Citation
The effects on children of witnessing/experiencing violence in their family of origin

Material Information

Title:
The effects on children of witnessing/experiencing violence in their family of origin
Creator:
Higgins, Gilda Joan
Publication Date:
Language:
English
Physical Description:
vi, 60 leaves : ; 29 cm

Subjects

Subjects / Keywords:
Abused children ( lcsh )
Children of abused wives ( lcsh )
Family violence -- Longitudinal studies ( lcsh )
Abused children ( fast )
Children of abused wives ( fast )
Family violence ( fast )
Genre:
Longitudinal studies. ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )
Longitudinal studies ( fast )

Notes

Bibliography:
Includes bibliographical references (leaves 55-60).
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Arts, Sociology.
General Note:
Department of Sociology
Statement of Responsibility:
by Gilda Joan Higgins.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
30834651 ( OCLC )
ocm30834651
Classification:
LD1190.L66 1993m .H54 ( lcc )

Full Text
THE EFFECTS ON CHILDREN OF WITNESS'ING/EXPERIENCING
VIOLENCE IN THEIR FAMILY OF ORIGIN
by
Gilda Joan Higgins
B.A., Metropolitan State College, 1983
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts
Sociology
1993


This thesis for the Master of Arts
degree by
Gilda Joan Higgins
has been approved for the
Department of
Sociology
by
Cruz Torres
7L> *
Date


Higgins, Gilda Joan (M.A., Sociology)
The Effects on Children of Witnessing/Experiencing
Violence in Their Family of Origin.
Thesis directed by Professor Richard Holmes Anderson
ABSTRACT
Being physically abused seemd to train people in
family violence. So does growing up in a house where the
mother and father hit each other. Each seems to make its
own contribution to training the next generation in
violence. Does it follow that children who experienced
both kinds of violence when they were growing up would
be the most violent of all? To date child abuse and
spouse abuse have been considered as two separate social
problems, each with its own set of causes and
characteristics, and each with its own corresponding
solutions. For the most part there has been no attempt
to view both problems as germinating from one root
system or cause. Experts in each field of service or
research have focused on their own area of concern,
either child abuse or spouse abuse, and have rarely
viewed each other's domain as congruent.
Method and Procedures: Data for the analysis was
collected from the General Social Surveys, 1972-1987:
iii


July 1987, which were conducted for the National Data
Program for the Social Sciences at the National Opinion
Research Center, (NORC) University of Chicago. The
questions taken from the NORC data all pertain to
violence, and incorporated the.word "hit" into each of
the questions. I used cross-tabulations to determine
whether respondents who had experienced violence (i.e.,
were hit) were more likely to condone the use of hitting
as a means of resolving conflict.
Results: The results of the analysis did show that
the percentage of those respondents who had been hit and
answered "yes" to whether it was "ok" to hit in certain
situations, were higher than those who had not been hit.
The analysis also showed that males who had been hit had
a higher percentage of "yes" answers to "whether it was
"ok" to hit" questions than did females who were in the
same Category. This significance was slight, however
Conclusions: The conclusions that were drawn from
the analysis are that individuals who have experienced
violence are more likely to condone of its use as a
means of resolving conflict than those who have not.
iv


This abstract accurately represents the content of the
candidate's thesis. I recommend its publication.
Richard Holmes Anderson


CONTENTS
CHAPTER
1. STATEMENT OF THE PROBLEM ....................... 1
2. LITERATURE REVIEW .............................. 5
Initial Effects of Domestic Violence .......... 8
Short Term Effects ........................10
Long Term Effects ....................... .13
3. THEORETICAL PERSPECTIVES ...................... 18
Social Learning Theory ....................... 18
Research Hypothesis ...........................25
4. METHODOLOGY ................................... 26
Data Source .................................. 26
Operationalization of the Variables.......27
Operationalization of the Hypothesis.......28
5. ANALYSIS and FINDINGS.......................... 30
Attitudes Toward Violence by
Experience of ViolenceAll Years ............. 31
Attitudes Toward Violence by
Experience of Violence-1984-
Controlling for SEX .......................... 39
6. SUMMARY AND CONCLUSIONS ....................... 51
REFERENCES
55


CHAPTER 1
STATEMENT OF THE PROBLEM
Violence against wives has a long history. Husbands
have long been considered as having the "right" to
beat their wives and children. Until the mid-1970's,
it was believed that spouse abuse affected only a
small and relatively unique group, and therefore was
not considered to be a social problem (Davis, 1989).
When wife abuse was discovered to affect more than a
select few, the research on this phenomenon
concentrated on the major players involved, the
husband and wife, and did not focus on the unintended
victims of this violence: the children who witnessed/
experienced it.
"Daddy, stop hitting Mamma." Or "Daddy, please
don't hit me any more, I promise I'll be good." Drive
down any street in America, more than one household in
six has been the scene of a spouse striking his or her
partner. Three American households in five (which have
children living at home), have reverberated with the
sounds of parents hitting their children (Straus and
Gelles, 1980). What are the consequences of children
having to witness the constant abuse of one parent
(usually the mother) by the other parent, and having to
1


experience the physical abuse themselves?
Violence is a learned behavior. Children who have
witnessed abuse or have been abused themselves are 1,000
times more likely to abuse a spouse or child when they
become adults than children raised in homes without
violence (Straus, Gelles, and Steinmetz, 1980). One
wonders why, when so many have expressed concern about
violence in television, it has taken so long for people
to voice concern about the consequences of children
seeing or being the victims of violence in their own
homes. The conventional theory is that the more violence
a child sees on television, the more he or she tends to
be violent, or is at least tolerant of violence
(Zuckerman and Zuckerman, 1985). If this is the case,
imagine the consequences of millions of children growing
up using violence on each other, and on their children.
Researchers who have studied child abuse continue
to find that children who were abused often grow up to
be abusing parents (Bakan, 1971; Gil, 1970; Kempe, et
al., 1962; Steele and Pollock, 1974). Research on
murderers finds that killers experienced more frequent
and severe violence as children than their brothers who
did not go on to commit a homicide (Palmer, 1962).
Examinations of presidential assassins or would-be
2


assassins also find these individuals sharing common
histories of violent upbringing. In his diary, Arthur
Bremer, Governor George Wallace's would-be assassin,
wrote, "My mother must have thought I was a canoe, she
paddled me so much." Lee Harvey Oswald, Sirhan Sirhan,
and Charles Manson all experienced violent childhoods
(Button, 1973).
A study of violent inmates in San Quentin prison
found that 100 per cent of them experienced extreme
violence between the ages of one and ten (Maurer, 1976).
Psychologist Ralph Welsh (1976) claims that he has never
examined or talked with a violent juvenile delinquent
who did not come from an extremely violent background.
Moreover, Welsh claims that even if the extreme violence
ceases before the child is four years old, the child
still is likely to exhibit violent tendencies as a
juvenile.
Violence in the streets, violence in the schools,
assassinations, murders, assaults, wife abuse, child
abuseare they caused by violence on television,
violence in the movies, permissive upbringing? These
probably contribute something. But the evidence appears
to support the notion that our homes, and how we raise
our children are the main sources of our violent society
-3


(Kalmus, 1984).


CHAPTER 2
LITERATURE REVIEW
Being physically abused seems to train people in
family violence. So does growing up in a house where the
mother and father hit each other. Each seems to make its
own contribution to training the next generation of
violence. Does it seem logical that children who witness
abuse between parents and are themselves abused would
grow up to be the most violent of all.?
Straus, Gelles, and Steinmetz (1980) suggests there
is striking evidence for the idea of social heredity in
violencethat violence by parents begets violence in
the next generation. When one member of a couple had
experienced being hit as a child and observing his or
her parents hitting each other, there was a one in three
chance that at least one act of violence had occurred
during the year.
The Straus, Gelles, and Steinmetz (1980) survey
conducted with 2,143 families found that couples who did
not hit each other had the lowest rate of abusive
violence toward their children.
Some of the learning about violence in the family
occurs by example: Children see their parents hitting
each other.
_5


Some of the learning about violence occurs as a
result of being the victim of violence: The more
children are hit by their parents, the more likely they
are to hit each other (Straus et al., 1980).
When a child grows up in a home where parents use a
lot of physical punishment and also hit each other, the
chances of becoming a violent husband, wife, or parent
are greatest of all: about one out of every four people
who grew up in these most violent households use at
least some physical force on their spouses in any one
year (Straus, et al., 1980).
According to these same researchers, the effect of
growing up in a violent home is even more predictive of
child abuse: over one out of every four parents who grew
up in a violent household were violent enough to risk
seriously injuring a child.
"Double Whammy." The term has been mentioned on
several occasions (Hughes et al., 1989). It refers to
the children who are doubly victimized not only being
exposed to interparental violence, but also by being
physically abused by one or both parents. The overlap of
spouse and child abuse occurring in the same home has
been estimated at between 40 and 60 percent of all
violent families (Forsstrom-Cohen and Rosenbaum, 1985;
.6


Pascoe,Hebbert, Perl, and Loda, 1981; Rosenbaum and
O'Leary, 1981). Wallerstein and Kelly (1980, pg.16) note
that over 40 percent of the children in their study had
exceedingly poor relationships with their fathers, and
that "parents who treated their children poorly were
more likely to treat their marital partners badly as
well." In their studies of battered wives, Walker (1984)
found that 53 percent, Giles-Sims (1985) found that 63
percent, and Bowker, Arbitell, and McFerron (1985)
report that 70 percent of abusive husbands also abused
their children. Rosenbaum and O'Leary (1981) studied
both wives and their abusers, using discordant but
nonviolent couples and satisfactorily married couples as
controls. They found that almost 82 percent of the men
who, as children, observed interparental violence were
also physically abused by one or both parents.
About half of the doubly victimized children in the
Pagelow (1982) survey of battered women appeared to have
been caught in the middle of an interparental attack or
attacks; one of the children died as a result of a
beating by his father. Based on their studies of abuse
victims in medical settings, Stark and Flitcraft (1983)
say that children of battered mothers are twice as
likely to be abuse victims, and their abusers are three
7


times more likely to be their fathers or father
substitutes than in families of nonbattered mothers.
Initial Effects of Domestic Violence On Children
The child's response to a violent home varies from
child to child. Problems that have been documented
include severe mood swings, sleep disturbances, somatic
complaints and school phobia. Hughes (1988) has
identified gender specific behaviors. Boys exhibited
violent behaviors while girls were withdrawn and
passive.
Reports on initial effects of domestic violence,
almost entirely based on case histories and observations
by parents and persons in the helping professions,
provide evidence that young children are highly
traumatized and exhibit a wide range of somatic and
psychological symptoms (Alessi and Hearn, 1984; Pfouts,
Schopler, and Henley, 1982; Stullman, Schoenberger, and
Hanks, 1987). "It is well documented that an abuser's
violence toward his mate tends to cause a variety of
psychological problems for their children," conclude Sun
and Thomas (1987, pg.3), after their survey of the
literature. Mothers are often unaware of the
victimization of their children while they are still
.8


living with their abusive spouses. At a shelter, Pagelow
(1982) observed a two year old boy bite his mother's leg
to get her undivided attention. The woman returned to
her abusive husband because she needed a "good home""
for her children, insisting that her husband was a "
very good father" despite the extreme violence he
unleashed against her and her child. Younger children,
especially boys, were more likely to act out, becoming
disobedient, defiant and destructive, while girls were
more likely to become clinging and dependent (Hughes et
al., 1989).
Children suffer a variety of severe social,
behavioral, and physical complaints from the trauma,
shock, fear, and guilt (Hershorn and Rosenbaum, 1985;
Levine, 1975; Westra and Martin, 1984). After employing
a variety of measures, clinicians Hilberman and Munson
(1977-78) observed that preschool-aged children of
battered mothers are intensely fearful, scream and
resist going to bed, and identify nighttime with the
occurrence of violence. Their physical problems include
such sleep difficulties as insomnia, sleepwalking,
nightmares, and bed-wetting (Hilberman and Munson,
1977-78). Children who try to intervene in violent
episodes subject themselves to possible injury; other
-9


typical responses are "immobilized shocked staring, and
running away and hiding" (Davidson, 1978, p. 119).
Older children may develop more sophisticated
methods or dealing with trauma; their initial reactions
are not always as clearly observable or documentable.
Hughes, Cole, and Ito (1988) studied two groups of
children from violent and nonviolent homes. When
discussing the "abuse/witness" children who reported
relatively minor distress, they suggested that "both
children and mothers may try to minimize the situation:
children in terms of experiencing negative feelings, and
the women in relation to the occurrence of violence...
Children may deny the presence of negative affect, due
to desensitization to violence, denial of it, or to
putting up a good front" (Hughes, Cole, and Ito, 1988,
p. 10). There is no doubt about whether or not children
suffer from domestic violence, the real question is how
long the effects last.
Short Term Effects
Observations of children in shelters produce
similar findings to those of initial effects. As they
grow older, children from violent families feel guilty
about their inability to prevent the violence and
protect their mothers and themselves. Some react by
ID


losing respect for their apparently helpless mother and,
finally, by developing anger toward her. Boys often
become aggressive and disruptive, fighting with their
siblings and schoolmates. Girls may become clinging,
withdrawn, passive, and anxious (Davidson, 1978;
Pagelow, 1982; Pizzey, 1974; Stullman, Schoenberger, and
Hanks, 1987 ).
The largest body of empirical research focuses on
the short-term effects of domestic violence. Many recent
studies employed samples that consisted of groups of
children who were observers only, observer-victims, and
control groups of children who were from nonviolent
homes (Hughes, 1988; Hughes, Rau, Hampton, and
Sablatura, 1985; Jaffe, Wolfe, Wilson, and Zak, 1985,
1986a, 1986b; Wolfe, Jaffe, Wilson, and Zak, 1985).
These researchers almost unanimously reported findings
of physical, social, and emotional problems in children
who were observers of parental violence, and even more
negative effects in those who were doubly victimized.
Some of this research examined differences by
gender among child observers. A sizeable number of
reports specifically point to negative effects on boys
(Hershorn and Rosenbaum, 1985) For example, Emery and
O'Leary (1982) found that marital discord was most
11


strongly related to behavior problems in boys. While the
children of battered women were, in comparison with
the control group, rated significantly higher in
behavior problems and lower in social competence, Wolfe,
Jaffe, Wilson, and Zak (1985) report that 34 percent of
the boys from violent homes, compared to 20 percent of
the girls, fall within the clinical range of behavior
problems. The researchers compared the children from
violent homes who tested as "normally adjusted" to
children who had severe behavior problems and found that
"children falling in the clinical range were more likely
to have been exposed to a higher frequency and intensity
of physical violence, and their mothers reported
negative life events over the past year" (p 661).
Children categorized as abused or exposed to abuse
were compared to control children by Jaffe, Wolfe,
Wilson, and Zak (1986a, 1986b) who found that boys from
violent homes differed significantly from boys raised in
nonviolent homes and also showed significantly more
behavipr problems than girls from violent homes. These
researchers noted that both boys and girls suffer
negative effects. While the boys indicate problems of
social competency, internalizing, and externalizing, the
girls indicate similar social competency problems and
12


problems relating to internalizing such as depression
and anxiety. In other words, boys tend to "act out,"
whether or not they do so in socially approved ways,
while girls tend to hold in their anxiety, fears, and
despair.
Restricting their sample to seventy-seven children
from violent homes, Davis and Carlson also found that
"the majority of these children, both girls and boys,
are experiencing serious problems" (1987 p. 288), but
the extent of the problems differed by age. Preschool
boys were more negatively affected than preschool girls,
but older girls, those in school, showed more negative
effects than same-age boys. School-aged girls scored
higher than boys on all measures of physical and
behavioral problems, including aggression, and lower in
social competence. The authors say that "the results of
this study strongly suggest that some girls may
experience even more difficulty than their male peers"
(Davis and Carlson, 1987 p. 288). Although some of the
early studies focused on all-male samples (Rosenbaum and
O'Leary, 1981), these results strongly suggest that both
sexes suffer from domestic violence but that its effects
are demonstrated in different ways at different ages.
Long Term Effects. A much smaller portion of the
13


literature focuses on long-term effects, largely because
of the difficulty in gathering samples of young adults
who were child victims and the scarcity of longitudinal
samples. There are a few studies available, and most
agree that child observers have a higher incidence of
social and behavioral problems than nonvictimized
controls, and that observer/victim children have the
highest incidence of these problems.
Forsstrom-Cohen and Rosenbaum's (1985) search of
the literature revealed almost no studies on long-term
effects of domestic violence on children. Influenced by
two previous studies, they hypothesized that problems
with anxiety, depression, and aggression are more common
among young adults who had observed parental violence.
Much of the literature dealing with long-term
effects consists of observations, theoretical pieces,
and retrospective studies that use principals of social
learning theory to support contentions that child
victims' same-sex role models taught them dysfunctional
patterns of behavior that are likely to be repeated in
their adult lives. Wallerstein (1988) reported her
observation that 40 percent of the children in her study
who saw marital violence as preschool children were
again involved in violence, either as victims or
14


perpetrators, when they became adolescents or young
adults.
Those children exposed to violence accept the
batterers as role models (Rosenbaum and O'Leary, 1981).
Children and young adults imitate the behavior of
aggressive models. Clinical reports indicate the
tendency of male children of abused women to act out
aggressively, frequently directing their aggression
toward the mother. The husband may provide a violent
role model for the male child, while exposure to an
abusive marital relationship may help the female view
violence toward women as normative (Rosenbaum and
O'Leary, 1981).
Straus, Gelles, and Steinmetz (1980) in a study
with American parents from intact families, found all
types of family violence to be positively correlated.
The study also revealed that when there is violence
between parents, there is likely to be violence between
other family members. Children of maritally abusive
couples mature into the next generation of abusive
husbands and abusive wives. There appears to be a
strong, positive relationship between spouse abuse and
child abuse. Eighty-two percent of husbands who had
witnessed parental spouse abuse were also victims of
15


child abuse at the hands of one or both parents
(Rosenbaum and O'Leary, 1981).
In a study with battered women, forty-three percent
of the women with children reported that their husbands
also abused the children (Prescott and Letko, 1977). One
sample of 44 battered women indicated that almost
one-fourth were beaten during pregnancy. Children of
violent marriages may also be viewed at-risk for adult
involvement as perpetrators or victims in family
violence.
The National Center on Child Abuse and Neglect
conducted a study in 1981 surveying five shelters.
Results indicate most of the women and children in the
shelter were from homes where violence and emotional
abuse were frequent and had persisted over several
years. Children in these violent homes not only
witnessed violence toward their mothers, but also
experienced violence directed toward themselves. Seventy
percent of the children had been abused or neglected.
Three groups are involved in this viscious cycle of
violence: the batterer, the mother, and the child.
As did others (Dobash and Dobash, 1979; Stark and
Flitcraft, 1981, 1983), Pagelow (1981a, 1981b) expressed
these ideas, but with a twist: They also argued against
16



a nondiscriminatory "cycle of violence" idea that had
great popular acceptance. Pagelow (1984) pointed out
that many people were combining two distinct phenomena
of childhood exposure to violencebeing an observer and
an observer-victim of parental violenceinto one, and
warned that these may lead to different adulthood
reactions strongly influenced by the sex of the child
and the sex of the violent parent. Studies of battered
women have found either no or very weak evidence that
little girls from violent homes grow up to become
battered wives or spouse abusers.
For example, the Pagelow study (1981a) found that
only about one fourth of the battered wives had observed
domestic violence as children, whereas more than half of
their abusers had observed their fathers beating their
mothers. In addition, 47 percent of the men had been
severely physically punished, in most cases by their
fathers or father figures. Many other studies found
strong evidence that a disproportionate percentage of
little boys from violent homes grow up to become abusive
husbands (Bowker, 1988; Dobash and Dobash, 1979; Walker,
1984).
17


CHAPTER 3
THEORETICAL PERSPECTIVES
Social Learning Theory
"I saw my Dad hit my Mom all the time. Whenever he
yelled my Mom and sisters would wait on him hand and
foot. That's how it is in my house,I yell, they jump.
I've been doing this all my life. I don't take no stuff
from nobody."
These statements are not unusual; most of us have
heard expressions similar to them sometime in our own
lives. The statement includes both imitating a high
status role model (his father) and reinforcement for his
bullying. This principle illustrates that behavior which
receives reinforcement is most likely to be repeated.
Social learning theory is defined as an integration of
differential association with differential
reinforcement, so that people with whom one interacts
are the reinforces that result in learning of deviant
and nondeviant behavior (Akers, 1973). According to
sociologist Ronald Akers,(1973) the type of behavior
that is consistently and most frequently reinforced by
others is the one most often exhibited.
Social learning theory has been widely proposed as
an explanation for spouse abuse but to a much lesser
18


degree for other types of family violence. One of the
basic propositions of social learning theory is that
reinforcement following behavior increases the
probability that the behavior will be repeated, and
another is that intermittently reinforced behavior is
the most difficult to extinguish (Hilgard and Bower
1966). Akers (1973) points out that two major parts of
the learning process are reinforcement and punishment.
Emphasis is placed on reinforcement rather than
punishment. When children observe aggressive behavior
being (in their minds) rewarded, they feel that is it
"ok" behavior.
In addition to reinforcement, modeling is another
important factor for learning, especially for children.
In a variety of social psychological tests, children
observing aggressive behavior not only showed they
remembered aggressive acts, but closely imitated them,
particularly When the acts were performed by a male
model (Bandura, 1973). The boys were more likely to
imitate aggression, spontaneously, but when they were
rewarded for imitation, children of both sexes increased
their performance until they were almost equally
aggressive (Bandura, 1973). As Bandura stresses, "people
learn more than they usually perform, unless given
19


positive incentives to do so" (Bandura, 1973. p. 160).
And the girls revealed they had learned but were
unwilling to demonstrate it until they were sure their
behavior would be rewarded.
Earliest socialization occurs in the home, thus
children receive reinforcement and/or punishment from
their adult models, and while they later learn
sex-appropriate behavior from many other social sources,
studies have shown that "children were much more
inclined to imitate a familiar aggressive model than an
unfamiliar one. This was especially true of boys who
performed approximately three times as many matching
responses as the girls..."(Bandura, 1973, p. 160).
According to Bandura's test findings, it made no
difference to either boys or girls if they had a
nurturant relationship with the model or not. In other
words, children, especially boys, are likely to imitate
behavior of a male model they know, but do not
necessarily like.
Other research findings that support social
learning theory are that if a models' behavior appears
to have functional valueif it achieves desired
resultsobservers have strong incentives to practice
these behavior patterns; and the more opportunity for
20


practice, whether physically or mentally rehearsed, the
more permanent the behavior patterns become (Bandura,
1973). If a young boy sees his father demand, and
receive, choice foods and personal service from others
in the family by force or threat of force, that kind of
behavior is likely to be perceived as functional. Later,
the boy practices his father's behavior with his younger
sibling or other children in the school yard. If that
behavior is reinforced, it proves functional for him, so
he is very likely to practice that behavior until it
becomes a permanent pattern. On the other hand, if his
sibling is bigger than he is, or he is smaller than his
target in the school yard, his behavior is unlikely to
be rewarded and is much less likely to be repeated.
There is one other possibility; the child may only
mentally rehearse, saying to himself, "When I get big
like Dad, people will give me what I want, or else!"
This child may grow up to be a man who never hits anyone
until he beats his wife.
Another important factor should be noted: attention
in the learning process. The first formulation of ideas
about appropriate or inappropriate behavior patterns is
established in the parental home while a young child is
in a relatively closed social system with fewer
21


distractions that impede learning, so that reinforcement
and models receive a major share of a child's attention.
Bandura had this to say:
The people with whom a person
regularly associated delimit the types of
behavior that he will repeatedly observe and
hence learn most thoroughly... The behavior of
models who possess high status in prestige,
power and competence hierarchies is more
likely to be successful and therefore to
command greater attention from others than
the behavior of models who are socially,
occupationally, and intellectually inept.
(1973, p. 175).
Children can mature in non-violent homes but become
violent adults, and this frequently occurs when their
reference groups approve of or use violence to achieve
their goals: their peers are high status models in their
eyes. This explains why children are more likely to
model their behavior after their fathers than their
mothers: In the hierarchical structure of the
patriarchal family, the husband/father position carries
the highest status in prestige and power (Dobash and
Dobash 1976,1978). Both boys and girls are more likely
to imitate a violent father than a non-violent mother,
especially if she is a victim of abuse, but the behavior
is less likely to be reinforced for girls when they
practice behavior that is considered sex-inappropriate.
Boys who are aggressive are much more likely to be
22


rewarded for such behavior, especially when they live in
an environment where traditional ideal of masculinity
include macho stereotypes (Bandura and Walters 1970).
It is true, that socialization of boys and girls
into adulthood contains many messages of sex-appropriate
behavior and that it is not rewarding, and sometimes
painful to deviate from cultural norms (Grossman 1977).
Little girls (and women) often witness modeled
aggression that is rewarding for others, but it takes
special inducements for them to act contrary to sex role
expectations. Bandura reports the results of his
experiments where children watched aggressive adult
models, and the different responses of boys and girls in
these tests:
Boys, who are generally encouraged to emulate feats
of physical prowess, spontaneously performed all they
had learned when they saw aggression well received. When
models were punished for their aggressive acts, boys
performed less than they had learned, but they later
readily exhibited additional imitative responses when
they produced rewarding results. By contrast, girls, for
whom physical aggression is traditionally regarded as
sex-inappropriate and hence negatively sanctioned, kept
much of what they had learned to themselves, regardless
23


of how the male models' behavior was treated. Their
learning was not manifested in action until they
received direct assurance that is was acceptable. In
predicting the occurrence of aggression, one should be
more concerned with predisposing conditions than with
predisposed individuals. Given that aggressive models of
conduct have been learned, social circumstances largely
determined whether and when they will be performed.
(Bandura, 1973)
Many authors who propose that stereotypical gender
role socialization along extremes of "masculinity" and
"femininity" is an important contributing factor in
family violence, particularly wife beating. Strauss
(1976) writes about cultural norms that legitimize
marital violence and discusses "compulsive masculinity",
which refers to the requirement that boys and men prove
to others that they are "real men" by their disdain for
any attributes considered "feminine." Pagelow (1978) did
a study which found that strict interpretation of
dichotomized sex roles literally set up men to be
aggressive (and if necessary, violent ), and women to be
passive, dependent, and frequently victimized by men.
But not all men or women follow sharply defined gender
roles, nor accept them as their definitions of manliness
24


or womanliness, so how can social learning theory
explain behavior of people who reject them? Despite
learning and social pressures that promote stereotypic
sex-role behavior, some men and women obviously see
these roles as dysfunctional. Some may have tried these
behavior patterns but found they were not rewarded or
that the behavior was not personally rewarding.
The single most serious criticism of social
learning theory is that one of its basic assumptions
depends on the human cognitive processes, on the idea
that people usually act in ways that they perceive to
serve their own best interests. Social learning theory
assumes rationality and some degree of thinking
concerning the costs/benefits ratio involved in the
action. It does not fully account for some types of
spontaneous, non-premeditated behavior that is
unrewarded and even self-punishing.
Research Hypothesis
From the previous review of the literature, the
following hypothesis was constructed for analysis in the
investigation: Those who were raised in an environment
where violence was implemented are more likely to
condone hitting (i.e., violence) as a means of conflict
and problem resolution.
25


CHAPTER 4
METHODOLOGY
Data Source
Data for the analysis was collected from the
General Social Surveys, 1972-1987: July 1987, which was
conducted for the National Data Program for the Social
Sciences at the National Opinion Research Center (NORC),
University of Chicago.
The General Social Surveys have been conducted
during February, March and April of 1972, 1973, 1974,
1975, 1976, 1977, 1978, 1980, 1982, 1983, 1984, 1985,
1986, and 1987. There are a total of 21,875 completed
interviews. Each survey is an independently drawn sample
of English-speaking persons 18 years of age or over,
living in non-institutional arrangements within the
United States.
The questions used from the NORC data all pertain
to violence. I hypothesized that those individuals who
had experienced violence (i.e., hitting) in their
families of origin, were more likely to approve of
hitting as a means of conflict resolution. Therefore,
the questions used for the analysis all had the word
"hit" incorporated into the variables. Between 1972 and
1987, the nine questions which had "hit" as a variable
26


were not asked each year. In 1987, the variables
"HITMARCH" AND "HITROBBR" were not included in the
questions on violence. In 1972, 1974, 1977, 1980, and
1985, no variables on hitting appeared.
Operationalization of the Variables
Before the analysis of the data can be undertaken,
each variable must be defined and operationalized as to
its use in this investigation. There were six dependent
variables used in this investigation. The first one was
"HITOK." Respondents were asked: "Is there ever a
situation in which you would approve of a man punching
an adult male stranger?" The second dependent variable
was "HITMARCH." Respondents were asked: "Would you
approve of hitting if the stranger was in a protest
march and was showing opposition to the other man's
views?" The third dependent variable was "HITDRUNK."
Respondents were asked: "Would you approve of hitting if
the stranger was drunk and bumped into the man and his
wife on the street?" The fourth dependent variable was
"HITCHILD." Respondents were asked: "Would you approve
of a man punching a stranger who had hit the man's child
after the child accidentally damaged the stranger,s
car?" The fifth dependent variable was "HITBEATR."
Respondents were asked: "Would you approve of a man
27


punching an adult male stranger who was beating up a
woman and the man saw it?" The sixth dependent variable
was "HITROBBR." Respondents were asked: "Would you
approve of a man punching an adult male stranger who had
broken into the man's house?"
There were three independent variables used for the
purpose of the investigation. The first independent
variable was "HIT." Respondents were asked: "Have you
ever been punched or beaten by another person?" The
second independent variable was "HITAGE." Respondents
were asked: "Did this happen to you as a child or adult?
The third independent variable was "HITNUM." Respondents
were asked:: "How many times would you guess you had
been hit?"
The analysis consisted of using all years that the
"HIT" variables were available between 1972 and 1987. In
addition to the aforementioned dependent and independent
variables, I used "SEX" as a controlling variable only
for the year 1984, as that was the latest year in the
NORC survey where all the "HIT" variables were used.
Operationalization of the Hypothesis
The variables used in the hypothesis are "being
abused" ( experience of violence) and condoning violence
(attitudes toward violence). The experience of
28



violence" for the purposes of this study will be the
independent variables of "HIT, HITAGE, and HITNUM." The
"attitudes toward violence" are represented by the
dependent variables of "HITDRUNK, HITCHILD, HITBEATR,
and HITROBBR."
Finally, sex was used as a control variable to
ascertain if gender had any bearing on attitudes toward
violence. I was mainly interested in the "HITOK to
HITROBBR by HIT to HITNUM" relationship, but I suspected
that sex might influence that relationship. Again, this
was done for the year of 1984 only, and the results were
not that significant.
29


CHAPTER 5
ANALYSIS AND FINDINGS
The analysis is divided into two major sections.
First, a discussion of the data. The information on all
the years will be presented first, followed by the
information for 1984 only, which included the control
variable, "SEX." The second section will be a brief
analysis to determine if the research hypothesis was
supported. It was easier to define the attitudes
(dependent variables) by the actual experience
(independent variables), rather than list each variable
more than necessary, I will be referring to them as
dependent and independent variables throughout this
study.
The data was analyzed using a cross tabulation. The
level of significance was established at <.05. One thing
should be noted here, my goal was to find out how likely
it was that a person who was hit would condone hitting.
Therefore, I concentrated my analysis only on the
respondents who thought it was appropriate to"HIT".
Attitudes Toward Violence by Experience of ViolenceAll
Years.
The first cross-tabulation was HITOK (dependent
variable) by HIT (independent variable). (See Table
30


2.1). As might be expected, a large majority of
respondents (73.3 percent) who had been punched or
beaten said it is ok to hit an adult male stranger.
However, not expected, was the fact that 63 percent of
the respondents who had never been punched or beaten
said that it was ok to hit an adult male stranger. The
significance level for this was .000.
The second cross-tabulation was HITOK (dependent
variable) by HITA6E (independent variable). Almost
seventy-four percent (73.9) of the respondents who had
been hit as a child said" yes", it was "ok" to hit an
adult male stranger in certain situations. Almost
seventy percent (69.4) of the respondents who had been
hit as an adult said "yes" to this question. As
expected, the largest percentage (78.3) to answer
affirmatively was the group of respondents who had been
beaten as an adult and a child. The significance level
was greater than .001.
The third cross-tabulation was HITOK (dependent
variable) by HITNUM (independent variable). As expected,
76.5 percent of the respondents who had been hit the
most (4 or more) said it is "ok" to hit a stranger.
Sixty-nine percent of the respondents who had been hit
once said "yes," it is ok to hit a stranger, and 73.8
31


Table 2.1
Attitude Toward Violence by Experience
of Violence
Experience of Violence
Percent Saying Yes: Have Been Hit Have Not Been Hit Total
Ok to hit in any situation? 73.3 (N=3943) 63.0 (N=6112) 66.7* (N=10055)
Ok to hit a protest marcher? 3.6 (N=192) 3.4 (N=317) 3.5 (N-509)
Ok to hit a drunk? 7.6 (N=399) 9.3 (N=865) 8.7* (N=1264)
Ok to hit a child if damaging car? 62.5 (N=3252) 50.9 (N=4664) 55.1* (N=7916)
Ok to hit if beating up woman? 85.9 (N=4422) 83.8 (N=7678) 84.5* (N=12100)
Ok to hit if robbing house? 88.1 (N=,4638) 83.6 (N=7797) 85.3* (N=12435)
*Significance: P=<.05
32


Table 2.1 Continued
Experience of Violence
Percent Saying Yes: Hit Once Hit 2-3 Times Hit 44- Times Total
Ok to hit in any situation? 69.3 (N=515) 73.8 (N=873) 76.5 (N=1090) 73 9* (N=2478)
Ok to hit a protest marcher? 3.8 (N=27) 3.3 (N=37) 4.7 (N=65) 4.0 (N=129)
Ok to hit a drunk? 8.5 (N=61) 7.0 (N=78) 7.9 (N=107) 7.7 (N=246)
Ok to hit a child if damaging car? 59.2 (N=419) 59.0 (N=655) 62.9 (N=846) 60.7 (N=1920)
Ok to hit if beating up woman? 87.4 (N=619) 85.1 (N=941) 85.2 (N=1129) 85.7 (N=2694)
Ok to hit if robbing house? 86.1 (N=619) 88.8 (N=1002) 88.0 (N=1198) 87.8 (N=2819)
*Significance:P=<.05
3-3


Table 2.1 Continued
Experience of Violence
Percent Saying Yes: Hit as Child Hit as Adult Hit as Both Total
Ok to hit in any situation? 73 9 (N=1685) 69.4 (N=1288) 78.3 (N=954) 73.3* (N=3927)
Ok to hit a protest marcher? 2.7 (N=61) 3.9 (N=71) 5.0 (N=60) 3.6* (N=192)
Ok to hit a drunk? 6.0 (N=133) filise, 9.0 (N=106) 7.6* (N=397)
Ok to hit a child if damaging car? 63.1 (N=1383) 59.5 (N=1078) 65.6 (N=774) 62.4* (N=3235)
Ok to hit if beating up woman? 86.7 (N=1892) 85.3 (N=1530) 85.0 (N=979) 85.8 (N=4401)
Ok to hit if robbing house? 87.8 (N=1958) 86.9 (N=1580) 90.6 (N=1083) 88.1 (N=4621)
*Significance:P=<.05
34


percent of the respondents who had been hit 2 to 3 times
said "yes". The assumption here is that the more an
individual is hit, the more that individual considers
hitting as appropriate behavior. The significance level
for this was .001.
The next six cross-tabulations have HITMARCH and
HITDRUNK as the dependent variables, and HIT, HITA6E,
and HITNUM as the independent variables, respectively. A
decrease in the percentages of respondents who have been
hit, who answered these questions with "yes" was noted.
Percentage ranges of 2.7 to 9.0 were charted for this
series of cross-tabulations for "yes" answers.
Respondents (as high as 97.3 percent) disagree that it
was "ok" to hit a drunk or a protest marcher.
The next cross-tabulation is HITCHILD (dependent
variable) by HIT (independent variable). Almost
sixty-three percent of the respondents (62.5) who had
been hit said "yes," it is ok for a stranger to hit a
child if the child accidentally damaged his car. On the
other hand, 50.9 percent of the respondents who had not
been hit said "yes" to this same question. The
significance level was greater than .000.
The next cross-tabulation is HITCHILD (dependent
variable) by HITAGE (independent variable). Again, the
35


results were as predicted. Those respondents who were
hit, both as an adult and as a child (65.6 percent of
them) said "yes" it is ok for a stranger to hit a child
if the child accidentally damaged his car. Sixty-three
percent of the respondents who had been hit as a child
said "yes" it is ok and 59.5 percent of the respondents
who had been hit as an adult said "yes". The results
were not statistically significant.
The next cross-tabulation was HITCHILD (dependent
variable) by HITNUM (independent variable). Again, those
respondents who had been hit the most (4 or more times)
had the highest percentage (62.9 percent) to agree that
it is ok to hit the child who had accidentally damaged a
strangers car. Fifty-nine percent of the respondents
who had been hit once said "yes," and 59 percent of the
respondents who had been hit 2 to 3 times responded with
"yes". The results were not statistically significant.
The next cross-tabulation was HITBEATR (dependent
variable) by HIT (independent variable). Eighty-six
percent of respondents who had been hit said "yes," it
is ok to hit a stranger if he is beating up a woman.
Eighty-four percent of the respondents who had never
been beaten said "yes." The significance level was .000.
The next cross-tabulation was HITBEATR (dependent
36


variable by HITAGE (independent variable). Eighty-seven
percent of the respondents who had been hit as a child
said "yes" it was ok to hit a stranger if he is beating
up a woman. Eighty-five percent of the respondents who
had been hit as an adult said "yes" it was ok, and 85
percent of the respondents who had been hit as an adult
and child said it was ok. Because of the slight
variation across rows, this association was not
statistically significant.
The next cross-tabulation was HITROBBR (dependent
variable) by HIT (independent variable). Of the
respondents who had been hit, 84 percent said "yes" it
is ok to hit a stranger if he is robbing your house.
Eighty-three point six percent of the respondents who
had never been hit said "yes". The significance level
was <.001.
The next cross-tabulation was HITROBBR (dependent
variable) by HITAGE (independent variable). Eighty-eight
percent of the respondents who had been hit as a child
said "yes", it was ok to hit a stranger if he is robbing
your house. Eighty-seven percent of the respondents who
had been hit as an adult said "yes" it was ok and
90.6percent of the respondents who had been hit as an
adult and a child said "yes". The significance level was
37


.001.
Finally, the last cross-tabulation was HITROBBR
(dependent variable) by HITNUM (independent variable).
Eighty-six percent of the respondents who had been hit
once said "yes", it was ok to hit a stranger who was
robbing your house, 88.8 percent of the respondents who
had been hit two or three times said "yes", and 88.0
percent of the respondents who had been hit four or more
times said "yes". The results were not statistically
significant.
In summarizing the attitudes toward violence by
experience of violence, it is interesting to note that
on almost all occasions, the majority of the respondents
who had been hit, condoned hitting. However, on the
questions that used HITMARCH and HITDRUNK, the opposite
was true. In these questions, an average of 94.5 percent
of the respondents who had been hit, answered "no", it
was not ok to hit a protest marcher who has opposing
views, and "no", it was not ok to hit a drunk who had
bumped into the man and his wife on the street. These
same variables elicited the same type of response for
1984 (which will be discussed later) and one can only
assume the reason for this response is the questions do
not present a threatening or antagonistic situation. All
38


of the other dependent variables seem to represent
conditions of conflict, whereas HITMARCH and HITDRUNK do
not.
Attitudes Toward Violence by Experience of
Violence-1984-Controlling for Sex
Before discussing attitudes toward violence by
experience of violence for 1984, controlling for sex, a
word of caution must be offered. It must be remembered
that the data is almost ten years old, and attitudes of
the respondents might be different today, considering
the nature of society. Thus, the process here was one of
studying the nature of violence and determining its
relationship to the attitudes toward violence. Again,
the latest information available in the NORC data that
had all the "HIT" variables, was for 1984.
From Tables 2.2 and 2.3, it can be seen that there
really was not a big difference between attitudes on
violence when the sex of the respondent was controlled
for. The majority of all men and women who had been hit
(punched or beaten) answered yes" to the HITOK
variable. There was statistically significant difference
for the males who were punched or beaten and answered
"yes" to the "HITOK" variable.
Here again, the same problem existed with the
39


Table 2.2
Attitude Toward Violence by Experience
of Violence 1984 Controlling for Sex
Experience of Violence
Percent of Men saying Yes: Have Been Hit Have Not Been Hit Total
Ok to hit in any situation? 72.6 (N=252) 55.8 (N=126) 66.0* (N=378)
Ok to hit a protest marcher? 3.9 (N=14) 5.5 (N=13) 4.5 (N=27)
Ok to hit a drunk? 6.3 (N=22) 14.2 (N=33) 9 5* (N=55)
Ok to hit if beating up woman? 87.6 (N=298) 82.7 (N=191) 85.6 (N=489)
Ok to hit if robbing house? 88.4 (N=306) 81.4 (N=188) 85.6 (N=494)
*Significance:P=<.05
40-


Table 2.2 Continued
Experience of Violence
Percent of Men Saying Yes: Hit Once Hit 2-3 Times Hit 4+ Times Total
Ok to hit in any situation? 64.8 (N=46 ) 74.1 (N=103) 74.0 (N=94) 72.1 (N=243)
Ok to hit a protest marcher? 5.5 (N=4) 3.5 (N=5) 3.9 (N=5) 4.1 (N=14)
Ok to hit a drunk? 8.3 (N=6) 4.9 (N=7) 6.5 (N=8) 6.2 (N=21)
Ok to hit child if damaging car? 62.0 (N=44) 56.8 (N=79) 62.4 (N=78) 60.0 (N=201)
Ok to hit if beating up woman? 90.3 (N=65) 87.0 (N=120) 86.6 (N=103) 87.5 (N=288)
Ok to hit if robbing house? 90.3 (N=65) 87.1 (N=122) 88.6 (N=109) 88.4 (N=296)
*Significance: P=<.05


Table 2.2 Continued
Experience of Violence
Percent of Men Saying Yes: Hit As Child Hit As Adult Hit As Both Total
Ok to hit in any situation? 71.1 (N-123) 71.4 (N=55) 76.3 (N=74) 72.6 (N=252)
Ok to hit a protest marcher? 1.1 (N=2) 10.3 (N=8) 4.0 (N=4) 3.9* (N=14)
Ok to hit a drunk? 5.0 (N=9) 7.7 (N=6) 7.6 (N=7) 6.3 (N=22)
Ok to hit a child if damaging car? 61.8 (N=107) 60.3 (N=47) 57.0 (N=53) 60.2 (N=207)
Ok to hit if beating up woman? 87.6 (N=156) 83.8 (N=62) 90.9 (N=80) 87.6 (N=298)
Ok to hit if robbing house? 85.1 (N=148) 92.1 (N=70) 91.7 (N=88) 88.4 (N=306)
*Significance: P=<.05
42


Table 2.3
Attitude Toward Violence by Experience
of Violence 1984 Controlling for Sex *
Experience of Violence
Percent of Wo- men Saying Yes: Have Been Hit Have Not Been Hit Total
Ok to hit in any situation? 65.0 (N=143) 57.6 (N=356) 59.5 (N=496)
Ok to hit a protest marcher? 4.1 (N=9) 1.8 (N=ll) 2.4 (N=20)
Ok to hit a drunk? 3.6 (N=8) 6.8 (N=42) 5.9 (N=50)
Ok to hit a child if damaging car? 60.1 (N=131) 49.4 (N=304) 52.2 (N=435)
Ok to hit if beating up woman? 84.8 (N=184) 85.7 (N=526) 85.4 (N=710)
Ok to hit if robbing house? 86.0 (N=191) 78.6 (N=491) 80.5 (N=682)
*Significance: P=<.05
43


Table 2.3 Continued
Experience of Violence
Percent of Wo- men Saying Yes: Hit Once Hit 2-3 Times Hit 4+ Times Total
Ok to hit in any situation? 58.0 (N=29) 63.6 (N=42) 69.7 (N=69) 65.1 (N=140)
Ok to hit a protest marcher? 8.0 (N=4) 3.0 (N=2) 3.1 (N=3) 4.2 (N=9)
Ok to hit a drunk? 7.7 (N=4) 1.5 (N=1) 3.1 (N=3) 3.7 (N=8)
Ok to hit a child if damaging car? 60.4 (N=29) 50.8 (N=33) 63.6 (N=63) 59.0 (N=125)
Ok to hit if beating up woman? 82.7 (N=43) 81.3 (N=52) 87.5 (N=84) 84.4 (N=179)
Ok to hit if robbing house? 83.0 (N=44) 86.2 (N=56) 86.7 (N=85) 85.6 (N=185)
*Significance: P=<.05
44


Table 2.3 Continued
Experience of Violence
Perccent of Wo- men Saying Yes: Hit As Child Hit As Adult Hit As Both Total
Ok to hit in any situation? 65.8 (N=50) 62.5 (N=70) 74.2 (N=23) 65.3 (N=143)
Ok to hit a protest marcher? 2.7 (N=2) 6.2 (N=7) 0 0 4.1 (N=9)
Ok to hit a drunk? 1.3 (N=1) 5.3 (N=6) 3.2 (N=1) 3.6 (N=8)
Ok to hit a child if damaging car? 68.9 (N=51) 50.0 (N=56) 77.4 (N=24) 60.4* (N=131)
Ok to hit if beating up woman? 89.5 (N=68) 82.3 (N=93) 85.2 (N=23) 85.2 (N=184)
Ok to hit if robbing house? 86.8 (N=66) 87.0 (N=100) 83.3 (N=25) 86.4 (N=191)
*Significance: P=<.05
45


HITMARCH and HITDRUNK variables. The cross-tabulations
containing these variables elicited the lowest
percentage rates of "yes" answers. One-hundred percent
of the females who were hit as an adult and child
answered "no" when asked if it was ok to hit a protest
marcher showing opposition to the other man's views, and
96 percent of the males answered "no" to this question.
The cross-tabulation for "HITCHILD" by "HIT" shows
60.2 percent of the males who had been punched or beaten
answered "yes", and 60.1 percent of the females who had
been punched or beaten answered "yes". There was
significance for the females at .01.
Males who had been punched or beaten as an adult
(92.1 percent) said it was "ok" to hit a stranger who
had broken into the man's house. This was the highest
"yes" response in the table. This was followed by 91.7
percent of the males who had been punched or beaten as
both a child and an adult, answering "yes" to the same
question. The highest "yes" for females was 87 percent
of the females who had been punched or beaten as an
adult, followed by 86.8 percent of the females who were
punched or beaten as a child answering "yes" that it is
ok to hit a stranger who had broken into the man's
house.
46


Similar to the findings in Table 2.1, it can be
seen that most (as high as 91.7 percent of the males for
1984) who were punched or beaten as an adult OR child,
and 90.6 percent of the respondents in general, who had
been punched or beaten as an adult AND child condoned
hitting (violence) as a means of resolving conflict in
some situations. And while the percentages of males and
females who have been hit, and condone (answered "yes")
are not appreciably different, the tables reflect that
more often, higher percentages of males answered "yes"
to the questions than did females.
Analysis of the Hypothesis. The previous analysis
of the relationship between physical violence and the
attitudes toward violence was presented in order to lay
the groundwork for the analysis of the research
hypothesis presented in Chapter 1. Before proceeding
with the analysis of the hypothesis, it is necessary to
point out that the research hypothesis in this
investigation deals with a specific set of conditions.
Those conditions are, that when people experience
violence, that experience determines their attitudes
toward violence. These conditions were selected because,
as the literature suggests, the likelihood of accepting
violence as a means of conflict resolution becomes
47


greater, the more that violence is experienced by the
individual.
The hypothesis states: Those who were raised in an
environment where violence is utilized are more likely
to condone the use of that same type of violence as a
means of conflict and problem resolution.
The expected relationship between experiencing
violence and one's attitude toward violence was borne
out in the analysis. The results do indicate that the
respondents who had experienced violence (i.e., being
punched or beaten, being punched or beaten as an adult,
child, or both, and being punched or beaten once, 2-3
times, or 4 or more times) approved of violence
(answered "yes"). Of the thirty-eight cross-tabulations
done for the year of 1984, in which sex was controlled
for, only twelve tabulations had responses where the
majority of the respondents said "no", it was not ok to
hit. Those were the "HITDRUNK" and "HITMARCH" dependent
variables. The general (all years) cross-tabulation had
similar results. Of the eighteen cross-tabulations done
for these variables, only six had responses where the
majority of the respondents answered "no" to the
questions on whether it was ok to hit. Again, this was
the "HITDRUNK" and "HITMARCH" dependent variables.
48


In further examining the results of this analysis,
there were higher percentages of men who had been hit
saying that hitting was ok in certain situations, than
the women in these same categories. The differences were
not that significant, but they were apparent. What does
this tell us? Does this mean that men who have
experienced violence are more likely to condone hitting
than women who have experienced the same kinds and
amounts of violence? Or, does it mean that, in general,
people who have been hit are more likely to say that
hitting is ok? Either way it is stated, I think the data
tends to support the premise that individuals who have
experienced violence as a means of conflict resolution,
are more likely to condone the use of violence.
This brings up the point of using data from a
survey that was not designed specifically for my study.
The best way to get survey data is to design and carry
out a survey focused on precisely the research you want
to study. Realistically, this was not possible.
Secondary analysis allowed me to do research that I
could not have otherwise done on my own. But it must be
kept in mind that the data were not collected
specifically for my own interests. The survey questions
may not have measured exactly what I wanted them to, but
45


I feel they have given a good indication of the
attitudes toward violence by individuals who have
experienced different amounts of violence at different
stages of their lives.


CHAPTER 6
SUMMARY AND CONCLUSIONS
Perhaps the most sinister aspect of violence is its
self-perpetuating character. Victims of family violence
seem to be at a higher risk to become future victims and
perpetrators. The inevitability of this connection is
sometimes exaggerated in popular discussion. Children
who experience and witness abusive behavior are by no
means DESTINED to grow up to be involved in abusive
relationships. Not all members of a particular family
constellation become abusers merely because they were
exposed to it in their family of origin. Some critics
(Dobash and Dobash, 1979) of the generational theory of
abuse point out that the learning theorists fail to
explain why, in a hypothetical family of four siblings,
where all have been exposed to violence, only one out of
the four may actually exhibit violent and abusive
behavior in their adult households. They contend that
because three siblings do not go on to repeat the
violence in their family backgrounds, the idea of a
generational transmission should be suspect and subject
to more sophisticated research before any more claims
can be made.
While this caveat seems to be based on a sound
5L


JO.
logic, it is unrealistic to pose such a criticism,
because all social science theory, unlike physical
scientific theories, are merely statistical trend
analysis. Therefore, taking one, or even a few isolated
families in order to question the validity of the
generational transmission of violence hypothesis, is
unsound.
Agreed, more controlled research containing two
statistically large groupsviolent and nonviolent is
desirable to make more statistically sounder predictions
with minimum standard deviation. However, a recent
survey conducted by the Colorado Domestic Violence
Coalition (1990) based on 150 cases selected randomly
from 1,000 cases, very strongly supports the
generational cycle hypothesis. For example, 81.1 percent
of the abusive partners came from homes in which they
themselves were abused, and had also witnessed their
fathers abusing their mothers. This supports that there
is a strong probability of generationally transmitted
abuse with early exposure to violence.
It does not explain why 100 percent of the abusers'
siblings do not grow up to beat their wives and
children, but it does substantiate that when there is a
high level of violence in the home, chances are four out
52


of five that at least one sibling will tend to become an
abuser. This would confirm, rather than devalue, the
likelihood of the cyclic nature of spouse and child
abuse.
When a reference is made to the transmission of
violence as generational, they are not ascribing to the
term the kind of certainty that would accompany a
hereditary disease for example. What is being said is
that if one considers the home environment as tantamount
to a school classroom, we could not possibly expect all
of the students (in the case of family violence, the
students being the children, and the parents being the
teachers) to learn everything that they observe or are
taught in exactly the same way, nor do we expect all of
the students will receive the same grades.
The cycle of violence makes intuitive and
theoretical sense. This may be the reason a cycle of
violence is sometimes used to explain more than it
actually does. People learn how to conduct intimate
relationships through the models they have. Certainly if
you have seen others use violence to handle conflict and
frustration, it will seem a more plausible and
legitimate answer when you too encounter difficulties.
If you have suffered family violence, you may also be
53


more tolerant of its use by others, against both
yourself, and others. To the extent that those who are
abused and witness abuse suffer from a sense of
powerlessness, stigma, and difficulty trusting others,
their healthy coping resources are impaired (Finkelhor,
1988). Violence is an ultimate resource, a tactic of
powerlessness and desperation.
The cycle of abuse is even more complex in that
there are many people who escape it. Many take from
their experiences a strong commitment never to inflict
such anguish on others. How people recover from the
trauma of the cycle of violence is not that well
understood. Perhaps that should be one of the major
tasks of future research.


REFERENCES
Akers, Ronal L. (1973). Deviant Behaviors. Bellmont,
Calif.: Wadsworth.
Alessi, J., & Hearn, K. (1984)." Group Treatment of
Children in Shelters for Battered Women." In A. Roberts
(Ed.), Battered Women and Their Families: Intervention
Strategies and Treatment programs, flew York: Springer.
Bakan, D. (1971). Slaughter of the Innocents: A Study of
the Battered Child Phenomenon. Boston: Beacon Press.
Bandura, Albert (1973). Aggression: A Social Learning
Analysis. Englewood Cliffs, N.J.: Prentice Hall.
Bowker, L. H., Arbitell, M., & McFerron, J. R.
(1988)."On the Relationship Between Wife Beating and
Child Abuse." In K. Yllo and M. Bograd (Eds.), Feminist
Perspectives on Wife Abuse. Newbury Park, Calif.: Sage
Press, .
Button, A. (1973). "Some Antecedents of Felonious and
Delinquent Behavior." Journal of Clinical Child
Psychology 2 (Fall):35-38.
Cappel, Charles, & Heiner, R. B., (1990). "The
Intergenerational Transmission of Family Aggression."
Journal of Family Violence, 5, (2), 135-152.
Colorado Domestic Violence Coalition Newsletter, Winter
1990.
Davidson, T. (1978). Conjugal Crime: Understanding and
Changing the Wife-Beating Pattern. New York: Hawthorn
Books.
Davis, Liane V. (1989). "Battered Women: The
Transformation of a Social Problem." Social Work,
July-Aug, 306-311.
Davis, L. V., & Carlson, B. D. (1987). "Observation of
Spouse Abuse: What Happens to the Children?" Journal of
Interspousal Violence, 2 (3), 278-291.
55


Dobash, R. E., & Dobash R., P. (1976). "The Importance
of Historical and Contemporary Context in Understanding
Marital Violence." Paper presented at the annual meeting
of the American Sociological Association, New York.
Dobash, R. E., & Dobash, R. P. (1978). "Wives: The
"Appropriate" Victims of Marital Violence." Victimology,
2 (3/4): 426-442.
Dobash, R. E., & Dobash, R. (1979). Violence Against
Wives: A Case Against the Patriarchy. New York: Free
Press.
Emery, R. E., & O'Leary, D. K. (1982). "Children's
Perceptions of Marital Discord and Behavior Problems of
Boys and Girls." Journal of Abnormal Child Psychology,
10(1), 11-22.
Finkelhor, Davis (1988). Stopping Family Violence:
Research Priorities for the Coming Decade. Newbury Park,
Calif.: Sage Publications.
Forsstrom-Cohen, B. & Rosenbaum, A.(1985). "The Effects
of Parental Marital Violence on Young Adults: An
Exploratory Investigation." Journal of Marriage and the
Family, May 267-472.
Gil, D. G. (1970). Violence against children: Physical
child abuse in the United States. Cambridge, Mass:
Harvard University Press.
Giles-Sims, J.(1985). Longitudinal Study of Battered
Children of Battered Wives." Family Relations, 34,
205-210.
Grossman, Beth (1977). "Children's Ideal Self-Which
Sex?" Paper presented at the annual meeting of the
American Psychological Association, San Francisco.
Hershorn, M., & Rosenbaum, A.(1985). "Children of
Marital Violence: A Closer Look a the Unintended
Victims." American Journal of Orthopsychiatry, 55(2),
260-266.
Hilberman, E., & Munson, K. (1977-78). "Sixty Battered
Women." Victimology: An International Journal, 2 (3-4),
460-463.
5.6


Hilgard, Ernest R., & Bower, G. H. (1966). Theories of
Learning. 3rd Edition, New York: Meredith.
Hughes, H. (1988). "Psychological and Behavioral
Correlated of Family Violence in Child Witnesses and
Victims." American Journal of Orthopsychiatry, 58 (1),
77-90.
Hughes, H. M., Cole, F. C., & Ito, E. S. (1988).
"Maternal Functioning and Family Violence as Predictions
of Child Maladjustment." Paper presented at the 9th
annual convention of the American Psychological
Association in Atlanta, Ga.
Hughes, H. M., Rau, T. J., Hampton, K. L., & Sablatura,
B. (1985) Effects of Family Violence on Child Victims
and Witnesses." Paper presented at the 93rd annual
convention of the American Psychological Association in
Los Angeles, Calif.
Hughes, Honore, M., Parkinson, D., & Vargo M.(1989).
"Witnessing Spouse Abuse and Experiencing Physical
Abuse: A Double Whammy?" Journal of Family Violence, 4
(2), 197-209.
Jaffe, P., Wolfe, D. A., Wilson, S. K. & Zak, L.(1985).
"Critical Issues in the Assessment of Children's
Adjustment to Witnessing Family Violence." Canada's
Mental Health, 33(4), 14-19.
Jaffe, P., Wolfe, D. A., Wilson, S. K., & Zak, L.(1986).
"Family Violence and Child Adjustment: A Comparative
Analysis of Girls' and Boys' Behavioral Symptoms."
American Journal of Psychiatry, 14(1), 74-77.
Jaffe, P., Wolfe, D., & Zak, L. (1986). "Similarities in
Behavioral and Social Maladjustment Among Child Victims
and Witnesses to Family Violence." American Journal of
Orthopsychiatry, 56, 142-146.
Kalmus, Debra K., (1984). "The Intergenerational
Transmission of Marital Aggression." Journal of Marriage
and the Family, 42, 11-19.
Kempe, C. H., et al. (1962)."The Battered Child
Syndrome." Journal of the American Medical Association,
181 (July): 17-24.
5.7


Levine, M. B.(1975). "Interparental Violence and Its
Effects on the Children: A Study of Fifty Families in
General Practice." Medicine, Science, and the Law, 15,
172-176.
Maurer, A., (1976)." Physical Punishment of Children."
Paper presented at the California State Psychological
Convention, Anaheim.
Pagelow, Mildred D. (1978). "Social Learning Theory and
Sex Roles: Violence Begins in the Home." Paper presented
at the North World Congress of Sociology, Uppsalel,
Sweden.
Pagelow, Mildred D. (1981a). "Sex Roles, Power, and
Woman Battering." In Lee H. Bowker (Ed.). Women and
Crime in America. New York: McMillan.
Pagelow, Mildred D. (1981b). Woman Battering: Victims
and Their Experiences. Newbury Park, Calif.: Sage Press.
Pagelow, Mildred D. (1982). "Children in Violent
Families: Direct and Indirect Victims." In Shirley Hill
and B. J. Barnes (Eds.), Young Children and Their
Families. Lexington, Mass.: D. C. Heath.
Pagelow, Mildred D. (1984). Family Violence. New York:
Praeger.
Palmer, S. (1962). The Psychology of Murder. New York:
Thomas Y Crowell Company
Pascoe, J. M. Hebbert J. Perl, T. & Loda, F. (1981).
"Violence in North Carolina Families Referred to a Child
Protection Team." North Carolina Medical Journal, Jan.,
35-37.
Pfouts, J., Schopler, J., & Henley, H. C. (1982).
"Forgotten Victims of Family Violence." Social Work, 27,
367-368.
Pizzey, E. (1974). Scream Quietly or the Neighbors Will
Hear. Short Hills, N' J.: Ridley Enslow.
38


Prescott, S., Letko, C. (1977). "Battered Women" in Roy,
Maria (Ed.) Battered Women: A Psycho-Sociological Study
of Domestic Violence. New York.: Van Nostrand.
Rosenbaum, A., & 0"Leary K, Daniel, (1981). "Children"
The Unintended Victims of Marital Violence." American
Journal of Orthopsychiatry, 51(64) 692-699.
Stark, E., & Flitcaft, A. (1981) "Woman-Battering, Child
Abuse, and Social Heredity." In N. Johnson (Ed.),
Marital Violence. London: Routledge & Kegan Paul.
Stark, E., & Flitcraft, A. (1983). "Social Knowledge,
Social Policy, and the Abuse of Women. The Case Against
Patriarchal Benevolence." In D. Finklehor, R. Gelles, G.
Hotaling, and M. Straus, (Eds.), The Dark Side of
Families: Current Family Violence Research. Beverly
Hills: Sage Press.
Steele, B. E., and Pollock C. B., (1974)." A Psychiatric
Study of Parents Who Abuse Infants and Small Children."
In R. E. Hefler, and C. H. Kempe (Eds.), The Battered
Child. Chicago: University of Chicago Press.
Straus, Murray A. (1976). "Sexual Inequality, Cultural
Norms, and Wife-Beating." In Emilio D. Vlano (Ed.),
Victims and Society. Washington D. C.: Visage Press.
Straus, Murray A., & Gelles, R. J. (1980). Physical
Violence in American Families. Risk Factors and
Adaptations to Violence in 8,145 Families. New
Brunswick: Transition Publishers.
Straus, Murray A., Gelles, R. J. & Steinmetz, S.K.
(1980). Behind Closed Doors: Violence in the American
Family. New York: Anchor Press.
Stullman, M. E., Schoenberger, A., & Hanks, S. E.
(1987). "Assessment and Treatment of the Child Witness
of Marital Violence." Paper presented at the 3rd Family
and Violence Research Conference, Durham, N. C.
Sun, M., & Thomas, E. (1987). Custody Litigation on
Behalf of Battered Women. New York: National Center on
Women and Family Law.
59


Walker, Lenore, (1984). The Battered Woman Syndrome. New
York: Springer.
Wallerstein, J. S., and Kelly, J. B. (1980). Surviving
the Breakup: How Children and Parents Cope With Divorce.
New York: Basic Books.
Wallerstein, J. S., (1988). "Keynote Address:
Association of Family and Conciliation Courts." Paper
presented at the spring conference of the Associaation
of Family and Conciliation Courts in Long Beach, Calif.
Welsh, R. S. (1976). "Severe Parental Punishment and
Delinquency: A Developmental Theory." Journal of Child
Psychology, 35(1): 17-21.
Westra, B., & Martin, H. P. (1984). "Children of
Battered Women." Maternal Child Nursing Journal, vol
10, n 1, 41-54.
Wolfe, D. A., Jaffe, P., Wilson, S. K. & Zak, L.
(1985). "Children of Battered Women: The Relation of
Child Behavior to Family Violence and Maternal Stress."
Journal of Counseling and Clinical Psychology, 33(5),
657-664.
Zuckerman, D. M. and Zuckerman, B. S. (1985).
"Television's Impact on Children", Pediatrics 75, Feb.,
233-40.
60