among 179 battered women (41 from shelters and 138 from community) and among
a comparison sample of 48 nonbattered but verbally abused women recruited from shelters,
support groups, therapists, and the community. On one measure of PTSD (i.e., PTSD
Self-Report Scale, Kemp, Rawlings, & Green, 1991), they found that 81% of battered women
and 63% of nonbattered women met criteria for PTSD; on the other (i.e., Fairbank's civilian
version of the Mississippi Scale for PTSD), they found that 43% of battered women were
categorized as having PTSD compared to 21% of nonbattered. They compared battered
women who met PTSD criteria to those who did not, finding that extent of physical and
verbal abuse, number of injuries, forced sex, and greater sense of threat discriminated the two
groups, with PTSD group scoring higher on these variables. In regression analyses, they also
182 A. Holtzworth-Munroe, N. Smutzler. and E. Sandin
found that negative life events and a history of child abuse predicted PTSD, while time out
of the relationship was negatively related to PTSD. In addition, perceived qualitative social
support (e.g., feeling supported by friends) was related to lower levels of PTSD, while
disengagement coping (e.g., wishful thinking, social withdrawal, problem avoidance, self-
criticism) was one of the strongest predictors of PTSD symptoms.
Riggs, Kilpatrick, and Resnick (1992) compared four groups of women who were victims
of violent crime (i.e., 14 had experienced marital rape, 10--stranger rape, 12--marital
assault, and 11--some other type of assault) to 96 non-victimized women. They included a
structured interview assessing the women's lifetime experience with criminal victimization,
the Impact of Event Scale, and the SCL-90-R (Derogatis, 1977), which assesses physical
symptomatology related to psychological distress. There were no group differences among
the four victimized groups when compared on both interview and self-report measures. This
study was innovative in that it compared victims of marital violence to women assaulted by
someone other than their husbands. Riggs et al. (1992) argued that "[F]ailure to compare
victims of spouse abuse to women assaulted by strangers reinforces the idea that spousal
violence is inherently different from criminal assault" (p. 284): their results "do not support
the commonly held belief that violence toward a woman that is perpetrated by her husband
is less traumatizing than violence at the hands of others" (p. 292). Yet, while it is important
to treat wife abuse similarly to other violent crimes, it is also important to realize that battered
women experience a unique set of circumstances. For example, they endure repeated abuse
from an intimate partner with whom they spend a great deal time (e.g., Browne, 1993).
Depression. Data from studies with no comparison samples suggest that battered women are
at high risk for depression. Using the Center for Epidemiological Studies Depression Scale
(CES-D; Radloff, 1977) in an early study, Rounsaville (1978) found that 53% of his sample
of self-selected psychiatric patients with a history of battering met the criteria for a diagnosis
of depression, and depression was the most commonly reported clinical disorder. Adminis-
tering the CES-D scale to 435 battered women, Walker (1984) found that the women's mean
score of 18.9 exceeded the high-risk cut-off score of 16 and was twice as high as scores in
epidemiological catchment studies. Using other measures, similar findings have emerged
from other studies with no comparison samples. Mitchell and Hodson (1983) found that
battered women (i.e., 60 women in shelter) reported levels of depressive symptomatology
comparable to those of psychiatric outpatients and significantly higher than those of nonpa-
tient females. Cascardi and O'Leary (1992) found that 52% of a shelter sample of 33 women
reported severe depressive symptomatology on the Beck Depression Inventory (Beck, Ward,
Mendelson, Mock, & Erbaugh, 1961 ). Likewise, Sato and Heiby (1992) reported that 47% of
battered women who had attended battered women's groups or sought shelter met the criteria
for clinical depression. Gleason (1993) reported that the prevalence rate of major depression
for battered women from both shelter and community samples was significantly higher than
for women in the NIMH Epidemiological Catchment Area Program. This finding held for all
three time periods assessed (i.e., lifetime, past 6 months and past 1 month). In fact, most of
the battered women met the criteria for major depression when diagnostic interviews
(Diagnostic Interview Schedule; Robins, Helzer, Crouhan, & Ratcliff, 1981) were conducted.
Specifically, 83% of shelter sample and 80% of the community sample had experienced
major depression in the last 6 months.
Researchers have also compared the prevalence of depression among battered versus
nonbattered women. In a study of 60 immigrant Hispanic women, using the CES-D, Perilla,
Bakeman, and Norris (1994) found that while 30 abused and 30 comparison women did not
differ in level of self-reported depression (i.e., both groups had high levels of depression),
level of depression was positively correlated with experience of husband abuse (r = .40).
Women and ChiMren 183
Findings from other studies of community samples do demonstrate group differences in rates
of depression. Using data from Straus's family violence national survey, Stets and Straus
(1990) found a high rate of depression among victims of marital assaults as compared to
nonassaulted respondents. In a longitudinal study of over 100 women in inner-city London,
Andrews (1995) found that the experience of abuse (primarily marital violence) in adulthood
was associated with depression: 61% of women who had experienced abuse had been
depressed compared with 25% of women who had not. Experiencing abuse as an adult was
related to both a single episode of depression and to chronic and/or recurrent depression:
however, it was not related to the number of depressive symptoms experienced in the worst
depressive episode. Campbell (1989) found that battered women were more likely than
women experiencing serious relationship problems to experience severe depression.
Two studies examining the relationship between husband violence and depression among
women seeking marital therapy have been conducted. Cascardi and O'Leary (1992) found
that women who had been the victims of marital aggression were significantly more likely to
report symptoms of moderate depression than either discordant/nonabused or nonabused/
nondiscordant wives. Christian, O'Leary, and Vivian (1994) found that husband aggression
was significantly related to wives' depressive symptomatology. In addition, multiple regres-
sion analyses showed that being a victim of physical aggression was a significant predictor
of depressive symptomatology for women.
Some researchers have gone beyond descriptions of prevalence of depression, to examine
the correlates of depression among battered women. Sato and Heiby (1992) examined the
relationship between various behavioral skills and depression in a sample of 136 battered
women. Four variables were uniquely associated with depressive symptoms. Battered women
who realistically assessed their battering relationship and who reported poor self-
reinforcement skills, personal or economic losses, and past histories of depression appeared
to be at particular risk for depression. Mitchell and Hodson (1983) examined 60 women from
battered women's shelters, and found that severity and frequency of aggression was related
to increases in depression. In addition, less active cognitive coping and more avoidance
coping were related to more severe depression: minimal personal resources, less help from
institutional sources, and higher levels of avoidance responses from friends were also related
to depression. Campbell (1989) attempted to predict grief (i.e., combination of depression
and stress-related physical symptoms) versus depression among 97 battered women recruited
from a variety of sources; outcome measures included the Beck Depression Inventory and a
modification of the SCL-90. She tbund that low self-esteem was the strongest predictor of
grief, along with relationship conflict (i.e., Conflict Tactics Scale: Straus, 1979) and low
self-care agency (e.g., agency beliefs and perceived ability to care for self); almost 50% of
the variance in grief was accounted lbr by these variables. She was also able to account for
about 57% of the variance in depression scores, with low self-esteem, low self-care agency,
and feelings of less control in the relationship predicting depressive symptoms.
Self-esteem. As reviewed here, the majority of studies suggests that battered women have
lower self-esteem than nonbattered women. However, there are a few exceptions. For
example, using a nonstandardized measure of self-esteem, Walker (1983) studied 435
battered women and found that battered women saw themselves more positively (e.g.,
stronger, more independent, and more sensitive) than other women and men. In addition,
Campbell (1989) compared battered women to other women with serious relationship
problems and found that while both groups scored below normative groups, they did not
differ from one another on a measure of self-esteem, suggesting that low self-esteem may be
related to relationship problems more generally, rather than being limited to the experience
of husband violence.
However, most researchers have found the expected group differences. Using data from
Straus's 1975 National Family Violence Survey, Mills (1984) found that female victims of
severe spousal violence were more likely to have lower self-esteem than were women in
nonviolent marriages. Mitchell and Hodson (1983), using the Rosenberg Self-Esteem Scale,
found that the 60 battered women in a shelter sample scored lower than a community sample
of women who had experienced at least one stressful event or life strain. Using the same
measure in a study of Latinos, Perilla, Bakeman, and Norris (1994) found that 30 abused
women reported significantly lower self-esteem than 30 comparison women. Using the
Barksdale Self-Esteem Evaluation (1972), Aguilar and Nightingale (1994) found that 48
battered women from counseling programs scored significantly lower than 48 nonbattered
community women. Among a group of 33 battered women who sought counseling, Cascardi
and O'Leary's (1992) multiple regression analysis showed that self-esteem scores contrib-
uted more unique variance to battering than depression scores (i.e., 32% vs. 5%, respective-
ly).
Other researchers have begun to examine which aspects of abuse experience correlate with
low self-esteem among battered women. In their study of 60 battered women in shelters,
Mitchell and Hodson (1983) found that severity and frequency of husband violence was
strongly related to lowered self-esteem. Using the Coopersmith Self-Esteem Inventory
(1975), Frisch and MacKenzie (1991) compared self-esteem scores of chronically abused and
formerly abused women. Chronically abused women scored significantly lower than formally
abused women. Similarly, Aguilar and Nightingale (1994) found that battered women who
were currently involved with an abusive partner had lower self-esteem than women who had
not been involved with their abusive partner for at least one year. Aquilar and Nightingale
(1994) also found that while emotional/controlling abuse was significantly related to lower
self-esteem scores, other types of husband abuse (e.g., physical, sexual/emotional) were not,
suggesting that certain abusive behaviors are more related to self-esteem than others.
Conclusions Regarding the Psychological Impact of Battering on Women

First, it is clear that battered women are at high risk for several psychological problems,
including PTSD, depression, and low self-esteem. Indeed, the findings reviewed suggest that
one- to two-thirds of battered women may evidence these problems.
Second, the majority of researchers view the psychological differences between battered
and nonbattered women as consequences of abuse rather than as a precipitating or causal
factor. As stated by Hotaling and Sugarman (1986) in their review, "The review of victim
characteristics makes it clear that the most influential victim precipitant is being female . . . .
Identifying female victims of violence is not possible if only her characteristics are consid-
ered . . . . These findings do not augur well for theoretical models of victimization that focus
upon characteristics of victims" (p. l l8).
Third, while much of the available research has simply documented prevalence of various
psychological problems among samples of battered women, more recent work has begun to
examine correlates of these psychological problems. For example, severity and chronicity of
abuse are positively correlated with psychological symptoms.
Fourth, many potentially important correlates have received little attention so far. For
example, researchers have begun to call for a need to understand how leaving an abusive
relationship affects a woman's psychological reactions to the abuse. Similarly, there is a need
to understand resiliency among battered women. "Failure to account for battered women who
do not possess the factor being assessed is also a common problem for researchers who report
a percentage possessing the factor in question as if that explains things while never stopping
to consider how all of the women without the factor fit into the picture" (Follingstad et al..
1988, p. 388). Finally, longitudinal research should examine the longer-term consequences of
husband violence. As noted in a recent meta-analysis of psychological distress associated
with interpersonal violence:
Clearly, reports of distress on diagnostic measures and measures of general psychopathology are
greatest right after the violence . . . . However, an overreliance on these measures really
obfuscates the question about long-term response to victimization, since these measures are in
some way tied to a crisis response. These measures essentially ignore the possibility that
psychological distress may have a qualitatively different developmental expression over time.
(Weaver & Clum, 1995, p. 134)
Skill Deficits
Researchers have also studied the problem solving and coping skills of battered women. Such
skills may be affected by marital violence and thus can be viewed as consequences of marital
violence. For example, a battered woman may develop learned helplessness after she finds
that none of her actions stop the violence; as part of that syndrome, she may have deficits in
problem solving skills (e.g., Walker, 1979). However, skills can also be viewed as potential
mediators of the effects of violence. For example, battered women with good coping and
problem solving skills might be expected to show fewer negative psychological effects from
battering than women without such skills.
Problem solving skills. A number of researchers has found that battered women have deficits
in their problem solving abilities. In one exception, Campbell (1989) reported that battered
women had thought of or tried more solutions to relationship problems than nonbattered
women with serious relationship problems; however, Campbell does not describe her mea-
sure of problem solving. Most researchers have found the opposite.
Claerhout, Elder, and Janes (1982) compared 14 battered and 20 nonbattered women, all
recruited from social service agencies. They asked women how they would respond to six
hypothetical situations depicting differing levels of conflict with a husband (e.g., husband
tells you that you cannot go shopping with a friend, husband has been drinking and starts an
argument where he begins to push and hit you). Relative to nonbattered women, battered
women generated fewer responses and their responses were less effective (e.g., avoidant,
dependent, or inappropriate/aggressive).
Launius and Jensen (1987) compared the problem-solving of 19 female students who had
been abused by their partners, 19 students who were not abused but in counseling, and 19
students who were neither abused by their partner nor in counseling. They borrowed the three
abusive situations used by Claerhout et al. (1982) and added three nonabusive situations. The
women were asked to generate options, as well as choose the most effective option, for each
situation. Battered women generated fewer total options, as well as less effective options, per
situation, than either the counseling or control groups. In addition, battered women, com-
pared to both the counseling and control groups, chose fewer effective options. These deficits
were evident for both the abusive and general situations.
Launius and Lindquist (1988) compared problem-solving skills of 22 battered women who
sought help from a shelter and 23 nonbattered women. The women were asked to generate
options to one abusive and one nonabusive hypothetical scenario. Their responses were
scored for number of options they generated as well as the amount of time they spent on each
task. For both situations, battered women spent less time generating responses. For the
nonabusive situation, the battered women produced fewer options; however, this was not the
case for the abusive situation.
Morrison, Van Hasselt, and Bellack (1987) recruited 26 maritally violent couples, 26
maritally distressed/nonviolent couples, and 15 satisfactorily married couples. On a role-
playing measure of assertiveness (which included marital and nonmarital situations), they
found that wives in the physically aggressive group exhibited significantly more speech
disturbances than did maritally distressed wives and were more compliant than satisfactorily
married wives.
Thus, it appears that battered women have problem solving skills deficits. They have
difficulty generating effective solutions to both violent and nonviolent relationship scenarios.
However, the methodology used in these studies limits the conclusions that can be drawn. For
example, it is questionable whether nonbattered women in the comparison samples could
imagine themselves in the hypothetical abusive situations. In addition, it is difficult to
determine what types of responses should be considered competent by abused women in
scenarios containing violence; for example, passive or compliant responses to active abuse
may be appropriate. As pointed out by Follingstad et al. (1988), " . . . the focus for the
[battered] woman after traumatization may be specifically on preventing danger rather than
on more global problem-solving" (p. 376).
Coping skills. Much of the work addressing coping styles of abused women is still theoretical
in nature (Follingstad et al., 1988). However, a few researchers have attempted to study the
coping strategies used by abused women.
Using the F-Copes scale, Finn (1985) compared the coping skills of 56 women from an
abuse shelter to reported norms of adult women. Battered women were significantly less
likely to use social and spiritual support and engaged in less reframing than the general
population of women. They also engaged in more passive appraisal. In a study reviewed
above, Kemp et al. (1995) found that, among battered women, disengagement coping (e.g.,
wishful thinking, social withdrawal, problem avoidance, and self-criticism) was a strong
predictor of PTSD symptoms.
Other research groups have examined coping as one variable among many to examine
more complex models of how battered women cope with their abuse. For example, Nurius,
Furrey, and Berlinger (1992) assessed coping capacities among abused women, nonabused
women, women whose partners had committed a sexual crime against their child, and women
whose partners had committed a sexual crime against a nonfamilial child. Coping capacity
was defined in terms of vulnerability factors (e.g., violence in family of origin and stressful
life events), coping resources (e.g., self-esteem, depression, dyadic adjustment, perceived
social support), and coping responses (e.g., blamed yourself, angry at abuser, took action
immediately after abuse). Examining vulnerability factors, relative to comparison women,
battered women were more likely to have a history of witnessing or experiencing abuse as
a child and reported higher levels of stressful life events. Battered women scored lowest on
measures of intrapersonal and interpersonal resources. Finally, regarding coping responses,
and relative to one or both of the sex offender groups, battered women were more afraid of
their partners; they also reported higher levels of self-blame and anger at their partner. The
greatest group differences were found for coping resource variables, leading the authors to
conclude that this "is consistent with the rationale (that) coping resources serve as a
mediating filter for the effects of vulnerability factors and create the social cognitive context
within which coping responses must take place." A lack of these resources "makes it difficult
to initiate and nurture support contacts to help reduce isolation and to augment sources of
reinforcement for desired behavioral change" (p. 238).
In a study reviewed above, Mitchell and Hodson (1983) assessed the impact of violence,
as well as personal resources, social support, institutional responsiveness, and coping on the
psychological health of 60 women from battered women's shelters. Psychological health was
defined in terms of depression, mastery, and self-esteem. The severity and frequency of the
aggression was strongly related to increases in depression and lowered self-esteem. In
addition, less active cognitive copingand more avoidance coping were related to more severe
depression. Minimal personal resources, less help from institutional sources, and higher
levels of avoidance responses from friends were also related to depression. Mitchell and
Hodson (1983) point to the importance of including measures of social support and coping
in order to appreciate the breadth of the psychological consequences related to battering.
Follingstad, Neckerman, and Vormbrock (1988) reviewed a number of theories relating to
coping strategies and styles (e.g., establishing meaning for the experience, using cognitive
defense styles, attribution of blame, learned helplessness, a focus on change and control,
handling anger, and positive strategies for ending the abusive relationship). They concluded,
"In reviewing the research, it is difficult at times to determine if reported characteristics of
battered women should be classified as coping mechanisms or as reactions to the victimiza-
tion as they often become seemingly intertwined over the process of coming to terms with
the abuse" (p. 379). In addition, because there are so many individualized responses to
coping, "[i]t may be fruitless and actually meaningless to look for one main reaction or one
main style of coping" (p. 387).
Cognitions
Researchers also have compared cognitions of battered and nonbattered women. In most of
this work, interest in cognitions stems from the belief that how a women thinks--about the
violence, her relationship, and herself--may influence the steps she is able to take to deal
with the violence in her life.
Attributions for husband violence. Researchers have conducted studies examining the
attributions, or explanations, battered women offer for husband violence. Holtzworth-Munroe
(1988) reviewed a series of these studies (e.g., Malhotra, 1984; Frieze, 1978, 1979, 1987;
O'Leary, Curley, Rosenbaum, & Clarke, 1985; Pagelow, 1981; Porter, 1986; Prescott &
Letko, 1977; Shields & Hanneke, 1983; Walker, 1983) and drew several conclusions.
First, in no study did the majority of female victims of marital violence blame themselves
for their husband's violence. In most studies, the women primarily held the husband
responsible for his violence. More recent research continues to support this conclusion (e.g.,
Bograd, 1988). For example, Cantos, Neidig, and O'Leary (1993) found that wives tended to
blame their spouses more than themselves for violent episodes.
A second general finding was that abused women may attribute causality to their relation-
ship, attributing their husband' s violence to mutual factors such as lack of agreement between
spouses. This conclusion is also supported by recent work. Cantos et al. (1993) found that
wives often offered explanations of mutual responsibility for relationship violence.
Holtzworth-Munroe, Jacobson, Fehrenbach, and Fruzzetti (1992) found that, relative to their
explanations for negative, but nonviolent husband behavior, wives were more likely to
attribute husband violence to the relationship.
A third conclusion reached by Holtzworth-Munroe (1988) was that abused women gen-
erally assume that the violence will not occur again (i.e., make unstable attributions). Recent
research continues to support this conclusion. For example, Holtzworth-Munroe et al. (1992)
found that abused wives believed husband's violence was less likely to occur in the future
than other negative, but nonviolent husband behavior. Bograd (1988) found that wives
frequently cited transient reasons (e.g., husband being drunk, having job problems) for
incidents of violence.
Another dimension of attributions that has more recently been examined is the perceived
intentions of husband violence. Bograd (1988) found that women were significantly less
likely than men (20% of women versus 80% of men) to attribute positive intent (e.g., to stop
a fight) to men's violence. Similarly, Holtzworth-Munroe et al. (1992) found that wives, but
not husbands, attributed more intentionality to husbands' violent behavior than to husbands'
nonviolent, negative behavior.
In addition, researchers have examined variables that seem to affect the types of attribu-
tions battered women offer for husband violence. They have found that wives' attributions
may be influenced by a variety of factors, including the severity of the marital violence she
experiences and her childhood experiences with family aggression. Holtzworth-Munroe
(1988) reviewed studies investigating how the severity of the marital violence affected
women's attributions (e.g., Frieze, Knoble, Zomnir, & Washburn, 1980; Frieze, 1979;Porter,
1983) and concluded that more severe abuse results in wives attributing greater blame to the
husband and less blame to themselves, and also to viewing the reasons for violence as more
stable. Further research has been generally consistent with this finding (e.g., Cantos et al.,
1993). Examining how childhood abuse relates to adult women's attributions for husband
violence, Andrews and Brewin (1990) found that women experiencing marital violence who
made internal and stable attributions of blame for the abuse were more likely than other
women to have been abused, sexually or physically, in childhood. Similarly, Cantos et al.
(1993) found that being abused as a child may increase wives' self blame for husband
violence.
The main purpose of examining wive's attributions for husband violence is the assumption
that the type of attributions offered may be related to women's reactions to the violence. In
one of the few empirical studies relating women's attributions to their responses to husband
violence, Andrews and Brewin (1990) studied women in inner-city London. They found that
among women who had experienced husband violence, women who made internal and stable
attributions of blame (i.e., characterological self-blame) for the abuse were more likely than
women who did not to have suffered persistent depression after the violent relationship
ended. However, in a study of 97 battered women, Campbell (1989) found that the women's
attributions of blame for relationship problems did not add significantly to the prediction of
depression among the women, once self-esteem, self-care agency, and feelings of control in
the relationship were used as predictors.
Sex role beliefs. Researchers have investigated the relationship between traditional sex role
beliefs of women and husband violence, since "it has been hypothesized that women
socialized to accept male-dominant family relationships or to hold traditional sex role
expectations may be more likely to experience violence in adult relationships" (Hotaling &
Sugarman, 1986, p. 106). A variety of hypothesized links between women's traditional sex
role beliefs and marital abuse has been offered. For example, Warren and Lanning (1992)
speculate that "strong traditional feminine traits (nurturant, compassionate, sympathetic,
yielding) without traditional masculine traits (assertive, independent, willing to take a stand)
may make a feminine-typed woman less able to protect and care for herself" (p. 6). As
another example, Schwartz and Mattley (1993) maintain that gender identity is a product of
the appraisals and subsequent actions of others, and thus speculate that a battering relation-
ship causes women to develop a narrowly feminine gender role identity. Thus, traditional sex
roles could increase risk of violence, could decrease likelihood that a woman would leave a
relationship once violence begins, and/or could be the result of experiencing marital violence.
At this time, the empirical evidence for a correlation between women's traditional sex role
beliefs and abuse is mixed. In their 1986 review, Hotaling and Sugarman found that 50% of
studies reported a positive relationship and 50% found no association, while no study found
a negative correlation of traditional sex role beliefs and marital violence.
Subsequent researchers have continued to investigate this link, with generally positive
results. For example, Warren and Lanning (1992) compared battered women from domestic
violence shelters to nonbattered women who contacted a mental health facility, assessing
gender identity using the Bem Sex Role Inventory. The battered women were more feminine
and the nonbattered women were either more masculine or androgynous. Using a different
questionnaire, Schwartz and Mattley (1993) compared women in shelters for domestic
violence to demographically matched nonbattered controls; they found that battered women
were lower in masculine traits, but not higher in feminine traits. Thus, both studies generally
support the notion that battered women hold more traditional gender roles.
As noted by Hotaling and Sugarman (1986), many of the available studies on sex role
traditionalism, including Warren and Lanning (1992) and Schwartz and Mattley (1992), do
not allow us "discriminate victims of wife abuse from women in nonviolent, highly conflicted
relationships" (p. 110). To do so, Apt and Hurlbert (1993) compared a sample of abused
women from couples in a marital violence treatment program to a group of nonbattered wives
from couples in marital therapy; all couples were recruited from a military setting. These
researchers assessed traditional versus feminist sex role beliefs and found that abused women
held a more traditional sex role ideology than nonabused, maritally distressed women.
Therefore, while the evidence is not entirely consistent, researchers have generally found
a positive relationship between women's traditional sex role beliefs and marital abuse.
However, the exact nature of this relationship has not yet been determined.
Dependency. While most theorists conceptualize wives' marital dependency in terms of

economic resources, some have speculated about cognitive or emotional dependency. It has
been hypothesized that wives with high marital dependency may be more tolerant of marital
abuse. Kalmuss and Straus (1990) examined the Straus national survey sample to investigate
the relationship of wives' objective and subjective marital dependency to their experience of
marital violence. Objective dependency was defined by economic factors, such as wives'
unemployment, while subjective dependency was conceptualized as the wives' perception
that marriage dissolution would hurt herself more than her husband (e.g., in terms of loss of
friends, finances, sex, angry relatives, and loneliness). Wives' subjective dependency, but not
objective dependency, was significantly related to experiencing minor violence. However, the
opposite pattern held for severe violence--objective dependency was significantly correlated
with experiencing severe violence, while subjective dependency was not. Overall, both
objectively and subjectively dependent wives experienced more abuse. The authors speculate
that dependent wives "tolerate m o r e . . , abuse than 'non-dependent' wives, perhaps because
they have (or perceive) fewer viable alternatives to marriage" (pp. 375-376).
Family of Origin
While the link between family of origin violence and a man's later use of marital violence
seems obvious, it also has been proposed that women who grew up in violent homes are at
risk to become adult victims of spouse abuse. A few ideas have been offered to explain how
one learns or inherits a vulnerability to aggression from their family of origin. For example,
girls growing up in violent homes may come to believe that marital violence is normal and
acceptable, thus tolerating such abuse when it occurs in their own marriages. Others have
theorized that women who have been victims of family of origin violence "may be provoking
their own victimization in their current families" (Cappell & Heiner, 1990, p. 136) to repeat
patterns learned while growing up.
In this section, we review studies comparing battered and nonbattered women's retrospec-
tive reports of childhood family of origin violence. Unfortunately, at this time, there is no
standardized measure of family of origin violence. Thus, some researchers asked whether the
subject witnessed parental violence, while others assessed whether the subject was a victim
of child abuse. The perpetrator (e.g., father or mother) and type of aggression (e.g., sexual or
physical abuse) examined varies across studies, as does the number of items used to assess
family of origin violence.
Studies finding a relationship. In their earlier review, Hotaling and Sugarman (1986)
concluded that of the 42 risk markers they examined, witnessing parental violence was the
only consistent predictor of being a victim of marital aggression.
Indeed, early studies often demonstrated a relationship between family of origin violence
and being a victim of husband violence. For example, based on interviews with 80 families,
Gelles (1976) reported that women who witnessed parental aggression were more likely to
be abused in their own marital relationship than women who had not witnessed parental
aggression. In addition, the more frequently a woman was hit by her parents, the more likely
she was to be hit by her husband. Women who had been hit by their parents were also more
likely to stay with their abusive partners. Similarly, Parker and Schumacher (1977) assessed
20 battered and 30 nonbattered women and found that if a respondent's mother had been
abused by her husband, she was significantly more likely to be a victim of abuse herself.
There was, however, no significant relationship between being victimized as a child and
being abused by a husband as an adult.
Similar findings have emerged in more recent studies. Langhinrichsen-Rohling, Neidig,
and Thorn (1995) studied military couples in therapy for domestic violence and found that
21% of the variance in being a victim of husband aggression was accounted for by having
been beaten by one's father. Cappell and Heiner (1990) found that, among a sample of 888
child-rearing families from the community, interparental aggression in the wife's family of
origin increased the likelihood of current wife abuse. In their study of 181 community
families, Doumas, Margolin and John (1994) found that for women, parental aggression was
predictive of being a victim of spousal abuse.
Studying 204 recently divorced mothers, Simons, Johnson, Beaman and Conger (1993)
assessed the relationship between harsh discipline by parents in childhood and having been
a victim of husband violence. The correlation between harsh maternal parenting and wife
abuse was significant while paternal harsh parenting was not significantly related to wife
abuse. Using structural equation modeling, Simons et al. (1993) found that the effect of harsh
parenting on wife abuse was mediated by the woman's aggressive, antisocial orientation
(e.g., adolescent antisocial behavior, being aggressive toward spouse, and substance abuse).
Specifically, both maternal and paternal harsh parenting styles were moderately associated
with women's aggressive/deviant style of behavior which, in turn, was strongly related to
wife abuse.
Studies not f i n d i n g a relationship. Other researchers have not found a link between family
of origin violence and the experience of spouse violence among women. For example,
focusing on the experience of child abuse in their review, Hotaling and Sugarman (1986)
concluded that experiencing violence as a child or adolescent was an inconsistent risk marker
of being a victim of husband abuse. Specifically, this variable produced nonsignificant results
when studies included a comparison group of women from nonviolent/distressed relation-
ships.
Failure to find a relationship between family of origin violence and current husband
violence has occurred in several studies using clinical samples. For example, in her study of
46 physically abused women and 12 emotionally, but not physically, abused women, Star
(1978) asked women about their experiences with childhood victimization. The two groups
did not differ significantly in either their reports of being a victim of childhood aggression or
witnessing parental aggression. In fact, a higher percentage of nonabused women reported
parental aggression (i.e., 50% vs. 35%, respectively). Rosenbaum and O'Leary (1981) found
a similar pattern of results in their study of 52 abused women, 20 nonviolent/distressed
women, and 20 nonviolent/nondistressed women. Across the three groups of wives, no
significant differences were found for either witnessing parental aggression or experiencing
child abuse. O'Leary and Curley (1986) also found no significant differences across groups
of 22 abused, 23 nonviolent/distressed, and 27 nonviolent/nondistressed women on measures
of either observing parental aggression or being beaten as a child.
Researchers studying community samples also have not consistently demonstrated a
significant relationship between family of origin violence and being a victim of husband
violence. Hotaling and Sugarman (1990) conducted a multivariate analysis of almost 700
women from the 1975 national survey (Straus et al., 1980) and found that neither witnessing
parental violence (i.e., frequency of father hitting mother and vice versa) nor experiencing
parental violence as a child (frequency of father's and mother's use of physical punishment)
differentiated women who were in nonabusive, verbally aggressive, mildly or severely
aggressive relationships.
Summary of family of origin studies. To summarize, studies examining the relationship

between family of origin violence and being a female victim of husband aggression have
produced mixed findings. At this time, it can not be definitively concluded that having grown
up in a violent home is a risk factor for becoming a victim of husband violence. On the other
hand, neither can we dismiss this proposed link. It should be kept in mind that many women
who grew up in violent homes do not experience later husband violence, while many women
who grew up in nonviolent homes do.
PSYCHOLOGICAL ABUSE
The present paper is primarily concerned with the impact of husband physical aggression.
Recently, however, researchers have begun to recognize the importance of psychological
abuse, and future researchers will most likely focus attention on this problem (e.g., a new
Journal of Emotional Abuse will soon be published). Thus, we include this section to briefly
draw attention to the issue of psychological abuse and its impact on battered women.
Defining and Measuring PsychologicalAbuse

Some researchers have used interviews with battered women and qualitative research
methods to study psychological abuse. In so doing, they have uncovered a consistent theme
in which "abusive behavior represents an attempt to control, dominate, or gain power over
one's partner" (Murphy & Cascardi, 1993, p. 90). Quantitative researchers, on the other hand,
have used objective measures such as behavioral checklists. Currently, seven different
objective measures have been developed and/or used to assess psychological abuse and
aggression--the Conflict Tactics Scales (CTS: Straus, 1979); the Index of Spouse Abuse Scale
(ISAS; Hudson & Mclntosh, 1981); the Wife Abuse Inventory (WAI: Lewis, 1985); the
Spouse-Specific Assertiveness Scale (SSAS; O'Leary & Curley, 1986); the Psychological
Maltreatment of Women Inventory (PMWI; Tolman, 1989); the Abusive Behavior Inventory
(ABI; Shepard & Campbell, 19921); and the Measure of Wife Abuse (MWA; Rodenburg &
Fantuzzo, 1993). For more information on these measures, the reader is referred to Murphy
and Cascardi (1993).
Unfortunately, none of the currently available measures has been widely adopted by
marital violence researchers; thus, the definition and measurement of psychological abuse
varies widely across studies. However, in the course of developing these measures, research-
ers have begun to delineate different types of psychological abuse. For example, Rodenburg
and Fantuzzo (1993) designed their measure (the MWA) to assess a broad range of abusive
behaviors, including sexual (e.g., "forced you to do unwanted sexual acts"), psychological
(e.g., "harassed you over the telephone"), and verbal (e.g., "told you that you were stupid")
abuse, and Shepard and Campbell (1992) developed the ABI to assess such varied emotion-
ally abusive behaviors as isolation (e.g., "tried to keep you from something you wanted to do,
like going out with friends"), threats (e.g., "used your children to threaten you, for example,
said he would leave town with the children"), use of male privilege (e.g., "refused to do
housework or childcare"), and economic control (e.g., "prevented you from having money for
your own use"). In one of the few studies to examine whether different types of psychological
abuse can be reliably discriminated, Tolman (1989) found that the items on his measure (the
PMWI) loaded onto two factors---emotional-verbal abuse (e.g., "yelled and screamed at
you") and dominance-isolation (e.g., "monitored your time and made you account for where
you were"). All of the CTS psychological abuse items Tolman (1989) borrowed for his
measure loaded on the emotional-verbal abuse factor, suggesting that therapists and research-
ers using only the CTS may not be assessing other important types of abusive behaviors.
Prevalence of Psychological Abuse

Much of the available prevalence data on psychological abuse comes from studies in which
the CTS was the primary measure. Murphy and Cascardi (1993) reviewed data from studies
using the CTS psychological' aggression items (i.e., insulted or swore at your partner; did or
said something to spite partner; stomped out of room, house or yard; sulked or refused to talk
about an issue). The percentage of couples reporting the occurrence of these behaviors ranged
from 89%-97% for couples seeking marital therapy (Barling, O'Leary, Jouriles, Vivian, &
MacEwen, 1987), 67%-87% for engaged couples (Barling et al., 1987), and 33%-58% for
participants in the Straus et al. national survey (Stets, 1990). These high percentages suggest
that items measured with the CTS" may be normative in our culture, occurring in a wide
variety of couples. These data also raise the question of when these types of behaviors should
be called "abusive." In general, researchers believe that they become abusive when they are
used to control the partner and/or given a background of physical aggression (i.e., a threat of
violence may be different for a wife who has been the victim of her partner's physical abuse
than for a wife who has not).
We do know that psychological abuse is very prevalent among couples experiencing
physical aggression. For example, Follingstad et al. (1990) assessed the prevalence of
emotional abuse among women who had been victims of physical abuse. They found that
99% of the women interviewed experienced at least some type of emotional abuse and 72%
experienced four or more types of abuse (i.e., threats of abuse, ridicule, jealousy, threats to
change marriage, restriction, and damage to property). Similarly, Hudson and Mclntosh
(1981) administered the Index of Spouse Abuse Scale to a sample from a battered women's
shelter and found a correlation of .86 between the physical abuse and psychological abuse
factors. Finally, as reviewed below, in Straus's 1985 national survey, Stets (1990) found that
less than 1% of respondents indicated that physical aggression had occurred without verbal
aggression.
In addition, some beginning work suggests that we may be able to differentiate the verbal
aggression engaged in by violent and nonviolent couples. Infante, Chandler, Rudd, and
Shannon (1990) interviewed a shelter sample of 40 abused women and a comparison
community sample of 64 nonabused women. First, they found that the abused women
reported greater numbers of both husband and wife verbally aggressive statements in past
marital disagreements than did the nonabused women (i.e., about 35 husband and 19 wife
verbally aggressive statements in the violent group vs. 4.5 husband and 3.5 wife verbally
aggressive statements in the nonviolent group); unfortunately, however, they asked abused
wives about violent disagreements and nonviolent wives about nonviolent arguments, so we
cannot conclude that violent and nonviolent husbands differ during nonviolent arguments.
Second, the groups could be discriminated (i.e., 95% of the nonabused women and 72.5% of
abused women were correctly classified) by type of husband, but not wife, verbal aggression;
abusive husbands were more likely than nonviolent men to use character attacks, curses, and
threats.
The Impact of Psychological Aggression on Battered Women

Available data also suggest that psychological aggression is very damaging, at least among
samples of physically battered women (e.g., Walker, 1979). For example, in their study of
women who had been victims of physical abuse, Follingstad et al. (1990) found that 72% of
the women believed that emotional abuse had a more severe negative impact on them than
physical aggression. This may not be surprising given the frequent occurrence of various
forms of emotional abuse--over 60% of the women reported that multiple forms of psycho-
logical abuse occurred weekly; these women also reported that emotional abuse escalated
over time. As noted by Murphy and Cascardi (1993), "it is possible that many of the effects
commonly observed in battered women are due to psychological abuse or combined effects
with physical abuse" (p. 103).
The Relationship Between Physical and Psychological Aggression

Recently, researchers have begun to explore the relationship between physical and psycho-
logical aggression. For example, Straus and Sweet (1992) investigated the relationship
between verbal aggression and correlates commonly associated with physical aggression
(e.g., demographic variables, alcohol). Using the CTS as a measure of verbal aggression
among 5,232 couples in the national family violence survey, verbal aggression was not
related to socioeconomic status or race. However, verbal aggression did decrease with age
and number of children in the family and increased with the use of alcohol and drugs. The
authors concluded that "verbal aggression is part of a pattern of abusive and problematic
interpersonal relationships within the family that has antecedents similar to those of physical
aggression" (p. 346).
Stets (1990) tested two explanations of how verbal and physical abuse might be related.
She argued that verbal and physical aggression could either be manifestations of the same
underlying phenomena, in which case verbal aggression would only occur in the context of
physical aggression, or they could be manifestations of two different phenomena where
verbal aggression proceeds, and progresses to, physical aggression. Using data from Straus'
1985 national survey, she found support for the latter explanation. In over 50% of the cases,
verbal aggression, as defined by the CTS, occurred without physical aggression. However,
the reverse was not true; less than 1% of respondents indicated that physical aggression
occurred without verbal aggression.
In a study of correlates of male aggression (sexual and nonsexual) towards women,
Malamuth, Linz, Heavey, Barnes, and Acker (1995) studied almost 150 men. They demon-
strated that men's use of verbal aggression was the only direct predictor of their use of
physical non-sexual aggression toward women, with verbal aggression accounting for over
30% of the variance in physical aggression.
Findings from a longitudinal study of almost 300 newly married couples similarly support
the notion that psychological aggression leads to physical aggression. Specifically, Murphy
and O'Leary (1989) found that husbands' use of psychological aggression at 18 months after
marriage significantly predicted their physical aggression at 30 months after marriage. Using
the same sample, O'Leary, Malone, and Tyree (1994) demonstrated that certain variables
(e.g., husband defendence and marital discord) predicted husbands' use of psychological
aggression which, in turn, predicted their later use of physical aggression.
Summary of Psychological Abuse

Only recently have researchers begun to study husbands' psychological abuse of women.
These initial studies indicate that psychological abuse is a major problem, being highly
prevalent among violent couples and resulting in negative consequences for battered women.
In addition, psychological abuse may precede or predict husband physical violence.
W H Y BATTERED W O M E N STAY
Do Battered Women Really Stay in Abusive Relationships?
One of the most commonly asked questions concerning marital violence is why battered
women stay in an abusive relationship (Okun, 1986). Schwartz (1988) suggests that a more
appropriate question may be "why do we assume that abused women stay in their marriag-
es?" (p. 241). In fact, based on interviews with 2,254 women in the National Crime Survey
(NCS), Schwartz (1988) found that a substantial number of victims of spousal assault were
either divorced (31.8%) or separated (46.9%) at the time of the interview. These rates were
significantly higher than those found among women who had been victims of a stranger
assault (19.1% divorced; 11.2% separated) and women who were not victimized (8.5%
divorced; 3.6% separated). Schwartz concluded that "the d a t a . . , are consistent with the
argument that in fact the majority of physically abused women do not stay in these
relationships. The recognition that a great many women do in fact take appropriate measures
to attempt to stop the violence caused by men in their lives is important to inform our general
attitude toward women" (p. 247). He further points out that continued emphasis on the
question of why women stay has "prevented the examination of, and led to a dearth of data
on, the problems of women abused by ex-spouses, or about the variety of ways in which
women take active and strong steps to improve their own lives" (p. 247).
Several researchers have reported similar data. For example, 62% of Strube and Barbour's
sample (1984) were separated or divorced from their abusive partners at a follow-up which
ranged from 1 to 18 months after intake. Herbert et al. (1991) found that, among 132 abused
women recruited from the community, almost 66% had left their abusive relationship at the
time of the assessment interview. Of the women who were still living with an abusive partner,
50% had left him on at least one previous occasion. Similarly, Okun (1986) reported that over
30% of the women in a sample of 187 shelter residents terminated their abusive relationships
directly after leaving the shelter and over 43% ended their relationships within 2 years. The
average number of separations among women who returned to their partner was 2.42
compared to 5.07 among the residents who did not return, suggesting that the decision to
leave an abusive relationship is an ongoing process which may require multiple attempts.
Thus, it appears that a large number of women do leave abusive relationships. It also is
apparent that women may require more than one shelter stay or separation before they
permanently terminate the relationship. Okun (1986) summarizes the importance of these
findings by stating "the process of separating and recohabiting by battered women is not an
endless back-and-forth shuttle in static equilibrium, but a progressive process toward change
in the violent relationship, including the termination of that relationship" (p. 118).
Is Leaving an Abusive Relationship a Safe Alternative?

Asking why women stay "implies that leaving is the safer alternative for all battered women"
(Okun, 1986). However, this may not be the case. In the only study comparing rates of
spousal homicide among cohabitating and separated couples, Wilson and Daly (1993) found
that, in three countries (Australia, Canada, and the U.S.), wives were at much higher (i.e., 2
to 4 times) risk of being killed by their husbands following separation than when living with
their partner. Similarly, in their investigation of family and intimate assaults, Saltzman and
Mercy (1993) found that over one-fifth of the fatal incidents occurred between people whose
relationships had terminated. They concluded:
This [finding] calls into question the presumption that an individual can always stop the violence
by terminating an abusive relationship. It also suggests that intervention strategies cannot focus
solely on intact family units or ongoing relationships. The effectiveness of strategies for
protecting estranged partners (e.g., court-issued temporary protective orders) must be seriously
examined and evaluated. (p. 72)
Factors Contributing to a Woman's Decision to Leave or Stay

Strube (1988) reviewed the empirical literature addressing the question of why women stay
in abusive relationships. He noted that many of the available studies suffered from such
methodological limitations as nonrandom samples, retrospective designs, and overreliance on
self-report measures. In addition, the samples studied (e.g., women from private social
service agencies, Gelles, 1976; women in shelters, Snyder & Sheer, 1981) and the length of
the foll0w-up period (e.g., from 6 weeks in Snyder & Sheer, 1981, to 18 months in Strube
& Barbour, 1983) varied considerably across studies. Many of the studies Strube (1988)
reviewed (e.g., Berk, Newton, & Berk, 1986; Giles-Sims, 1983; Hilberman & Munson, 1978;
Kortlath, 1979; Labell, 1979; Pfouts, 1978; Rounsaville, 1978; Snyder & Fruchtman, 1981:
Stone, 1984; Walker, 1983) provided indirect evidence about why women stay or leave (i.e.,
the purpose of these studies was not to determine the antecedents of relationship status).
Other studies (e.g., Gelles, 1976; Snyder & Scheer. 1981; Strube & Barbour, 1983, 1984)
were designed to identify factors contributing to a woman's decision to remain in or leave an
abusive relationship.
Across the studies reviewed, Strube (1988) concluded that "these studies paint a picture of
women who lack the economic means to leave an abusive relationship, are willing to tolerate
abuse so long as it does not become too severe or involve the children, and who appear to
be very committed to making their relationship last" (p. 240). While the Strube (1988) review
is more comprehensive, in this paper, we review some of the available data regarding a few
of the major factors believed to influence a woman's decision to stay in or leave an abusive
relationship.
Financial influences. Strube and Barbour (1983, 1984) conducted two of the first studies to
systematically study the relationship between economic dependence and women's decisions
to leave abusive relationships. In both studies, economic dependence was related to a
woman's decision to stay in an abusive relationship; women who were employed were more
likely to leave an abusive relationship, whereas women who remained in abusive relation-
ships attributed economic hardship as at least one factor for staying. Similarly, among 312
battered women, Aguirre (1985) found that if the husband was the sole source of income, a
woman was more likely to return to him after seeking help from a shelter. Okun (1986) also
found a significant relationship between "the relative economic position of the battered
woman and her violent mate" (p. 113). Specifically, 54.5% of women who had a higher
income than their husbands immediately terminated the relationship, compared to 27.7% of
the women whose husbands earned equal or greater pay. In addition, if the husband was not
gainfully employed, 48% of the wives immediately terminated the relationship whereas
26.6% of women whose husbands were full-time or self- employed terminated cohabitation.
In the only study in which women were directly asked whether finances influenced their
decision, Herbert, Silver, and Ellard (1991) reported that 39.5% of the abused women in their
sample stated they would leave their partner if they had the resources to do so; however,
52.6% of the women stated they would not leave their partner given the resources.
In summary, it appears that financial status influences a woman's decision to remain in an
abusive relationship. However, these results must be interpreted with caution, since the
majority of the women in these studies were in battered women shelters. These shelters often
provide services to lower income women who may not have friends or family to rely on
during a crisis. Thus, there is little information as to how women from higher income
brackets cope with wife abuse.
Severity and frequency o f violence. Empirical studies have provided mixed and contradic-
tory evidence for the notion that the more severe the marital violence, the more likely the
woman is to leave. In an early study of 41 women, Gelles (1976) found that the less severe
and less frequent the violence, the more likely a woman was to stay with her partner.
However, some researchers have found the opposite. For example, among 350 women in
shelters, Pagelow (1981) found that the more severe the physical injuries, the longer women
remained with their spouses. Likewise, Hilbert and Hilbert (1984), studying 35 women from
a shelter, found that "the more severe the abuse and the longer the length of the relationship,
the more likely the victim will return to her batterer" (p. 295). In contrast, at least two
researchers have found insignificant relationships between severity of abuse and the women' s
decision to stay or leave. Aquirre (1985), in a study of 312 battered women entering shelters,
found that the number of injuries a women incurred was not a significant predictor of leaving
the relationship, and Schwartz (1988) reported that the extent of a woman's injuries was not
a significant predictor of whether the women were divorced or separated from their husbands.
Based upon clinical interviews with battered women, Ferraro and Johnson (1983) specu-
lated that it may not be frequency or severity of violence that serves as an impetus for a
woman's decision to leave a relationship, but rather "a sudden change in the relative level of
violence. Women who suddenly realize that battering may be fatal may reject rationalizations
in order to save their lives" (p. 331). They also suggest that "as battering progresses, periods
of remorse may shorten, or disappear, eliminating the basis for maintaining a positive outlook
on marriage" (p. 332). Thus, focusing exclusively on severity and frequency of violent
episodes may offer too simplistic of an explanation of why women choose to stay or leave.
Rather, it is may be a significant variation in the p a t t e r n of abuse that instigates a separation.
Marital commitment. Loseke and Cahill (1984) challenge the notion that staying in an
abusive relationship is in fact a "deviant" response. They argue that terminating any type of
intimate relationship is difficult, noting that "battered women are simply defined as assaulted
wives who remain with assaultive mates, and their relationships are portrayed as no more
than victimizing processes . . . One effect of the victimism practiced by the experts on
battered women is that possible experiential and behavioral similarities between battered
women and other persons are overlooked . . . However, even a cursory review of the
sociological literature on marital stability and instability suggests that, at least in regard to
their reluctance to leave their mates, battered women are quite similar to both other women
and to men" (p. 304).
In two studies, Strube and Barbour (1983, 1984) sought to determine whether battered
women's psychological commitment to their relationships influenced their decision to remain
with their abusive partners. In both studies, they found that measures of commitment were
related to a woman's decision, with higher commitment (i.e., "love" or length of time in the
relationship) being related to staying.
OTHER FACTORS AFFECTING WOMEN'S DECISION TO STAY OR LEAVE

Interestingly, researchers often approach the question of "Why battered women stay" as if the
woman was the only person participating in the process. For example, investigators rarely
consider how help obtained by women may impact on the choice of remaining or leaving the
relationship. Indeed, researchers are just recently beginning to consider how shelters and
community service agencies impact a battered woman's decision making process, both in
terms of whether she returns to her partner and what steps she must take to survive without
him (Aguirre, 1985; Berk, Newton, & Berk, 1986; Hamilton & Coates, 1993; McKeel &
Sporakowski; Sirles, Lipchick, & Kowalski, 1993; Sullivan, Basta, Tan, & Davidson, 1992;
Vinton, 1992; Wilson, Baglioni, & Downings, 1989).
Similarly, we found that researchers rarely consider the batterer's influence over his
partner's decision to return to the relationship. However, the few existing findings suggest
that this may be an important factor to examine. For example, Schutte, Malouff, and Doyle
(1988) found that all 24 battered women in their shelter sample reported that their partner had
attempted to persuade them to return to the relationship. The majority of women were
influenced by such things as their partners' promises to change, apologies for the abuse,
denial of responsibility, withholding of contact with children, gifts and promises, reported
difficulties of independent living, reminders of family responsibilities, threats to harm
someone, and mentioning family responsibilities. Similarly, Gondolf (1988), in a study of
6,612 women leaving Texas shelters over an 18-month period, found that the best predictor
of returning to an abusive husband was the husband's having sought help or entering
counseling. These findings suggest that battered women may place their hope in the notion
that their husbands will change, although many batterers are not helped by therapy (see
Rosenfeld, 1992).
Summary. In summary, battered women appear to have many understandable reasons for
remaining with an abusive partner, including fear of further abuse, economic dependence,
commitment to one's spouse, and the belief that the partner will change. Yet, as noted above,
such findings must be placed in the context that many battered women do eventually leave
their abusers. Rather than continuing to narrowly examine why battered women stay, future
researchers should consider broader issues, including what factors help women to end the
abuse in their lives through any means, whether that includes leaving or other changes in the
relationship.
SUBTYPES OF BATTERED W O M E N
It is important to realize that battered women are not homogeneous groups. As reviewed here,
two typology studies have made it clear that samples of battered women are heterogeneous.
Snyder and Fruchtman (1981) identified five clusters of battered women based on inter-
views with 119 women entering a shelter. Women in Type One (28%) were rarely physically
abused. Their relationships with their abusers were stable, and they were the most likely to
return to the perpetrator after a brief stay in the shelter; they were the most likely to attribute
the violence to alcohol or other external variables. Conversely, Type Two women (33%) w e r e
victims of severe abuse, and were generally involved in "unstable, explosive" relationships
that involved multiple separations; they were least likely to return to the assailant. Type Three
women (19%) endured the most extensive and chronic abuse; at follow-up, they were
unlikely to be living with the batterer. Type Four women (11%) were the least likely to be
abused; rather, they sought protection for their children, who were abused. Finally, the
women in Type Five (9%) had experienced much family of origin violence as well as abuse
by other assailants, and were hypothesized to view violence as a standard aspect of life. These
women usually returned to the perpetrator.
Follingstad, Laughlin, Polek, Rutledge, and Hause (1991) conducted a similar study, using
various sources for subject recruitment (e.g., women's shelters, stores, emergency rooms).
Results from interviews with 234 women revealed five clusters. Type One women (14%)
endured severe and frequent abuse. They were likely to have been victims of violence as
children, to make external attributions for the abuse, to be relatively satisfied with their
relationships, and to accept violence in their relationships. In contrast, Type Two women
(23%) endured the least frequent and least severe violence. Their abuse was short-term, and
the women were likely to stay in the violent relationship. Type Three women (22%)
encountered a moderate frequency of abuse. Many of these women also had been abused as
children, were likely to minimize the perpetrators' accountability for the violence, and
accepted the occurrence of violence in their relationships. Type Four women (38%) experi-
enced a high frequency of abuse of moderate severity. They were likely to have been abused
as children, but they did not excuse the violence, despite their propensity to remain in the
relationship. Type Five women (3%) were frequently abused, but the violence decreased over
time. These women tended to stay in the relationships because they had no other resources.
These typologies demonstrate that battered women vary with regard to the violence they
experience and other relevant variables (e.g., history of abuse, likelihood of returning to the
batterer). Understanding such differences in the experiences of battered women may help us
to design interventions geared to the needs of individual battered women. However, two
limitations of the current research must be noted. First, at this stage, we do not understand
the extent to which subtypes of women simply reflect differences in the types of batterers
women marry. For example, to what extent is the subgroup of severely battered women
simply women who happened to have married severely violent men? Second, it will be
important for future typology researchers to conduct longitudinal studies. As reviewed above,
leaving an abusive husband is often a lengthy process that evolves over time. Thus,
cross-sectional typology research that categorizes women based on whether or not they stay
with their partner may not accurately reflect the ultimate outcome for women in various
subgroups.
EFFECTS OF MARITAL VIOLENCE ON CHILDREN

While spousal physical aggression has obvious and direct negative consequences on spouses,
the children in maritally violent homes may also be victims. Based on the Straus et al. (1980)
national survey data, Carlson (1984) estimated that at least 3.3 million children live in
maritally violent homes. In this section, we review some of the research examining how these
children are affected by interparental aggression.
General Consequences
A series of research reviews on the effects of marital conflict (Davies & Cummings, 1994;
Emery, 1982; Emery, 1989; Grych & Fincham, 1990; Markman & Leonard, 1985; Rutter,
1994; Reid & Crisafulli, 1990) and marital violence (Fantuzzo & Lindquist, 1989; Jaffe,
Women and Chihtren 199
Wolfe, & Wilson, 1990; McDonald & Jouriles, 1991; Rosenberg, 1987; Rosenberg &
Rossman, 1990) on children have been written. We draw heavily from these reviews and refer
the reader to them for more detailed information.
Previous reviewers have generally concluded that children from maritally violent homes
suffer from a variety of problems. McDonald and Jouriles (1991) state that "controlled
research in this area, in general, converges with clinical reports and indicates that children of
mothers in shelters for battered women are at risk for developing behavior p r o b l e m s . . , the
percentage of children of maritally aggressive families reported to exhibit behavior problems
at clinical levels ranges from 25% to 70% [compared to] 10% to 20% of children in
non-clinic, community comparison samples" (p. 189). Reviewers note that the resulting
problems for children may not only include behavior problems, but also emotional distress
and impaired functioning in areas such as self-esteem, social competence, and social
problem-solving (e.g., Fantuzzo & Lindquist, 1989; Rosenberg & Rossman, 1990).
Factors Affecting Impact

While marital violence has a generally detrimental impact upon children, many variables
mediate the specific effects. These factors include dimensions of the conflict itself (e.g.,
intensity, frequency), child related variables (e.g., gender, age), and stressors associated with
marital aggression (e.g., child abuse, parental psychopathology).
Dimensions of Conflict
Marital distress versus marital violence. Since marital violence often co-occurs with marital
distress (see review in third paper of this series), researchers have tried to determine which
of these factors is primarily responsible for the negative consequences seen among children.
In general, reviewers concur that negative outcomes for children are primarily explained by
exposure to overt interparental conflict rather than marital distress or divorce (Emery, 1982;
Grych & Finchaml 1990). For example, Jouriles, Murphy, and O'Leary (1989) found that,
after controlling for marital distress, child's age and sex, and the interaction of child's age
and marital discord, marital aggression still contributed unique variance to the assessed child
problems. Rosenbaum and O'Leary (1981) found a trend for boys from maritally violent,
distressed homes to be more deviant on measures of conduct and personality disorders than
boys from nonviolent, but maritally distressed homes. Using maternal reports, Johnson and
O'Leary (1987) compared conduct-disordered and nonconduct-disordered girls; the parents
of the two groups of girls did not differ in marital satisfaction, but mothers of the conduct-
disordered girls scored higher on a measure of open marital conflict observed by the child.
While Porter and O'Leary (1980) found no correlations between marital distress and
children's problems, overt marital conflict occurring in front of the child was significantly
correlated with many behavior problems for boys.
Intensity o f conflict. Intense conflict, such as physical aggression, has a greater impact upon
children than milder conflict (Cummings, Zahn-Waxler, & Radke-Yarrow, 1981). Davies and
Cummings (1994) reviewed several studies and found that "among the various forms of
[marital] conflict, physical aggression carded the greatest risk for children in terms of
negative reactions and adjustment problems" (p. 394).
This has been demonstrated in studies using simulated adult conflicts to study children's
reactions. For example, Cummings, Vogel, Cummings, and El-Sheikh (1989) found that
scenarios depicting nonverbal, verbal, or verbal-physical anger between adult actors were all
perceived as negative and elicited negative emotions in children; however, the verbal-
physical expressions of anger were perceived as most negative, and responses of distress
were especially strong in children from maritally violent homes. Similarly, after presenting
boys with simulated marital conflicts, O'Brien, Margolin, John, and Krueger's (1991) found
that boys from physically aggressive homes showed more intense emotional and cognitive
reactions (e.g., avoidance of dealing with the conflict, reported attempts to intervene) than
boys from only verbally aggressive and low conflict homes. Also, Cummings, Pelligrini,
Notarius, and Cummings (1989) found that children from homes with interparental physical
aggression displayed more preoccupation (e.g., observing the interaction), concern and social
support seeking (e.g., signs of distress, expressions of concern, seeking comfort from the
mother), and attempts to comfort and protect the mother in response to observed conflict.
The importance of intensity of conflict has also been demonstrated in other types of
studies. For example, Fantuzzo, DePaola, Lambert, Martino, Anderson, and Sutton (1991)
found that exposure to interparental verbal conflict was correlated with moderate levels of
conduct problems, while exposure to both verbal and physical conflict was correlated with
clinical levels of conduct problems and moderate levels of emotional problems. In a study
measuring adolescents' satisfaction with their family relationships, researchers found that
satisfaction decreased and anger towards parents increased linearly as the level of familial
violence escalated from nonviolence to verbal violence to verbal and physical violence
(Martin, Schumm, Bugaighis, Jurich, & Bollman, 1987).
Frequency of conflict. Frequency of interparental conflict is an important aspect of conflict,

since repeated conflict "appears to sensitize children to conflict, and is associated with a
greater incidence of behavior problems" (Fincham & Osborne, 1993, p. 80). This variable is
important to understanding the effects of marital violence, since it is assumed that children
from violent homes are more frequently exposed to parental conflict.
Hanson, Saunders, and Kistner (1992) reported that "frequency of interparental conflict
was the most important predictor of depression, externalizing behavior problems, and
negative views towards marriage for the females" (p. 435). Similarly, Hershorn and Rosen-
baum (1985) found that increased exposure to marital discord and violence was related to
behavioral and emotional problems among children. Studying children in shelters for battered
women, O'Keefe (1994) discovered that "children showing more severe disturbance were
more likely to have witnessed a greater frequency/severity of marital violence" (p. 72).
Similarly, Cummings et al. (1989) divided a sample of children based on the frequency of
interparental fighting. Increased frequency of exposure increased the likelihood of negative
emotional reactions and active involvement in the conflict. Negative reactions to repeated
conflict were also found by Cummings, Iannotti, and Zahn-Waxler (1985), who examined
children's reactions to various types of interactions between adult actors; exposing children
to angry adult conflicts, they found that reactions of distress and aggression became more
pronounced with repeated exposure to conflict.
Resolution of conflict. The way in which conflicts are resolved is a mediator of children's
reactions to parental conflict and seemingly relevant to a discussion of marital violence, since
the evidence suggests that children in violent homes often do not observe successful conflict
resolution, given their parents' lack of communication and problem-solving skills.
Martin et al. (1987) found that adolescents from verbally violent or verbally and physically
violent families were less likely to report successful outcomes from attempts at conflict
resolution than adolescents from nonviolent families. O'Brien et al. (1991) compared
reactions to simulated parental conflict of boys from physically aggressive, verbally aggres-
sive, and nonaggressive homes. They found that boys from nonaggressive homes consistently
made more predictions of positive outcome and positive evaluations, and offered more
adaptive problem solving ideas than boys from the homes characterized by verbal or physical
aggression. Using simulated adult conflicts, Cummings et al. (1989) found that children
perceived unresolved anger more negatively and experienced greater anger and distress when
witnessing unresolved anger than when witnessing resolved anger. Cummings, Ballard,
El-Sheikh, and Lake (1991) extended this research by exposing children to videotaped
conflict interactions between adult actors with different degrees of resolution. They found that
children's negative reactions were inversely related to the degree to which the conflict was
resolved.
Child awareness o f conflict. The impact of marital conflict upon children may also be
mediated by whether the children are aware of the conflict or not. As noted above, the most
negative impact is from conflict that is openly hostile (e.g., marital violence) rather than from
covert dissatisfaction. While parents may believe that they are successfully shielding their
child from interparental aggression, it must be remembered that even if children do not
directly witness violence, they may overhear the altercation or see consequences (e.g.,
injuries or intimidation). In addition, there is little consistency between what children report
that they observe and what parents report that their children know (Jaffe, Wolfe, & Wilson,
1990; Rosenberg & Rosman, 1990). Even between parents, there is only fair agreement as to
whether the child either saw or heard interparental aggression (O'Brien, John, Margolin, &
Erel, 1994). Thus, children's level of awareness is rarely accurately assessed.
Child Related Variables

Gender. Marital distress and background stressors have generally been found to have
differential effects for boys and girls, namely, externalizing problems in boys and internal-
izing problems in girls (Emery, 1982; Reid & Crisafulli, 1990). The identical pattern is often
seen when examining children from maritally violent homes, as concluded by Fantuzzo and
Lindquist (1989) in their review. However, other reviewers suggest that there is a broader
impact, with both boys and girls manifesting internalizing and social competence problems,
while boys also display a greater degree of externalizing behaviors (McDonald & Jouriles,
1991).
Many researchers have examined samples consisting only of boys and have consistently
found detrimental effects of witnessing marital violence. For example, examining incarcer-
ated boys, Lewis, Shanok, Pincus, and Glaser (1979) found that the children who had a more
extensive history of aggressive acts were more likely to have experienced abuse and
witnessed spousal abuse. Similarly, Rosenbaum and O'Leary (1981) compared boys from
spouse abusive families to boys from control families; there was a trend for boys from
abusive families to be more deviant on factors such as conduct disorder, personality disorder,
inadequacy-immaturity, and subcultural delinquency. Also in a sample of boys, Hershorn and
Rosenbaum (1985) found that witnessing parental marital discord and violence correlated
with maternal reports of behavioral and emotional problems.
Most researchers have studied samples that included both boys and girls and compared the
effects of exposure to spouse abuse on boys and girls. A number of researchers have found
a higher likelihood of males manifesting externalizing behavior as a result of marital
violence, with females demonstrating more internalizing problems. For example, Carlson
(1990) studied adolescents in residential treatment agencies; adolescent males exposed to
marital aggression were significantly more likely than females to have run away and report
suicidal thoughts, and tended to be more likely to hit their mothers. Also studying an older
sample (i.e., college students), Silvern, Karyl, Waelde, Hodges, Starek, Heidt, and Min
(1995) found that retrospective reports of parental marital violence were more strongly
related to internalizing symptoms (i.e., depression, low self-esteem, trauma) for women than
men. Among a sample of children in battered women's shelters, Hughes and Barad (1983)
found that boys received higher behavior problem scores than girls from staff members and
mothers; girls received higher scores on scales such as anxiety and worry/oversensitivity,
based on children's self-report. Cummings, Iannotti, and Zahn-Waxler (1985) also found
differential gender effects when exposing children to angry verbal interactions of adult actors.
Girls manifested more distress, while boys were more aggressive following exposure.
Similarly, Cummings, Vogel, Cummings, and El-Sheikh (1989) found boys reported more
anger than girls in reaction to videotaped angry interactions between adult actors. Interest-
ingly, Jouriles and Norwood (1995) only found expected gender differences in externalizing
behavior among families experiencing more extreme, rather than less extreme, husband
violence.
Other research supports the notion that while boys may experience more externalizing
problems, both boys and girls experience internalizing problems and social competence
problems (McDonald & Jouriles, 1991). For example, using maternal reports, Jaffe, Wolfe,
Wilson, and Zak (1986) found that boys and girls from violent homes both showed
internalizing problems and lower levels of social competence, while the boys additionally
demonstrated externalizing behavior problems.
A few studies have, surprisingly, yielded opposite results, with girls manifesting greater
externalizing behaviors as a result of marital violence. For example, O'Keefe (1994) studied
children residing at shelters for battered women. She found that both boys and girls displayed
elevated scores on both externalizing and internalizing behavior problems, but the family
violence variables of witnessing parental aggression and experiencing mother-child aggres-
sion were better predictors of externalizing behavior in girls rather than boys. Also using a
sample of children in shelters for battered women, Christopoulos, Cohn, Shaw, Joyce,
Sullivan-Hanson, Kraft, and Emery (1987) found that both boys and girls showed elevated
internalizing and externalizing scores; however, boys from their demographically matched
(e.g., low income) community comparison group also were elevated on externalizing scores.
Examining a low income, multiple stressor group, Spaccareli, Sandier, and Roosa (1994)
used multiple regression to control for risk factors in an inner city sample of children and
found that spousal violence contributed significant unique variance to girls' conduct prob-
lems and showed a trend towards affecting girls' depression, as assessed by self-report.
Spousal violence did not contribute unique variance to boys' self-report of depression,
self-esteem, competence, or conduct problems as compared to the inner city controls. It may
be that in such highly stressful environments (e.g., poverty, exposure to inner city violence),
any unique effects attributable to marital aggression are difficult to detect, especially for boys.
As Christopoulos et al. (1987) assert, in such studies, the comparison group children "were
experiencing a number of family adversity factors that previously have been found to be
predictive of antisocial behavior in boys" (p. 616).
Age. Another variable that may mediate the effect of marital violence is the child's age. Age
may have a unique main effect, or may interact with gender.
Hughes and Barad (1983) examined children residing in a shelter for battered women. On
a self-report measure of self esteem, children in grades 4-7 had average scores compared to
appropriate norms, children in grades 1-3 obtained low average scores, and preschoolers
scored well below average. Porter and O'Leary (1980) found that overt marital conflict
observed by children correlated with measures of conduct disorder and total pathology in
boys 5-10 years old, while boys 11-16 scored higher on scales of personality disorder,
inadequacy-immaturity, socialized delinquency, and total pathology; no significant results
were found for girls in the two age groups. In a study of children of battered women residing
in a shelter, Davis and Carlson (1987) found that girls aged 6-11 had lower social compe-
tencies and higher behavior problems, aggression, depression, and somatic problems than
boys aged 6-11; the opposite pattern was found in preschool children, with boys manifesting
lower social competence, and higher behavior problems, aggression, depression, and somatic
complaints. In research examining the longterm effects of parental marital violence on grown
children, Silvern, Karyl, Waelde, Hodges, Starek, Heidt, and Min (1995) demonstrated that,
among college students, retrospective reports of parental physical aggression were signifi-
cantly related to current trauma symptoms for both men and women and to current levels of
depression and low self-esteem for women; however, the relationships were lessened once
child abuse and parental alcohol use were statistically controlled.
Based upon these studies, no clear conclusions can be drawn about the specific mediating
effects of age on child problems. However, it does appear that types of problems manifested
by children in violent homes will vary with age and developmental level and that there may
be a relationship between parental aggression and psychological symptoms even among
young adults.
Race. While this variable has rarely been directly investigated, some studies have found
differential effects for race. For example, Stagg, Wills, and Howell (1989) surveyed preschool
children entering a shelter for battered women with their mothers. Based upon maternal
report, all children had elevated behavior problems and impaired social competencies, but
white males demonstrated the highest level of impairment. Similarly, in a study of marital
violence, O'Keefe (1994) found that, based upon maternal reports, white children exhibited
higher levels of externalizing problems than African American children. At this point, further
research is needed before definitive conclusions can be reached.
Other Stressors
Marital violence is often found in combination with other stressors, such as shelter
residency, parental divorce, parent-child aggression, parental psychopathology, social isola-
tion, and low socioeconomic status. Researchers have rarely factored out these stressors to
assess the unique variance of marital violence.
A notable exception is a study by Spaccarelli et al. (1994). They studied inner-city children
and found that spouse abuse correlated with single-parent status, lower family income, lower
maternal education level, frequent family moves, parental divorce, parental drinking, and
father's incarceration. After controlling for all these risk variables, parental violence still
contributed unique variance to predicting girls' conduct problems, but not boys' adjustment.
In a study comparing children in shelters to children in violent homes, with comparable levels
of interparental aggression, and to a control group, Fantuzzo et al. (t991) found that the
children residing in shelters manifested higher internalizing problems, lower social compe-
tency, and lower perceived maternal acceptance than the home violence or control group.
These studies suggest that multiple stressors act in a cumulative fashion to impact child
adjustment.
Child abuse. Children, perhaps boys especially (Jouriles & LeCompte, 1990), of maritally
violent parents are at a higher risk for being abused themselves. Indeed, Fantuzzo and
Lindquist (1989) note that "Straus et al. (1980) found that there was a 129% greater chance
of child maltreatment in a home where conjugal violence was present" (p. 89).
In their review of the literature, Fantuzzo and Lindquist (1989) found that child abuse and
neglect were not assessed in 75% of the studies conducted to examine the effects of marital
violence on children. However, in studies that do assess both child abuse and witnessing
marital violence, researchers generally find that exposure to spousal abuse only or being a
victim of abuse only are both uniquely related to child behavior problems, and that the
combination of these factors results in even more serious consequences for children (e.g.,
Davis & Carlson, 1987; Jaffe, Wolfe, Wilson, & Zak, 1986; Miller, Handal, Gilner, & Cross,
1991; Hughes, Parkinson, & Vargo, 1989; O'Keefe, 1995; Silvern et al., 1995).
A few researchers have begun to examine the predictors of child abuse among families
experiencing marital violence. For example, Jouriles and LeCompte (1990) found that child
gender made a difference---husband violence correlated with fathers' and mothers' aggres-
sion toward sons but not daughters. In a within family comparison study, Jouriles and
Norwood (1995) studied 48 families with both a boy and a girl where the mother had sought
help at battered women's shelters. They found that in the families experiencing more
extreme, but not less extreme, marital violence, a gender difference emerged---parents
engaged in more aggression toward boys than toward girls; this gender difference was only
partially explained by the fact that the boys had more externalizing behavior problems than
the girls (i.e., parents are more likely to physically aggress against children who engage in
problematic externalizing behavior). Among a sample of 184 children living with their
mothers in a battered women's shelter, O' Keefe (1995) found that boys were more likely than
girls to be victims of parental aggression; significant predictors of parental aggression toward
children included higher levels of spousal violence and child aggression, lower levels of
marital satisfaction, and a poorer quality father-child relationship.
Maternal depression. Just as depression and marital distress are related, depression is a
frequent consequence for battered women (see review earlier in this paper). Thus, the
behavior problems seen in children of violent marriages could also be a consequence of
parental depression. After reviewing the relevant literature, Fincham and Osborne (1993)
concluded that "parental depression does not account for the correlation between marital
discord and child adjustment, [however] work by Rutter (1971) suggests that depression
facilitates or potentiates the relationship."
Fincham and Osborne (1993) reviewed studies that examine the effects of marital distress
and parental mental health. For example, Shaw and Emery (1987) found independent effects
of parental conflict and maternal depression on child behavior problems, while also finding
increased predictive ability when these two factors were considered simultaneously. Evalu-
ating the impact of marital distress and parental psychopathology on children's behavior,
Emery, Weintraub, and Neale (1982) found a relation between marital distress and parental
psychopathology, and also found that both factors affected children's school behavior. Smith
and Jenkins (1991) found that maternal mental health (e.g., depression and drinking habits)
affected children's behavior and, after controlling for maternal mental health, the researchers
still found an effect on children's behavior for marital distress. These studies generally
indicate both an independent and combined effect of maternal mental health and marital
distress on children's behavior.
Several mechanisms have been proposed to explain the association between marital
conflict and children's behavior problems. These mechanisms are more thoroughly discussed
by previous reviewers (Davies & Cummings, 1994; Emery, 1982, 1989; Fantuzzo &
Lindquist, 1989; Fincham, Grych, & Osborne, 1994; Grych & Fincham, 1990; Jaffe, Wolfe,
& Wilson, 1990; Markman & Leonard, 1985; McDonald & Jouriles, 1991).
A primary mechanism discussed is modeling, in which children directly learn aggressive
behavior and the attitude that violence is appropriate, but fail to learn prosocial behavior,
since they lack models of this. Alternatively, parental aggression may serve as a stressor that
results in behavior problems for the child, especially when compounded with such factors
such as maternal depression or social isolation. Multiple stressors may have a summative or
potentiating effect, leading to greater risk of child problems.
Other proposed mechanisms describe the role of the child's cognitive and emotional
processing of the marital violence in mediating his/her responses to marital aggression. As an
example, it has been proposed that the appraisals children make regarding marital conflict
determines child outcome. Children's appraisals for the conflict could involve fear for the
safety of the self or parent, fear of the loss of the parent, self-blame, or helplessness,
increasing the risk of child reactions of fear, anxiety, aggression, or low self-esteem.
Altematively, the behavior problems seen in the children from maritally violent homes may
be an active coping response to spousal abuse, possibly serving to express the child's
negative emotions, obtain reassurances, or reduce the marital conflict by diverting the
attention of the parents away from the marital dispute. The behavior problems may also be
a simple effect of increased negative emotional arousal in the child from exposure to spousal
abuse, increasing likelihood of conduct problems if s/he does not possess the ability to
regulate emotions.
As a specific example, Davies and Cummings (1994) posit an "emotional security
hypothesis." Children develop a sense of emotional security from the quality of parent-child
and interparental relations. This "emotional security is a paramount factor in children's
regulation of emotional arousal and organization and in their motivation to respond in the
face of marital conflict" (p. 388). As another example, consider the "cognitive-contextual
framework" model offered by Grych and Fincham (1990). They conceive of interparental
conflict "as a stressor that leads to an attempt by the child to understand and cope with the
conflict" (p. 278). In this model, "both cognition and affect serve appraisal functions and
guide the child's coping behavior" (p. 278).
A final set of mechanisms concerns the possible effects of the deterioration of parent-child
relations. Child abuse and/or negativity towards children is frequently found in maritally
violent homes, which may directly lead to behavior problems in the child. Also, high
interparental conflict may result in altered, inconsistent, or opposing child discipline, due to
stress on the parents or lack of agreement between spouses. Maritally violent couples may be
emotionally unavailable for their children, since "frequent conflict may be emotionally
draining to parents and thus reduce their ability to recognize and respond to their children's
emotional needs . . . . fathers in particular may tend to withdraw from their children when they
are dissatisfied with their marriages" (Fincham, Grych, & Osborne, 1994, p. 134). Finally,
abused mothers may suffer increased physical and psychological symptoms, which may
negatively affect the mother-child relationship (e.g., see above section on maternal depres-
sion).
Although most of these proposed mechanisms and models receive empirical support, based
upon researched correlations between child behavior problems and the various hypothesized
causal factors, many of these links are post hoc, and there is no definitive proof for any causal
pathways (Fincham, Grych, & Osborne, 1994).
Buffers
Consistent with the theory that multiple stressors compound to cause behavior problems in
children, researchers have posited factors that may serve as buffers mitigating the negative
repercussions of marital violence. According to Jaffe, Wolfe, and Wilson (1990, p. 73), "in
a review of the stressors of childhood, Garmezy (1983) found that protective factors of the
children could be divided into three categories: (a) dispositional attributes of the child (e.g.,
ability to adjust to new situations); (b) support within the family system (e.g., good
relationship with one parent); and (c) support figures outside the family system (e.g., peer,
relatives)." Indeed, reviewers agree that outside support and/or at least one goodparent-child
relationship help to protect the child from the stress of interparental conflict (Grych &
Fincham, 1990; Emery, 1982, 1989; Jaffe, Wolfe, & Wilson, 1990). For example, Burman,
John, and Margolin (1987), using mothers', fathers', and children's reports, suggest that a
warm parent-child relationship buffered the impact of distressed and conflictual marriages on
children's adjustment.
METHODOLOGICAL CONSIDERATIONS
Methodological limitations of past studies may influence interpretations that can be drawn
from the available research on husband violence. In this section we will briefly mention some
of the major problems in this literature.
Sampling issues are a problem in research examining the effects of marital violence on
women and children. For example, while researchers studying violent husbands now rou-
tinely include a maritally distressed but nonviolent comparison sample, most researchers
examining battered women have not, leaving unanswered the question of how the effects of
marital violence differ from the effects of being in a unhappy but nonviolent relationship. In
addition, many studies of battered women have included only battered women from shelters,
who may differ in many important ways (e.g., socioeconomic status, level of fear, ability to
leave) from battered women in the community. Similarly, findings gathered from samples of
battered women in treatment may not generalize to battered women who have not sought
help. On a positive note, researchers studying battered women are beginning to consider how
various dimensions of the battering (e.g., severity, frequency, whether or not the woman is
still in the relationship) are related to variables of interest. Most researchers studying children
from violent homes have studied children residing in battered women's shelters or clinic
populations; thus, their findings may not generalize to other samples or may exaggerate the
effects of marital conflict. In particular, children in shelters not only come from violent homes
but are now also exposed to a variety of new stressors, such as being removed from their
neighborhood and school and thus their normal sources of support.
There are also a series of problems with the measures used in many past studies. In an
improvement over early research, most researchers are now using standardized measures.
However, most researchers continue to rely almost exclusively on self-report questionnaire
measures, which have all of the problems associated with self-report and, in many cases,
retrospective reporting. As a specific example, consider the fact that the most frequently used
tool for assessing child problems is maternal report. Not only does this confound reporting
source (i.e., mother) for both the dependent and independent variables (i.e., marital violence;
child outcomes) but, in addition, several reviewers have suggested that abused mothers may
judge their children's behavior more negatively than other sources (Emery, 1982; Grych &
Fincham, 1990; Fantuzzo & Lindquist, 1989; McDonald & Jouriles, 1991 ; Reid & Crisafulli,
1990), which may exaggerate the correlations found between marital conflict and child
problems.
Another major problem with this research area is the fact that most researchers have used
correlational study designs. These studies simply do not provide any information about
causality or even temporal sequence. Until additional longitudinal studies are conducted, it
will be impossible to know which of the reviewed variables are simply correlates of marital
violence versus which are precursors or consequences of marital violence.
SUMMARY
Studies of how battered and nonbattered women differ have been controversial, since the
findings can easily be used to point to the deficits of battered women, implying that they are
to blame for their abuse. However, researchers in this area are increasingly sensitive to such
issues. For example, studies of posttraumatic stress disorder clearly link the psychological
distress found a m o n g battered w o m e n to the abuse they have suffered, relating information
about battered w o m e n to the more established trauma literature. As another example, findings
relating degree of s y m p t o m a t o l o g y to the a m o u n t of abuse experienced support that n o t i o n
that the problems of battered w o m e n are the consequence, not the cause, of violence.
O n the other hand, the correlational nature of the research does leave questions about
temporal sequence and causality u n a n s w e r e d and will continue to lead to theoretical debates
about these issues. It can be asked why all correlates of violence found a m o n g husbands are
assumed to be causes of violence, while m a n y of the same correlates of violence found
a m o n g battered w o m e n are assumed to be the consequences of violence. Future researchers
need to address such questions.
Researchers studying children are not asked to engage in such debates, as most of us
willingly accept a relatively unidirectional model positing that marital violence causes
problems a m o n g children rather than vice versa. The effects of marital violence a m o n g
children is a relatively new and undeveloped research area; yet clear and consistent findings
have emerged, demonstrating the detrimental effects that violence can have on children.
However, as discussed above, better methods must be used in future research in this area. In
addition, while m u c h interesting theoretical work has been done, researchers in this area must
continue to grapple with questions as to the relationship between marital distress, marital
conflict, and marital violence, relative to their effects on children. Is marital violence simply
an extreme form of marital distress and conflict, or does it u n i q u e l y impact children?
In both of these research areas, more work is needed that considers and tests the
m e c h a n i s m s by which violence has its effects. Related to this, we should give more attention
to the w o m e n and children who are relatively resilient and m a n a g e to survive under the
circumstances of living in an abusive home. Clearly, m a n y w o m e n and children are adversely
affected b y h u s b a n d violence. W h a t is less clear is how to buffer w o m e n and children from
these effects and how to help them recover from the trauma they have experienced.
REFERENCES
Aguilar, R. J., & Nightingale, N. N. (1994). The impact of specific battering experiences on the self esteem of abused
women. Journal of Family Violence, 9(1), 81-95.
Aguirre, B. E. (1985). Why do they return'? Abused women in shelters. Social Work, 30. 350-354.
Andrews, B. (1995). Bodily shame as a mediator between abusive experiences and depression. Journal of Abnormal
Psychology, 104, 277-285.
Andrews, B., & Brewin, C. R. (1990). Attributions of blame for marital violence: A study of antecedents and
consequences. Journal of Marriage and the Fami(v, 52, 757-767.
Apt, C., & Hurlbert, D. E (1993). The sexuality of women in physically abusive marriages: A comparative study.
Journal of Family Violence, 8(1), 57-69.
Astin, M. C., Lawrence, K. J., & Foy, D. W. (1993). Posttraumatic stress disorder among battered women: Risk and
resiliency factors. Violence and Victims, 8(1 ), 17-27.
Barksdale, L. S. (1972). Building self-esteem. Idyllwild, CA: Barks~ale Foundation.
Bading, J., O'Leary, K. D., Jouriles, E. N., Vivian, D., & MacEwen, K. E. (1987). Factor similarity of the Conflict
Tactics Scales across samples, spouses, and sites: Issues and implications. Journal of Family Violence, 2, 37-54.
Beck, A., Ward, C., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for measuring depression.
Archives of General Psychiato', 4, 561-571.
Berk, R. A., Newton, P. J., & Berk, S. E (1986). What a difference a day makes: An empirical study of shelters for
battered women. Journal of Marriage and the Family, 48, 481---490.
Bograd, M. (1988). How battered women and abusive men account for domestic violence: Excuses, justifications,
or explanations? In G. T. Hotaling, D. Finkelhor, J. T. Kirkpatrick, & M. A. Straus (Eds.), Coping with family
violence: Research and policy perspectives. Newbury Park, CA: Sage.
Browne, A. (1993). Violence against women by male partners prevalence, outcomes, and policy implications.
American Psychologist, 48, 1077-1087.
Burman, B., John, R. S., & Margolin, G. (1987). Effects of marital and parent-child relations on children's
adjustment. Journal of Family Psychology, 1, 91-108.
Campbell, J. (1989). A test of two explanatory models of women's responses to battering. Nursing Research, 38,
18-24.
Cantos, A. L., Neidig, E H., & O'Leary, K. D. (1993). Men and women's attributions of blame for domestic
violence. Journal of Family Violence, 8, 289-303.
Cappell, C., & Heiner, R. B. (1990). The intergenerational transmission of family aggression. Journal of Family
Violence. 5, 135-152.
Carlson, B. E. (1984). Children's observations of interpersonal violence. In A. Roberts (Ed.), Battered women and
their families (pp. 147-167). New York: Springer.
Carlson, B. E. (1990). Adolescent observers of marital violence. Journal of Family Violence, 5, 285-299.
Cascardi, M., & O'Leary, K. D. (1992). Depressive symptomatology, self-esteem, and self-blame in battered women.
Journal of Family Violence, 7, 249-25.
Christian, J.L., O'Leary, K.D., & Vivian, D. (1994). Depressive symptomatology in maritally discordant women and
men: The role of individual and relationship variables. Journal of Family Psychology, 8, 3242.
Christopoulos, C., Cohn, D. A., Shaw, D. S., Joyce, S., Sullivan-Hanson, J., Kraft, S. E, & Emery, R. E. (1987).
Children of abused women: Adjustment at time of shelter residence. Journal of Marriage and the Family, 49,
611~19.
Claerhout, S., Elder, J., & Janes, C. (1982). Problem-Solving skills of rural battered women. American Journal of
Community Psychology, 10, 605-612.
Coopersmith, S. (1975). Developing motivation in young children. Palo Alto Press: Consulting Psychologists Press.
Cummings, E. M., B allard, M., El-Sheikh, M., & Lake, M. (1991 ). Resolution and children's responses to interadult
anger. Developmental Psychology, 27, 462-470.
Cummings, E. M., Iannotti, R. J., & Zabn-Waxler, C. (1985). Influence of conflict between adults on the emotions
and aggression of young children. Developmental Psychology, 21,495-507.
Cummings, E. M., Vogel, D., Cummings, J. S., & El-Sheikh, M. (1989). Children's responses to different forms of
expression of anger between adults. Child Development, 60, 1392-1404.
Cummings, E. M., Zahn-Waxler, C., & Radke-Yarrow, M. (1981). Young children's responses to expressions of
anger and affection by others in the family. Child Development, 52, 1274-1282.
Cummings, J. S., Pellegrini, D. S., Notarius, C. I., & Cummings, E. M. (1989). Children's responses to angry adult
behavior as a function of marital distress and history of interparent hostility. Child Development, 60, 1035-1043.
Davies, E T., & Cummings, E. M. (1994). Marital conflict and child adjustment: An emotional security hypothesis.
Psychological Bulletin, 116, 387-411.
Davis, L. V., & Carlson, B. E. (1987). Observation of spouse abuse: What happens to the children? Journal of
Interpersonal Violence, 2, 278-291.
Derogatis, L. R. (1977). SCL-90: Administration, scoring, and procedural manual-1 for the R (revised) version.
Baltimore, MD: Johns Hopkins School of Medicine.
Doumas, D., Margolin, G., & John, R. (1994). The intergenerational transmission of aggression across three
generations. Journal of Family Violence, 9, 157-175.
Emery, R. E. (1982). Interparental conflict and the children of discord and divorce. Psychological Bulletin, 92,
310-330.
Emery, R. E. (1989). Family violence. American Psychologist, 44, 321-328.
Emery, R., Weintraub, S., & Neale, J. M. (1982). Effects of marital discord on the school behavior of children of
schizophrenic, affectively disordered, and normal parents. Journal of Abnormal Child Psychology, 10, 215-228.
Fairbank, J. A. (n.d.). Civilian M-PTSD. Research Triangle, NC: Research Triangle Institute.
Fantuzzo, J. W., & Lindquist, C. U. (1989). The effects of observing conjugal violence on children: A review and
analysis of research methodology. Journal of Family Violence, 4, 77-94.
Fantuzzo, J. W., DePaola, L. M., Lambert, L., Martino, T., Anderson, G., & Sutton, S. (1991). Effects of interparental
violence on the psychological adjustment and competencies of young children. Journal of Consulting and Clinical
Ferraro, K. J., & Johnson, J. M. (1983). How women experience battering: The process of victimization. Social
Problems, 30, 325-339.
Fincham, E D., & Osborne, L. N. (1993). Marital conflict and children: Retrospect and prospect. Clinical
Psychology Review, 13, 75-88.
Fincham, E D., Grych, J. H., & Osborne, L. N. (1994). Does marital conflict cause child maladjustment? Directions
and challenges for longitudinal research. Journal of Family Psychology, 8(2), 128-140.
Finn, J. (1985). The stresses and coping behavior of battered women. Social Casework: The Journal of Contem-
porary Social Work, 66(6), 341-349.
FoIlingstad, D. R., Laughlin, J. E., Polek, D. S., Rutledge, L. L., & Hause, E.S. (1991 ). Identification of patterns of
wife abuse. Journal of Interpersonal Violence, 6(2), 187-204.
Follingstad, D. R., Neckerman, A. E, & Vormbrock, J. (1988). Reactions to victimization and coping strategies of
battered women: The ties that bind. Clinical Psychological Review, 8, 373-390.
Follingstad, D. R., Rutledge, L. L., Berg, B. J., Hause. E. S., & Polek, D. S. (1990). The role of emotional abuse
in physically abusive relationships. Journal of Family Violence. 5(2), 107-120.
Foy, D. W., Sipprelle, R. C., Rueger, D. B., & Carroll, E. M. (1984). Etiology of posttraumatic stress disorder in
Vietnam veterans: Analysis of premilitary, military, and combat exposure influences. Journal of Consulting and
Clinical Psychology, 52, 79-87.
Frieze, I. H. (1978). Self Perceptions of Battered Women. Cited in Cohn and Giles-Sims.
Frieze, I. H. (1979). Perceptions of battered wives. In I. H. Frieze, D. Bar-Tal, & J. S. Carroll (Eds.), New approaches
to social problems: Applications of attribution theory (pp. 79-108). San Francisco, CA: Jossey-Bass, Inc.
Frieze, I. H. (1987). Perceptions of battering by battered women. Paper presented at the Third National Family
Violence Research Conference, July 1987, University of New Hampshire.
Frieze, I. H., Knoble, J., Zomnir, G., & Washburn, C. (1980). Types of battered women. Paper presented at the Annual
Research Conference of the Association for Women in Psychology, Santa Monica, California.
Frisch, M. B., & MacKenzie, C. J. ( 1991 ). A comparison of formally and chronically battered women on cognitive
and situational dimensions. Psychotherapy. 28, 339-344.
Garmezy, N. (1983). Stressors of childhood. In N. Garmezy & M. Rutter rEds.), Stress, coping, and development in
children (pp. 43-84). New York: McGraw-Hill.
Gelles, R. J. (1976). Abused wives: Why do they stay. Journal of Marriage and the Family, 38, 659-668.
Giles-Sims, J. (1983). Wife battering: A systems theorv approach. New York: Guilford.
Gleason, W. J. (1993). Mental disorders in battered women: An empirical study. Violence and Victims, 8, 53~56.
Gondolf, E. W. (1988). The effect of batterer counseling on shelter outcome. Journal of Interpersonal Violence, 3,
275-289.
Grych, J. H., & Fincham, E D. (1990). Marital conflict and children's adjustment: A cognitive-contextual
framework. Psychological Bulletin, 108(2), 297-290.
Hamilton, B., & Coates, J. (1993). Perceived helpfulness and use of"professional services by abused women. Journal
of Family Violence, 8, 313-323.
Hanson, R. E, Saunders, B. E., & Kistner, J. (1992). The Relation between dimensions of interparental conflict and
adjustment in college-age offspring. Journal of Interpersonal Violence, 7, 4352,53.
Herbert, T. B., Silver, R. C., & Ellard, J. H. (1991). Coping with an abusive relationship: I. How and why do women
stay? Journal of Marriage and the Family, 53, 311-325.
Hershorn, M., & Rosenbaum. A. (1985). Children of Marital Violence: A Closer look at the unintended victims.
American Journal of Orthopsychiatry, 55. 260-266.
Hilberman, E., & Munson, K. (1978). Sixty battered women. Victimology: An International Journal, 2(3-4),
460-470.
Hilbert, J. C., & Hilbert, H. C. (1984). Battered women leaving shelter: Which way do they go? A discriminant
function analysis. The Journal of Applied Social Sciences, 8, 291-297.
Houskamp, B. M., & Foy, D. W. (1991 ). The assessment of posttraumatic stress disorder in battered women. Journal
of Interpersonal Violence, 6, 367-375.
Holtzworth-Munroe, A. (1988). Causal attributions in marital violence: Theoretical and methodological issues.
Clinical Psychology Review, 8, 331-344.
Holtzworth-Munroe, A., Bates, L., Smutzler, N., & Sandin, E. (1995). A brief review of the research on husband
violence. Part 1: Maritally violent versus nonviolent men. Manuscript under review.
Holtzworth-Munroe, A., Waltz, J., Jacobson, N., Monaco, V., Fehrenbach, & Gottman, J. (1992). Recruiting
nonviolent men as control subjects for research on marital violence; How easily can it be done? Violence and
Victims, 7, 79-88.
Holtzworth-Munroe, A., Jacobson, N. S., Fehrenbach, P. A., & Fruzzetti, A. (1992). Violent married couples"
attributions for violent and nonviolent self and partner behaviors. Behavioral Assessment, 14, 53-64.
Horowitz, M. J., Wilner, N., & Alvarez, W. (1979). Impact of event scale: A measure of subject stress. Psychosomatic
Medicine, 41,209-218.
Hotaling, G., & Sugarman, D. (1986). An analysis of risk markers in husband to wife violence: The current state of
knowledge. Violence and Victims, 1, 101-124.
Hotaling, G., & Sugarman, D. (1990). A risk marker analysis of assaulted wives. Journal of Family Violence, 5,
1-13.
Hudson, W. W., & Mcintosh, S. (1981). The assessment of spouse abuse: Two quantifiable dimensions. Journal of
Marriage and the Family, 43, 873-885.
Hughes, H. M., & Barad, S. J. (1983). Psychological functioning of children in a battered women's clinic. American
Journal of Orthopsychiatry, 53, 525-531.
Hughes, H. M., Parkinson, D., & Vargo, M. (1989). Witnessing spouse abuse and experiencing physical abuse: A
"'double whammy"? Journal of Family Violence, 4, 197-209.
Infante, D. A., Chandler, T. A., Rudd, J. E., & Shannon, E. A. (1990). Verbal aggression in violent and nonviolent
marital disputes. Communication Quarterly, 38, 361-371.
Jaffe, P. G., Wolfe, D. A., & Wilson, S. K. (1990). Children of battered women. Sage Publications: Newbury Park,
Jaffe, P., Wolfe, D., Wilson, S. K., & Zak, L. (1986). Family violence and child adjustment: A comparative analysis
of girls' and boys' behavioral symptoms. American Journal of Psychiatry. 143, 74-77.
Johnson, P. L., & O'Leary, K. D. (1987). Parental behavior patterns and conduct disorders in girls. Journal of
Abnormal Child Psychology, 15, 573-581.
Jouriles, E. N., & LeCompte, S. H. (1990). Husbands" aggression toward wives and mothers' and fathers' aggression
toward children: Moderating effects of child gender. Journal of Consulting and Clinical Psychology, 59, 190-192.
Jouriles, E. N., Murphy, C. M., & O'Leary, K. D. (1989). Interspousal aggression, marital discord, and child
problems. Journal of Consulting and Clinical Psychology, 57, 4 5 3 4 5 5 .
Jouriles, E. N., & Norwood, W. D. (1995). Physical aggression toward boys and girls in families characterized by
the battering of women. Journal of Family Psychology, 9, 69-78.
Kalmuss, D., & Straus. M. A. (1990). Wife's marital dependency and wife abuse. In M. A. Straus & R. J. Gelles
(Eds.), Physical Violence in American Families: Risk Factors and adaptations to violence in 8,145families. New
Brunswick, NJ: Transaction Publishers.
Kemp, A., Green, B. L., Hovanitz, C., & Rawlings, E. I. (1995). Incidence and correlates of posttraumatic stress
disorder in battered women: Shelter and community samples. Journal of Interpersonal Violence, 10, 43-55.
Kemp, A., Rawlings, E., & Green, B. (1991). Post-traumatic stress disorder (PTSD) in battered women: A shelter
sample. Journal of Traumatic Stress, 4, 137-148.
Kortlath, M. J. (1979). Alcoholism in battered women: A report of advocacy services to clients in a detoxification
facility. Victimology: An International Journal, 4, 292-299.
Labell, L. S. (1979). Wife abuse: A sociological study of battered women and their mates. Victimology: An
International Journal 4, 258-267.
Langhinrichsen-Rohling, J., Neidig, P., & Thorn, G. (1995). Violent marriages: Gender differences in levels of
current violence and past abuse. Journal of Family Violence, 10, 159-175.
Launius, M. H.. & Jensen, B. L. (1987). Interpersonal problem-solving skills in battered, counseling, and control
women. Journal of Family Violence, 2, 151-162.
Launius, M. H., & Lindquist, C. U. (1988). Learned helplessness, external locus of control, and passivity in battered
women. Journal of hTterpersonal Violence, 3, 307-318.
Lewis, B. Y. (1985). The wife abuse inventory: A screening device for the identification of abused women. Social
Work, 30, 32-35.
Lewis, D. O., Shanok, S. S., Pincus, J. H., & Glaser, G. H. (1979). Loseke, D. R., & Cahill, S. E. (1984). The social
construction of deviance: Experts on battered women. Social Problems, 31, 296-310.
Malamuth, N. M., Linz, D., Heavey, C. L., Barnes, G., & Acker, M. (1995). Using the confluence model of sexual
aggression to predict men's conflict with women: A ten year follow-up study. Journal ofPersonali~ and Social
Malhotra, A. K. (1984). The relationship of attributions for physical abuse to learned helplessness in battered
women. Dissertations Abstracts International 44(11), 3533-B.
Markman, H. J., & Leonard, D. J. (1985). Marital Discord and Children at Risk. In W. K. Frankenburg (Ed.). Early
Identification of children at risk (pp. 59-77). New York: Plenum Publishing.
Martin, M. J., Schumm, W. R., Bugaighis, M. A., Jurich, A. P., & Bollman, S. R. (1987). Family Violence and
Adolescents" Perceptions of Outcomes of Family Conflict. Journal of Marriage and the Family, 49, 165-171.
McDonald, R., & Jouriles, E. N. (1991). Marital aggression and child behavior problems: Research findings,
mechanisms, and intervention strategies. The Behavior Therapist, 189-192.
McKeel, A. J., & Sporakowski, M. J. (1993). How shelters counselors' views about responsibility for wife abuse
relate to services they provide to battered women. Journal of Family Violence, 8, 101-111.
Miller, T. R., Handal, E J., Gilner, E H., & Cross, J. E (1991). The relationship of abuse and witnessing violence
on the Child Abuse Potential Inventory with Black adolescents. Journal of Family Violence, 6, 351-363.
Mills, T. (1984). Victimization and self-esteem: On equating husband abuse and wife abuse. Victimology: An
International Journal 9, 254-261.
Mitchell, R. E., & Hodson, C. A. (1983). Coping with domestic violence: Social support and psychological health
among battered women. American Journal of Communi~ Psychology, 11,629-654.
Morrison, R. L., Van Hasselt, V. B., & Bellack, A. S. (1987). Assessment of assertion and problem-solving skills in
wife abusers and their spouses. Journal of Family Violence. 2. 227-238.
Women and Children 21I
Murphy, C., & Cascardi, M. (1993). Psychological aggression and abuse in marriage. In R. L. Hampton, T. E
Gollotta, G. R. Adams, E. H. Potter, & R. P. Weissberg (Eds), Family violence: Prevention and treatment. Sage:
Newbury Park.
Murphy, C. M., & O'Leary, K. D. (1989). Psychological aggression predicts physical aggression in early marriage.
Journal of Consulting and Clinical Psychology, 57, 579-582.
Nurius, P. S., Furrey, J., & Berliner, L. (1992). Coping capacity anaong women with abusive partners. Violence and
Victims, 9, 229-243.
O'Brien, M., John, R. S., Margolin, G., & Erel, O. (1994). Reliability and diagnostic efficacy of parents' reports
regarding children's exposure to marital aggression. Violence and Victims, 9, 45-62.
O'Brien, M., Margolin, G., John, R. S., & Krueger, L. (1991). Mothers' and sons' cognitive and emotional reactions
to simulated marital and family conflict. Journal of' Consulting and Clinical Psychology, 59, 692-703.
O'Keefe, M. (1994). Linking marital violence, mother-child/father-child aggression, and child behavior problems.
O'Keefe, M. (1995). Predictors of child abuse in maritally violent families. Journal of Interpersonal Violence, 10,
3-25.
Okun, L. E. (1986). Woman abuse: Facts replacing myths. Albany, NY: State University of New York Press.
O'Leary, K. D., & Curley, A. D. (1986). Assertion and family violence: Correlates of spouse abuse. Journal of
Marital and Family Therapy, 281-289.
O'Leary, K. D., Curley, A., Rosenbaum, A., & Clarke, C. (1985). Assertion training for abused wives: A potentially
hazardous treatment. Journal of Marital and Family Therapy, 1I, 319-322.
O'Leary, K. D., Malone, J., & Tyree, A. (1994). Physical aggression in early marriage: Prerelationship and
relationship effects. Journal of Consulting and Clinical Psychology 62, 594-602.
Pagelow, M. D. (1981 ). Factors affecting women's decisions to leave violent relationships. Journal of Family Issues,
2, 391~-14.
Parker, B., & Schumacher, D. N. (1977). The battered wife syndrome and violence in the nuclear family of origin:
A controlled pilot study. Victimology: An International Journal, 67, 760-761.
Pence, E. (1989). Batterer programs: Shifting from community collusion to community confrontation. In Caesar, P.L.
& Hamberger, L.K. (Eds), Treating men who batter: Theory, practice and programs. New York: Springer.
Perilla, J.L., Bakeman, R., & Norris, F. H. (1994). Culture and domestic violence: The ecology of abused Latinas.
Violence and Victims, 9, 325-339.
Pfouts, J. H. (1978). Violent families: Coping responses of abused wives. Child Welfare, 57, 101-111.
Porter, C. A. (1983). Blame, depression and coping in battered women. Reported in Miller & Porter, 1983.
Porter, C.A. (1986). Coping and perceived control in battered women. Unpublished manuscript.
Porter, B., & O'Leary, K. D. (1980). Marital Discord and childhood behavior problems. Journal of abnormal Child
Prescott, S., & Letko, C. (1977). Battered women: A social psychological perspective. In M. Roy (Ed.), Battered
women: A psychological study of domestic violence (pp. 72-96). New York: Van Nostrand-Reinhold Company.
Radloff, L. S. (1977). The CES-D scale: A self-report depression scale for research in the general population. Applied
Psychological Measurement, 1,385--401.
Reid, W. J., & Crisafulli, A. (1990). Marital discord and child behavior problems: A meta-analysis. Journal of
Abnormal Child Psychology, 18. 105-117.
Reynolds, R., & Siegle, E. (1959). A study of casework with sado-masochistic marriage partners. Social Casework.
40, 545-551.
Riggs, D. S., Kilpatrick, D. G., & Resnick, H. S. (1992). Long-term psychological distress associated with marital
rape and aggravated assault: A comparison to other crime victims. Journal of Family Violence, 7, 283-296.
Robins, L., Helzer. J., Crouhan, J., & Ratcliff, K. (1981). The NIMH Diagnostic Interview Schedule. Archives of
General Psychiat~, 38, 381-389.
Rodenburg, F., & Fantuzzo, J. (1993). The measure of wife abuse: Steps toward the development of a comprehensive
assessment technique. Journal of Family Violence, 8, 203-228
Rosenbaum, A., & O'Leary, K. D. (1981). Children: The unintended victims of marital violence. American Journal
of Orthopsychiatry, 51,692-699.
Rosenberg, M. S. (1987). Children of battered women: The effects of witnessing violence on their social problem-
solving abilities. The Behavior Therapist, 4, 85-89.
Rosenberg, M. S., & Rossman, B. B. R. (1990). The child witness to marital violence. In Treatment of family
violence. New York: John Wiley & Sons.
Rosenfeld, B. D. (1992). Court-ordered treatment of spouse abuse. Clinical Psychology Review, 12, 205-226.
Rounsaville, B. J. (1978). Theories in marital violence: Evidence from a study of battered women. Victimology: An
International Journal, 3, 11-31.
Rutter, M. (1971 ). Parent-child separation. Psychological effects on the children. Journal of Child Psychology and
Psychiatry. 12, 233-260.
Rutter, M. (1994). Family discord and conduct disorder: Cause, consequence, or correlate? Journal of Family
Saltzman, L. E., & Mercy, J. A. (1993). Assaults between intimates: The range of relationships involved. In A.V.
Wilson (Ed.), Homicide: The victim~offender connection. Eastern Kentucky: Anderson Publishing Co.
Sato, R. A., & Heiby, E. M. (1991 ). Correlates of depressive symptoms among battered women. Journal of Family
Violence, 7, 229-245.
Saunders, D. G. (1994). Posttraumatic stress symptom profiles of battered women: A comparison of survivors in two
settings. Violence and Victims, 9, 31--43.
Schutte, N. S., Malouff, J. M., & Doyle, J. S. (1988). The relationship between characteristics of the victim,
persuasive techniques of the batterer, and returning to a battered relationship. The Journal of Social Psychology,
128, 605-610.
Schwartz, M. D. (1988). Marital status and women abuse theory. Journal of Family Violence, 3, 239-248.
Schwartz, M. D., & Mattley, C. L. (1993). The battered women scale and gender identities. Journal of Family
Violence, 8, 277-287.
Shaw, D. S., & Emery, R. E. (1987). Parental conflict and other correlates of the adjustment of school age children
whose parents have separated. Journal of Abnormal Child Psychology, 15, 269-281.
Shepard, M. E, & Campbell, J. A. (1992). The abusive behavior inventory A measure of psychological and physical
abuse. Journal of Interpersonal Violence, 7, 291-305.
Shields, N. M., & Hanneke, C. R. (1983). Attributions process in violent relationships: Perceptions of violent
husbands and their wives. Journal of Applied Social Psychology, 13, 515-527.
Silvern, L., Karyl, J., Waelde, L., Hodges, W. E, Starek, J., Heidt, E., & Min, K. (1995). Retrospective reports of
parental abuse: Relationships to depression, trauma symptoms and self-esteem among college students. Journal
of Family Violence, 10, 177-201.
Simons, R. L., Johnson, C., Beaman, J., & Conger, R. (1993). Explaining women's double jeopardy: Factors that
mediate the association between harsh treatment as a child and violence by a husband. Journal of Marriage and
the Family, 55, 713-723.
Sirles, E. A., Lipchik, E., & Kowalski, K. (1993). A consumer's perspective on domestic violence interventions.
Smith, M. D. (1990). Patriarchal ideology and wife beating: A test of a feminist hypothesis. Violence and Victims,
5(4), 257-273.
Smith, M. A., & Jenkins, J. M. (1991). The effects of marital disharmony on prepubertal children. Journal of
Abnormal Child Psychology, 19, 625-644.
Snell, J., Rosenwald, R., & Robey, A. (1964). The wife beater's wife: A study of family interaction. Archives of
General Psychiatry, 11, 107-113.
Snyder, D. K., & Fruchtman, L. A. (1981). Differential patterns of wife abuse: A data-based typology. Journal of
Consulting and Clinical Psychology, 49, 878-885.
Snyder, D. K., & Scheer, N. S. (1981). Predicting disposition following brief residence at a shelter for battered
women. American Journal of Communi~ Psychology, 9, 559-566.
Spaccarelli, S., Sandler, I. N., & Roosa, M. (1994). History of spouse violence against mother: Correlated risks and
unique effects in child mental health. Journal of Family Violence, 9, 79-98.
Spitzer, R., & Williams, J. B. S. (1985). Instruction manual for the Structured Clinical Interview for DSM-III-R
(SCID). New York: New York State Psychiatric Institute.
Stagg, V., Wills, G. D., & Howell, M. (1989). Psychopathology in early childhood witnesses of family violence.
Topics in Early Childhood Special Education, 9(2), 3-87.
Star, B. (1978). Comparing battered and non-battered women. Victimology: An International Journal 3, 32-44.
Sugarman, D. B., & Hotaling, G. T. (1989). Violent men in intimate relationships: An analysis of risk markers.
Journal of Applied Social Psychology, 19, 1034--1048.
Stets, J. E. (1990). Verbal and physical aggression in marriage. Journal of Marriage and the Family, 52, 501-514.
Stets, J.E., & Straus, M.A. (1990). Gender differences in reporting marital violence and its medical and psycho-
logical consequences. In M.A. Straus & R.J. Gelles (Eds.), Physical violence in American families: Risk factors
and adaptation to violence in 8,145families. New Brunswick, NJ: Transaction Publishers.
Stone, L.H. (1984). Shelters for battered women: A temporary escape from danger or the first step toward divorce.
Victimology: An International Journal 9, 284--289.
Straus, M. A. (1979). Measuring intrafamily conflict and violence: The Conflict Tactics (CT) Scales. Journal of
Straus, M. A.. & Gelles, R. J. (1986). Societal change and change in family violence from 1975 to 1985 as revealed
by two national surveys. Journal of Marriage and the Family, 48, 465-479.
Straus, M. A., & Sweet, S. (1992). Verbal/Symbolic aggression in couples: Incidence rates and relationships to
personality characteristics. Journal of Marriage and the Family 54, 346-357.
Straus, M. A., Gelles, R. J., & Steinmetz, S. K. (1980). Behind closed doors: Violence in the American family. Garden
City, NY: Doubleday Press.
Strube, M. J. (1988). The decision to leave an abusive relationship: Empirical evidence and theoretical issues.
Psychological Bulletin, 104, 236-250.
Strube, M. J., & Barbour, L. S. (1983). The decision to leave an abusive relationship: Economic dependence and
psychological commitment. Journal of Marriage and the Family. 45, 785-793.
Strube, M. J., & Barbour, L. S. (19841. Factors related to the decision to leave an abusive relationship. Journal oJ'
Sullivan, C. M., Basta, J., Tan, C., & Davidson III, W. S. (19921. After the crisis: A needs assessment of women
leaving a domestic violence shelter. Violence and Victims, 7. 267-275.
Tolman, R. M. (1989). The development of a measure of psychological maltreatment of women by their male
partners. Violence and Victims, 4(31, 159-177.
Vinton, L. (1992). Battered women's shelters and older women: The Florida experience. Journal of Family Violence,
7, 63-72.
Walker, L. E. (1979). The battered woman. New York: Harper & Row.
Walker, L. E. (19831. The battered woman syndrome study. In Finkelhor, D., Gelles, R. J., Hotaling, G. T.. & Straus,
M. A. (Eds.), The dark side of families: Current family violence research. Sage: Beverly Hills.
Walker, L. E. (1991). Post-traumatic stress disorder in women: Diagnosis and treatment of battered women
syndrome. Psychotherapy; 28, 21-29.
Warren, J., & Lanning, W. (19921. Sex role beliefs, control, and social isolation of battered women. Journal of
Family Violence, 1-8.
Weaver, T. L., & Clum, G. A. (1995). Psychological distress associated with interpersonal violence: A meta-analysis.
Clinical Psychology Review, 15, 115-140.
Wilson, M. N., Baglioni, Jr., A. J., & Downing, D. (1989). Analyzing factors influencing readmission to a battered
women's shelter. Journal of Family Violence, 4, 275-283.
Wilson, M., & Daly, M. (1993). Spousal homicide risk and estrangement. Violence and Victims, 8, 3-16.

A Brief Review of The Research On Husband Violence: Amy Holtzworth-Munroe, Natalie Smutzler, and Elizabeth Sandin

Uploaded by

Copyright:

Available Formats

A Brief Review of The Research On Husband Violence: Amy Holtzworth-Munroe, Natalie Smutzler, and Elizabeth Sandin

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

A Brief Review of The Research On Husband Violence: Amy Holtzworth-Munroe, Natalie Smutzler, and Elizabeth Sandin

Uploaded by

Copyright:

Available Formats

Aggression and Violent Behavior. Vol. 2, No. 2, pp.

A BRIEF REVIEW OF THE RESEARCH

Amy Holtzworth-Munroe, Natalie Smutzler,

A B S T R A C T , This is the second article, in a series of three, reviewing currently available

BATTERED VERSUS NONBATTERED W O M E N

Correspondence should be addressed to A. Holtzworth-Munroe, Indiana University, Department of

Conclusions Regarding the Psychological Impact of Battering on Women

Dependency. While most theorists conceptualize wives' marital dependency in terms of

Summary of family of origin studies. To summarize, studies examining the relationship

Defining and Measuring PsychologicalAbuse

Prevalence of Psychological Abuse

The Impact of Psychological Aggression on Battered Women

The Relationship Between Physical and Psychological Aggression

Summary of Psychological Abuse

Is Leaving an Abusive Relationship a Safe Alternative?

Factors Contributing to a Woman's Decision to Leave or Stay

OTHER FACTORS AFFECTING WOMEN'S DECISION TO STAY OR LEAVE

EFFECTS OF MARITAL VIOLENCE ON CHILDREN

Factors Affecting Impact

Frequency of conflict. Frequency of interparental conflict is an important aspect of conflict,

Child Related Variables

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.