abuse. Walker (1979) also found that most of the abused women in her study
described the incidents of psychological abuse as their worst abuse experiences.
McCue (1995) stresses that targeted or repeated emotional abuse can severely
affect the victim’s sense of self and reality and result in psychological distress.
Sexual abuse involves forcing an intimate partner, through the use of
verbal or physical threats, to participate in sexual activities against the person’s
will (Carden, 1994; McCue, 1995). Forms of sexual abuse are commonly a part
of violence in the home and may happen with other forms of abuse (Campbell,
2002). The acts of sexual abuse may include making jokes about women and sex
17
in the presence of the victim, touching the victim sexually in uncomfortable
ways, withholding sex and affection, showing jealousy or engaging in sadism
and deriving pleasure from it (McCue, 1995). Many studies reveal that sexual
abuse has resulted in a higher incidence of health complaints and more
gynecological problems including abdominal pain, urinary problems, decreased
sexual desire and depression (Campbell & Soeken, 1999; Eby, Campbell,
Sullivan, & Davidson, 1995; Jamieson & Steege, 1997).
Physical abuse involves the use of physical force by the abuser to control
the abused spouse or intimate partner (Carden, 1994; McCue, 1995). It may start
with overt violence or intentional attack on the victim’s body causing minor
assaults (such as painful pinching or squeezing) leading to severe injuries (such
as homicide). Physical abuse includes pushing, grabbing, slapping, biting,
pushing or assault with a weapon. Studies documented that physical abuse can
cause physical injuries such as pain, broken bones, facial trauma, and tendon or
ligament injuries (Vavarro & Lasko, 1993; Zachariades & Koumarra-Agouridaki,
1990).
4. Explanatory models of intimate partner abuse
The causes of intimate partner abuse may not be known, but there are many
theories to explain the existence and extent of the problem. Some theories focus
18
on the patriarchal infrastructure of political, cultural, and social relations that
sanction intimate partner abuse. Other theories focus on the individual and look
for personal explanations such as the use of alcohol or drugs, the victim’s actions,
mental illness, stress, frustration and violent families of origin. However, none of
these theories appear to accurately describe the cause of intimate abuse.
Researchers have concluded that there may not be one single cause of intimate
partner abuse (Campbell & Fishwick, 1993; Hasselt, Morrison, Bellack &
Hersen, 1988) and the cause of intimate partner abuse is likely to be complex
and multi-factorial. Some of the explanatory models of intimate partner abuse
will now be discussed.
4. 1. The “Individual” model
This model mainly focuses on individual psychology and behaviour
in explaining intimate partner abuse. Based on psychodynamic theory, it is
suggested that mental illness in either the abuser or the abused may be the
predisposing factor (McCue, 1995). According to this theory, women in
abusive relationships are supposed to be primary masochistic, and
therefore invite and even enjoy abuse. They apparently intentionally
provoke their partners to attack in order to satisfy their own innate desires
and masochistic needs to be hurt and punished. Such abuse assures their
19
husbands that they exercise control and power and allows them to release
natural aggression (Fagan & Browne, 1994). Research on the concepts of
dominance and power found that higher levels of dominance were
associated with higher levels of violence (Straus & Gelles, 1990). It is
suggested that the male batterer may feel insecure about his own
masculinity; has vulnerable self-concept and pathological jealousy; is
unable to communicate feelings or identify feelings in others; subscribes
to traditional attitudes regarding male dominance; has a desire for control
over women and children, and yet has a complex mix of helplessness,
powerlessness and inadequacy that causes violent behaviour (McCue,
1995). The afore-mentioned explanations of intimate partner abuse have
been criticized as being based on overgeneralization from small and
non-representative clinical samples, and obtained under extremely
selective circumstances (Dutton, 1995). Thus, the validity and reliability
of such studies have been questioned.
4.2. The ‘Social-psychological’ perspective
Intimate partner abuse is also explained the by the Social Learning
Theory (Buadura, 1979; Dutton, 1995; Stith & Farley, 1993). According to
the Social Learning Theory, violence is a learned behaviour. The abusive
20
behaviour is often learned when a man witnesses violent acts by a male
authority figure in his childhood. He may perceive violence as a method of
conflict resolution and a means of maintaining power and control in
intimate relationships, especially when these violent acts are often
rewarded by immediate gratification (Jasinski & Kantor, 1998). According
to Bandura’s (1979) social–learning analysis of aggression, males who
have witnessed inter-parental violence during childhood may be
predisposed to abusive behaviour in their adult intimate relationships.
Hotaling, Straus and Lincoln (1990) reported that the experience of being
assaulted or living in a multi-assaultive family could lead to different types
of violent behaviour in later life. Mihalic and Elliott’s (1997) analyzed a
National Youth Survey of 360 female and 290 male respondents for
marital violence. Respondents who reported marital violence were further
examined with regard to their prior exposure to violence including
witnessing parental violence, child and adolescent abuse and victimization,
and involvement in violence during adolescence. The result showed that
males and females were affected differently by their prior experience with
violence in childhood and adolescence, suggesting that social learning was
an important factor in marital violence.
21
4.3 Sociocultural model
In this perspective, society’s cultural norms are considered to be a
pre-disposing factor for intimate partner abuse. In accordance with the
sociocultural belief, cultures that approve the use of violence tend to have
the highest rate of women violence (McCue, 1995). The sociocultural
perspective considers violent behaviour to be the result of conditions that
inhibit the biological, psychological and social needs of human beings
(Gelles and Straus, 1988). For example, if a person is frustrated at work
and feels that it is not acceptable to react to frustration with anger in the
workplace, that person may choose to express feelings of frustration at
home. Another sociological perspective on intimate partner abuse is
offered by the Social Exchange Theory (Gelles and Straus, 1988), which
advocates that people may use violence in the family when there appears
to be a reward for violent behaviour and when the costs of violence do not
exceed the rewards. In a study on violence in American society, Gelles and
Straus (1988) found that in the context of wife abuse, male batterers suffer
little cost by using violence; due to the inequalities of women in terms of
physical size, social and economic status in comparison with men. On the
22
other hand, there are immediate rewards for men for being violent,
including working off anger, gratification of personal needs, having power
to control and influence their intimate partners, and enhancement of
self-esteem. As a result, men are more likely to use violence on women in
the family, as argued by these authors.
4.4 System Theory
The System Theory is another approach to explain the process of
interaction within a family that leads to intimate partner abuse. According
to the system theory, violence is a product of the family system. In each
family, there are established rules of behaviours for each individual
member and interaction patterns within the system. Thus, if one family
member challenges the established rules of the family system, another
family member will take corrective action to maintain the equilibrium of
the system. This corrective action is taken to establish the latter member’s
power position and is done through an increase in violent behaviour.
Therefore, intimate partner abuse violence is perceived as a way for the
husband to cope with tension, and resolve conflicts within the family
system, as well as to restore the system’s equilibrium (McCue, 1995).
23
4.5. The Patriarchal model
According to the patriarchal ideology, men are highly valued and
are the household heads. It has been further suggested that the patriarchal
ideology supports much of our social structure, which creates and
maintains male domination over women (Dobash & Dobash, 1979). As a
result, men wield power over women and the latter have been devalued as
secondary and inferior, thus encouraging men to use physical force to
control or gain power over their wives or intimate partners (McCue, 1995;
Yllo & Straus, 1990). Men’s authoritative position is further reinforced by
religious teachings, society, and the economic and legal systems. A study
of Yllo & Straus (1990) has shown that there is a linear relationship
between patriarchal norms and violence against wives; and if there are
more patriarchal norms about marital power, there is more wife abuse. The
same authors have concluded that society’s tolerance of wife beating is a
reflection of the patriarchal norms that support male dominance in
marriage.
Emerson Dobash and Russell Dobash, proponents of the feminist
theory on violence against women, argue that patriarchy contributes to
intimate partner abuse and that patriarchy is encouraged in our economic
24
and social system (Dobash & Dobash, 1979). This is because a husband
has been defined as the dominant, strong, authoritarian, aggressive and
rational provider for the family while the wife has been assigned to take a
dependent, passive, submissive, soft role. Under this model, women are
encouraged to remain at home and care for their husbands and children,
while the husbands stay outside to earn money to provide a living for their
family.
Power in marital relationships has been suggested to be related to
intimate partner abuse in that violence toward the wife is high when the
decision making power between the spouses is not equal (Yllo & Straus,
1990). Empirical research by Straus, Gelles, and Steinmetz (1988) has
shown that the rate of wife beating in couples of husband dominance was
50% higher than wife-dominant couples, and more than 300% higher
when compared with egalitarian couples. A study by Tang (1999) in
Hong Kong also found that husband-dominance resulted in more severe
violence, while verbal and physical aggression was less prevalent in
egalitarian marriages.
Female status has been the subject of research in several studies.
Campbell (1992) found no linear correlation between female status and
25
rates of intimate partner abuse. Levison (1989), in a cross-cultural research
on the link between violence against women and sexual inequality, found
that societies that enjoy sexual equality between spouses tend to have less
intimate partner violence. Yllo and Straus (1990), in a study on the
relationship of patriarchy and wife assault, reported that the rate of wife
assault was higher in lowest and highest female status societies. The
authors argue that high female status breaks down the patriarchal norms
and that, men have to resort to violence to bolster their threatened
masculinity. While in the lowest female status society, men usually use
force to keep women in their lower status place in the family. Yllo and
Straus (1990) thus conclude that societal power imbalance is associated
with intimate partner abuse even though the mechanism that generates the
violence is not clear.
Patriarchal theory has been used frequently to explain intimate
partner abuse in Chinese society (Bullough & Ruan, 1994; Rimont, 1991;
Tang, 1991). This is because the Chinese culture is traditionally patriarchal,
as influenced by Confucian teachings (Chan & Lee, 1995; Tang, 1994).
The ethical norm of Confucianism for a proper family life is a patriarchal,
patrilineal and patrilocal family system. Such a system has rooted Chinese
26
women in an inferior, dependent, obedient and submissive role while
Chinese men are superior and authoritarian. Traditionally, a Chinese
woman should follow and obey her father in childhood, her husband in
adulthood, and her eldest son in old age. Besides, a respectable woman
should possess the four virtues of fidelity, tidiness, propriety in speech and
commitment in needlework. A good women should follow this social
norm and to be a good mother and good wife. Wife beating is legitimized
in a traditional Chinese family and defended as the “rules of the family”.
In the case of wife beating, the husband may believe the wife is his
property and he has the right to beat her for not complying with his needs
or her prescribed role (Chan & Lee, 1995; Tang, 1994). To what extent do
the traditional Chinese patriarchal values contribute to wife abuse in Hong
Kong remains unclear, and further studies are warranted.
With the open policy in recent years of accepting immigrants from
Mainland China, the number of new immigrants has increased steadily in
Hong Kong. Fu (1988) reported that more immigrants have sought help at
refuge shelters for battered women due to financial difficulties and poor
relationships with their husbands. Further, Man and Wu (1996) found that
Chinese women called for police help more frequently, and that may be
27
related to their lesser resources in Hong Kong.
Since there is no single explanation for the cause of intimate partner
abuse, researchers have tried to find the risk factors, that is, the
characteristics associated with the abuse and the increased likelihood that
the abuse would occur (Kantor and Jainski, 1998). The risk factors for
wife abuse have been identified as childhood violence, low income and
education of the man, alcohol use by the man, communication skill
deficits, rigid sex-role attitudes, personality trait disorders, child abuse, hot
temper, stress and depression (McCue, 1995). Local statistics (Harmony
House, 2001) indicate that apart from male chauvinism, work pressure,
financial problems, sex problems, conflict over task division, extra-marital
affairs, psychiatric problems, drug addiction, alcoholic, gambling, jealousy
and in-law problems have been linked with intimate partner abuse.
5. Cycle of violence
Abused women have shown themselves to be resilient and resourceful
despite their sad experiences. Lampbert (1996) found that women would
develop agency in abusive relationships to halt, change and cope with the
violence. A study by Campbell, Rose, Kub and Nedd (1998) found that in
response to battering, women behave in a complicated and non-violent manner
28
with evidence of resistance and resourcefulness. Walker (1979) categories the
violence process into three phases, namely the tension-building phase, the acute
battering incident, and a calm period; and describes the process as a cycle of
violence. According to Walker, during the tension-building phase, women
would try to calm the batterer but even then, the man may show a gradual
escalation of tension through discrete and harmful actions. However, the
women may maintain an unrealistic belief that she can control the man. During
the acute battering incident, the women would be severely shaken and
frightened by the sudden outburst. Finally, during the calm period, the abused
women truly want to forgive their abusers and believe their partners’ apologies.
This honeymoon phase gives hope to the abused women that the abuse, rather
than the relationship will end (Walker, 2000).
6. Coping of abused women
Self-concept, that is, how a battered woman’s perception of herself as a
woman, a wife and a mother, could be a powerful coping mechanism by
influencing her interpretation of the nature and reasons for her abuse. The
differences in self-concept among abused women may lead to minimization or
a maximization of the psychological effects of the abuse and further influence
the appropriateness of action taken. Poor self-concept has been identified in
29
women who are staying in the abusive relationship (Gelles & Straus, 1990).
The resilience and resourcefulness of abused women is one of the
strategies to cope with the abusive relationship. The women’s responses to
battering are purposeful and develop active strategies for protection and
survival (Campbell, Rose, Kub and Nedd, 1998; Tiwari, Wong & Ip, 2001). A
study of Lempert (1996) found women would strategize and develop
contingencies in abusive relationships to halt, change and cope with the
violence.
Emotional attachment has been said to be one of the reasons for women
staying with their abusers. Emotional attachment and dependency often persist,
at least for sometime, even when the relationship is distressed or terminated.
Emotional attachment convinces the women staying with their abusers that
every abuse is an individual incident and that their husbands will reform
(Dutton, 1993). During the honeymoon phase of Walker’s (1979) cycle of
violence, women tend to respond to the violence with dependence behaviors,
and loyalties to marriage and family values. In this case, the feeling of agony
and rejection is minimized.
Learned helplessness has also been suggested as one of the women’s ways
of coping with the abuse (Walker, 1979). This suggests that the abused women
30
soon learn that their voluntary responses really do not make much difference in
what happens to them, and that changing their own behaviour would not
change their life situation or their abuser’s behaviour. The notion of learned
helplessness has been challenged by subsequent studies on abused women’s
responses, which show that abused women are resilient and resourceful despite
life’s challenges (Campbell, Rose, Kub and Nedd, 1998; Lampbert, 1996).
7. Impact of abuse on women
The increase in physical and mental health problems of abused women is
well documented (Dienemann, 2000; McCauley et al., 1995; Robert, Lawrence,
Williams, & Raphael, 1998). Studies have shown that physical health
consequences of abuse include disability, arthritis, chronic pain, migraine,
frequent headaches, stammering, sexually transmitted infections, HIV, chronic
pelvic pain (Coker, Smith, Lesa, King & McKeown, 2000; Eby, Campbell,
Sullivan & Davidson, 1995) and even risk factors for homicide (Campbell &
Soeken, 1999). Other studies reported psychological health consequences of
abuse including generalized anxiety, dysthymia, depression, phobias, current
harmful alcohol consumption and psychoactive drug dependence with
long-term reactions to violence (Robers, Lawrence, Williams & Raphael,
1998).
31
8. Effect of abuse on pregnant women
Abuse during pregnancy increases the risk to mother and fetus. A study by
Wester, Chandler and Battistutta (1996) indicated that pregnant women who
experienced violence had a higher incidence of miscarriage, abortion and
neonatal death. In other studies, violent attacks have been linked to intrauterine
growth retardation, and low birth weight infant (McFarlane, Parker & Soeken,
1996). Psychological abuse, which causes the production of catecholamines in
response to stress, may also inhibit placental perfusion resulting in fetal
hypoxia (Newton & Hunt, 1984). However, in a study by Leung, Wong, Leung
and Ho (2001) on pregnant abused women in Hong Kong no adverse effects of
domestic violence on pregnancy outcomes were found. On the other hand,
another study of 838 Chinese postnatal women by Leung, Kung, Lam, Leung
and Ho (2002) demonstrated the relationship of abuse in pregnancy and
postnatal depression suggesting that abused women are more likely to develop
postnatal blues and depression.
9. Definition of quality of life
There is no universal definition of quality of life. It varies according to
different cultures and persons; and may vary between academic disciplines
(Bowling & Windsor, 2001). Quality of life may be defined as an individual’s
32
satisfaction with the life domain, which he or she considers important
(Hornquist, 1982 as cited by Bowling & Windsor, 2001). Kaplan (1991)
considers quality of life as health and physical functioning, while a more recent
definition by the World Health Organization suggests that quality of life is a
composite measure of physical, mental and psychological well being, together
with happiness, satisfaction and gratification with all-important aspects of life
(WHO, 1985).
10. Health related quality of life
Health can affect the overall quality of life. Health related quality (HRQL)
is defined as the impact of perceived health on an individual’s ability to live a
fulfilling life (Bullinger, 1993). Corresponding to an interest in the quality of
life in the medical domain, instruments for assessing the quality of life had
been developed. There are disease specific instruments and generic instruments
that can be applicable to all populations. The multi-dimensional concept of the
health related quality of life consists of at least three aspects of the quality of
life, which include emotional well being, physical well being and social
functioning (Power, Bullinger, Harper & The WHO Quality of Life Group,
1999). How an individual perceives the performance of each of those activities
may be assessed separately by measuring the satisfaction for each domain. In
33
this study, the concept of health related quality of life refers to the eight
concepts derived by Ware and Sherbourne (1992). They include: 1) limitations
in physical activities because of health problems; 2) limitations in social
activities because of physical or emotional problems; 3) limitations in usual
role activities because of physical health problems; 4) bodily pain; 5) general
mental health (psychological distress and well being); 6) limitations in usual
role activities because of emotional problems; 7) vitality (energy and fatigue);
and 8) general health perceptions.
Summary
Despite an increased attention to intimate partner abuse, there is a dearth
of such study in Hong Kong. The negative mental and physical health
consequences of intimate partner abuse are widely recognized and can cause
significant impact on the quality of life of the abused. Data on the quality of life
of abuse women in Hong Kong, however, are lacking. Without such meaningful
measures, the effectiveness of interventions for spouse abuse could not be
accurately estimated. Hence, the purpose of this study is to explore the types of
intimate partner abuse in a group of pregnant women in Hong Kong and to
investigate the effect of the abuse on the women’s quality of life.
34
CHAPTER 3
Methodology
1. Study Design
This was a secondary data analysis of a pilot study of an exploratory and
descriptive study design to understand the existence, patterns of intimate partner
abuse in Hong Kong Chinese families, and the quality of life of abused women.
The study was a cross sectional study of a group of postnatal women who were
identified as ‘abused’ during their antenatal period. The study used a quantitative
approach consisting of closed-end questions to identify the nature and incidence
of the abuses to which the women were subjected and their self-perception
towards their own bio-psychosocial health. As such, the impact of intimate
partner abuse on the quality of life was explored.
2. Sampling
Since the study aimed to explore the magnitude of intimate partner abuse
in Hong Kong Chinese women and the effect of abuse on the abused women’s
quality of life, Chinese women were the target population. The sample was
drawn from women identified as positive using the Abuse Assessment Screening
35
Questionnaire (AAS) in a previous study conducted by some members of the
research team (Leung, Leung, Lam & Ho, 1999). The following selection
criteria were used for the present study:
Inclusion criteria: Chinese women who had reported being abused based on the
AAS; and they were able and willing to communicate their abuse to the
researcher.
Exclusion criteria: Women who were mentally incapacitated or psychological
unstable were excluded from the study.
Convenience sampling was used to recruit the women for the study, which
was carried out in one of the general hospitals under the Hong Kong Hospital
Authority.
3. Data Collection
This is was a secondary data analysis. Data collection of primary study
was carried out in the first two weeks of July 2001. Twenty-nine women
consented to participate in the study after a full explanation of the nature and
purpose of the study, the risks and benefits, the rights of research subjects, and
the confidentiality of the data collected. After obtaining the woman’s written
consent, two questionnaires were administered, the Chinese version of the
Conflict Tactics Scale (CTS) and the Chinese (HK) version of the SF-36.
36
Additionally, an investigator-developed demographic sheet was used to assess
the women’s biosocial characteristics including age, number of years living in
Hong Kong, number of children, marital status, occupation of husband/partner,
education level of husband/partner, drinking habits of husband/partner, and
whether husband/partner was happy with the gender of the baby.
4. Instruments
The Chinese version of the Conflict Tactics Scale (CTS) (Straus, 1979)
was used to explore the women and their spouses’ conflict management, with
two additions: sexual abuse and severity of injury. The Conflict Tactics Scale
(Appendix 1), a 7-point, 19-item Likert-type questionnaire, measures the use of
reasoning, verbal/psychological aggression and physical violence in resolving
conflict in a marital, cohabiting, or dating relationship. The women were asked
to indicate which of the tactics their intimate partners had used during times of
conflict in the past year and how often the tactics were used. The reasoning
subscale (Appendix A, item a, b, c) refers to the intellectual approach in which
disputes are resolved through rational discussion, argument, and reasoning. The
verbal subscale (Appendix A, items d, e, f, g, h, I, j) involves the use of verbal
and /or nonverbal acts that symbolically hurt the intimate partner or the use of
threat to the intimate partner. Items of the verbal abuse subscales include:
37
insulted or swore at, sulked and/ or refused to talk; stomped out of the room or
house; did or said something to spite; and threatened to hit or throw something at.
The physical subscale (Appendix A, items k, l, m, n, o, p, q, r, s) refers to the act
of physical violence against an intimate partner as a means of resolving conflicts.
The physical subscale is further divided into minor and severe violence. Items of
minor violence (Appendix A, items k, l, m) include: threw something at; pushed,
grabbed, or shoved; and slapped or spanked the intimate partner. Items of severe
physical violence (Appendix A, items n, o, p, q, r, s) include: kicked, bit, or hit
with a fist; hit or tried to hit with an object; beat up; threatened with a
weapon/knife/chopper; and used with a weapon/knife/chopper. The woman was
asked to indicate the number of times each action occurred during the past
twelve months, ranging from “Never” to “More than 20 times”. An extra item
was added to the CTS, asking the woman to indicate if she had been subjected to
any sexual abuse and the frequency of it should it occur. Finally, an additional
section was added to find out the severity of the injury, which ranged from minor,
moderate, severe physical hurt to permanent physical damage.
The reliability of CTS, as indicated by the internal consistency, ranges
from 0.50 to 0.88 for the three subscales. The reliability coefficient is higher for
the verbal aggression and violence scales but lower for the reasoning scale
38
(Straus, 1979). Although research evidence indicates high reliability and validity
of the CTS (Straus, 1990), limitations are acknowledged including the ignoring
of the context of the violence and the extent of the injury and the likelihood of
social desirability bias (Straus, 1990). The Chinese translation of the CTS, as
used in this study, shows satisfactory reliability, which ranges from 0.76 to 0.86
(Tang, 1994).
The SF-36 is a 36-item Short Form Health Survey developed by Ware,
Snow and Kosinski (1993) to measure health related quality of life. The SF-36
consists of eight multi-item scales: ten items on physical functioning, four items
on role limitation due to physical problems, two items on bodily pain, five items
on general health, four items on vitality, two items on social functioning, three
items on role limitation due to emotional problems and five items on mental
health. The scores of the items in each scale have to be summated and
transformed into a scale score that has a standardized range from zero to 100.
Higher scale scores indicate better quality of life. As health-related quality of life
is relative rather than absolute, the results should be interpreted in the light of
what is normal for the same population (Gandek & Ware, 1998). For this reason,
a Chinese (HK) version of the SF-36 was developed and validated for Chinese
people in Hong Kong, with normative values for the local Chinese adults (Lam,
39
Lauder, Lam & Gandek, 1999).
The SF-36 is a validated, internationally used generic measure of quality
of life on different health conditions (Lam, Gandek, Ren, Chan, 1998; Thumboo
et al., 2001). There is evidence supporting its validity and reliability (Ware &
Sherbourne, 1992). The Chinese (HK) version of the SF-36 was tested on 236
Chinese people in Hong Kong in 1996 and the conceptual validity and construct
validity were confirmed (Lam, Gandek, Ren, Chan, 1998). Besides, the Chinese
(HK) version of the SF-36 was also validated and normed by compared to the
general population in 1996 (Lam, Lauder, Lam & Gandek, 1999).
5. Data Analysis
The Software of Statistical Package for the Social Sciences (SPSS) version
10 was used to analysis the raw data gathered. The statistical methods used to
analyze the data were descriptive statistics and the Spearman Test. The
Spearman’s correlation was calculated and the significance level was set at
p<0.05. It was used to test the correlation between the nature of intimate partner
abuse and their partners’ demographic data. The partner’s demographic data
include education level, drinking habits, satisfaction towards the sex of their
children, and perception towards sexual equality. For the SF-36 items, the
respondents’ responses were calculated and expressed as mean score for each of
40
the subscales, before comparing the mean scores with those of the normative
values. The effects of intimate partner abuse on the respondents’ quality of life
were also analyzed using the Spearman Test.
The strength of the relationship between the demographic variables and
the types of abuse, as well as that between the types of abuse and quality of life
subscales were expressed as correlation coefficient. Correlation coefficient is a
useful index to show the strength and direction of a relationship between two
variables. It is very important to keep in mind that correlation coefficient reflects
relationship, not cause and effect (Polit & Hungler, 1997).
The possible values for a correlation coefficient range from –1.00 through
0.0 to +1.00 (Portney & Watkins, 2000). A correlation of 0.0 indicates no
relationship, whereas a correlation approaching +1 indicates strong relationship
or perfect correlation. The plus sign indicates a positive correlation, whereas the
minus sigh indicates a negative correlation (Portney & Watkins, 2000). A
correlation that is published without a plus or minus sign is usually interpreted as
positive.
There are no widely accepted criteria for defining a strong, moderate, or
weak association. However, in health science studies, correlations ranging from
0.00 to 0.25 indicate little or no relationship; range from 0.25 to 0.50
41
demonstrates a fair degree of relationship; those from 0.50 to 0.75 suggest
moderate to good relationship; and values above 0.75 are considered as good to
excellent (Munro & Page, 1993). It should be pointed out that perfect correlation
is rarely found in research. The magnitude of the relationship is indicated by
how close the correlation comes towards the absolute value of 1.00.
The direction of a relationship between two variables is indicated by a
positive (+) or negative (-) correlation. A positive correlation means that as the
value of one variable increases, the value of the other variable also increases.
In contrast, a negative correlation means that as the value of one variable
increases, the value of the other variable decreases.
6. Ethical consideration:
An approval to conduct the study was sought and obtained from the
Institutional Review Board of the hospital. As explained earlier, a full
explanation of the nature and purpose of the study and the risks and benefits that
may be associated with the study was given. The potential participants were
informed that participation in the study was entirely voluntary and even after
agreeing to join the study; they might withdraw from it at any time without
having to give reasons and without causing any adverse effect to their treatment.
They were also assured that the data collected would be kept confidential.
42
Informed consent was obtained by the signing of a consent form. Anonymity
was ensured since no names were elicited on the questionnaires to protect the
identity of the participants (Nieswidadomy, 1993). Confidentiality of the
collected data was assured by confining it to the access of the research team
members for the purpose of completing of the study.
43
CHAPTER 4
Results
This chapter reports on the results obtained from 28 respondents who
participated in the study. The questionnaire completed by one of the respondents
was incomplete and was therefore excluded from the analysis. The information
obtained was divided into: (1) demographics, (2) incidence and nature of abuse
reported, and (3) health-related quality of life.
1. Demographic data
1.1 Respondents’ personal particulars
1.1.1 Marital status: 27 (96.4%) of the respondents were married and
1 (3.6%) was single.
1.1.2 Number of children: 20 (71.4%) of the respondents had 1 child, 7(25%)
had 2 children, and 1 (3.6%) had 3 children.
1.1.3 Years of living in Hong Kong: 26 (92.8 %) have lived in Hong Kong for
more than 7 years, 1 (3.5 %) within 1-7 years, and 1 did not answer the
question (Figure 1).
44
1-7 years
4%
>7 years
92%
no answer
4%
Figure 1. Years of the abused women living in Hong Kong
_________________________________________________________________
1.2 Partners’ particulars
1.2.1 Occupation: 5 (18%) respondents reported their partners’ occupation as
professional, 2 (7.14%) as associate professional, 10 (35.72%) as skilled
worker, 9 (32.14%) as semi-skilled, 1 (3.57%) was unskilled worker
and 1 (3.57%) was a student (Figure 2).
45
18%
7.14%
35.72%
32.14%
3.57% 3.57%
0%
5%
10%
15%
20%
25%
30%
35%
40%
Percentage of partners
professional associate
professional
skilled
worker
semi-skilled
worker
unskilled
worker
others
Partners' occupations
Figure 2. Occupations of the abused women’s partners
_________________________________________________________________
1.2.2 Education level: 6 (21%) of the respondent’s partners had attained
university level education, 12 (42%) had completed Form 5, 7 (25%) had
completed Form 3, 2(7.1%) had only received primary education
(Figure 3).
46
Form 7
4%
University
21%
Form 3
25%
Primary
7%
Form 5
43%
Figure 3. The abused women’s partner’s education level
________________________________________________________________
1.2.3 Drinking habit: 7 (25%) of the partners were reported to have a habit of
drinking alcohol while the remaining (75%) were reported to have no
such habit.
1.2.4 Satisfaction towards the sex of their children: 23 (82.1%) of the partners
were reported to be satisfied with their children’s sex (Figure 4).
47
Don't m ind
14%
Satisfied 82%
No answer
4%
Figure 4. The abused women’s partners’ satisfaction towards their
children’s sex
________________________________________________________
1.2.5 Partners’ perception towards sexual equality: 21 (75%) of the
partners were reported to agree with sexual equality, 6 (21.4%) were
reported to support the dominance of men over women, and 1 (3.6%) was
reported to be uncertain.
2. Incidence and nature of abuse
2.1 Verbal abuse
Verbal abuse was reported by 26 (92.8%) respondents. As shown
in table 3, the 2 most frequent verbal abuse were “sulking or refusing to
talk” (mean=4.18), and “stomping out of the room” (mean=2.71) whilst
“crying” was the least used tactic (mean=0.21).
48
Table 3.
Incidence of verbal abuse of 28 respondents
Behaviour
Never Once Twice 3 to 5
times
6 to
10
times
11 to
20
times
More
than
20
times
Mean Standard
Deviation
Insulted or
swore at you
19
67.8
%
2
7.1
%
6
21.4
%
1
3.6
%
0
0%
0
0%
0
0%
0.64
1.06
Sulked or
refused to
talk about
the issue
9
32.1
%
1
3.6
%
6
21.4
%
4
14.3
%
6
21.4
%
0
0%
1
3.6
%
4.18 5.56
Stomped out
of the room
or house
16
57.1
%
1
3.6
%
1
3.6
%
6
21.4
%
3
10.7
%
0
0%
1
3.6
%
2.71 5.13
Cried
25
89.2
%
2
7.1
%
1
3.6
%
0
0%
0
0%
0
0%
0
0%
0.21 0.79
Did or said
something to
Spite you
23
82.1
%
1
2. 6
2%
1
3.6
%
2
7.1
%
0
0%
0
0%
1
3.6
%
1.29 4.78
Threatened
to hit or
throw
something at
you
27
96.4
%
1
3.6
%
0
0%
0
0%
0
0%
0
0%
0
0%
2.12 0.19
49
2.2 Minor physical abuse
Minor physical abuse was reported by 4 of the respondents. The
types of minor physical violence reported were “throwing, smashing,
hitting or kicking something” (mean = 0.5), “throwing something at the
respondent” (mean = 3.57) and “pushing, grabbing or shoving the
respondent” (mean = 0.43), as shown in Table 4.
Table 4.
Incidence of minor physical abuse of 28 respondents
Behaviour
Never Once Twice 3 to 5
times
6 to 10
times
11 to 20
times
More
than
20
times
Mean Standard
deviation
Threw or
smashed or
hit or kicked
something
22
78.5
%
2
7.1
%
2
7.1
%
2
7.1
%
0
0%
0
0%
0
0%
0.5 0.14
Threw
something at
the
respondent
27
96.4
%
1
3.6
%
0
0%
0
0%
0
0%
0
0%
0
0%
3.57 0.19
2.3 Severe physical abuse
27 (96.4%) of the respondents reported that their partners had
never used violent behaviour such as kicking or beating with fist, hitting
50
or trying to hit with something, beating up, threatening with knife or
weapon, or using a knife or weapon within the twelve months prior to the
interview. 1 (3.6%) of the respondents did not answer the questions in
this category, hence, it was not possible to assess the incidence for this
respondent.
2.4. Incidence of sexual abuse
Only 1 (3.6%) respondent indicated that she had been sexually
abused up to 3-5 times within the last year. 1 respondent did not answer
the question at all, and the remaining 26 (2%) respondents reported no
sexual abuse.
2.5 Correlation of nature of abuse and their intimate partners’
particulars
The correlation of the intimate partners’ abuse and their partners’
particulars were as follows:
2.5.1 Verbal abuse: There were no correlations between verbal abuse and the
partner’s occupation, education level, alcohol habits, satisfaction with
children’s sex or perception towards sexual equality, as shown in Table 5.
51
Table 5.
Correlation of verbal abuse with intimate partners’ demographic particulars
Correlation
and level of
significance
Occupation Education
level
Alcoholic
habit
Satisfaction
towards
their
children’s
sex
Perception
towards
sexual
equality
Correlation(rs)
-0.243 0.105 -0.032 -0.147 0.065
Level of
significant (p)
(2- tailed)
0.213 0.596 0.872 0.454 0.741
2.5.2 Physical abuse: There were no correlations between minor physical
violence and the partner’s occupation, education level, alcohol habit,
satisfaction with children’s sex or perception towards sexual equality, as
shown in Table 6.
52
Table 6.
Correlation of minor physical abuse with intimate partners’ demographic particulars
Correlation
and
Significance
Level
Occupation Education
Level
Alcoholic
habit
Satisfaction
towards their
children’s
sex
Perception
towards
sexual
equality
Correlation (rs)
-0.02 0.346 -0.012 0.202 0.153
Level of significant
(p)
(2- tailed)
0.993 0.072 0.951 0.303 0.436
2.5.3 Sexual abuse: There were no correlations between sexual abuse and the
partner’s occupation, education level, alcohol habits, satisfaction with
children’s sex or perception towards sexual equality, as shown in Table 7.
Table 7.
Correction of sexual abuse with intimate partners’ demographic particulars
Correlation
and
Significant level
Occupation Education
level
Alcoholic
habit
Satisfaction
towards their
children’s sex
Perception
towards
sexual
equality
Correlation (rs)
0.111 0.026 -0.116 -0.041 -0.115
Level of
significant
(p)(2- tailed)
0.580 0.896 0.564 0.840 0.567
53
3. Health-related quality of life
3.1 General health
11 (39.3%) respondents reported that their general health was
good while 10 (35.7%) reported as fair (Figure 5).
3.60%
21.40%
39.30%
35.70%
0%
0.00%
5.00%
10.00%
15.00%
20.00%
25.00%
30.00%
35.00%
40.00%
Percentage of abused women
Excellent Very good Good Fair Poor
Present health condition
Figure 5. Women’s self perception towards their health
_______________________________________________________________
3.2 Perception towards present health as compared with last year
15 (53.6%) respondents perceived that their health condition was
the same as the past year and 7(25%) perceived that their health was
worse than the past year (Figure 6).
54
7.10% 14.30%
53.60%
25%
0%
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
Percentage of abused women
Much
better
Somewhat
better
Same Somewhat
worse
Much
worse
Health condition
Figure 6. The abused women’s perception towards their present health as
compare with last year
________________________________________________________________
3.3 Daily activities
3.3.1 Degrees of limitation: None of the respondents had any limitation in
activities such as bathing, walking one block or climbing a flight of stairs.
2 (7.1%) respondents reported a lot of limitation in vigorous activities, 1
(3.57%) respondent reported a lot of limitation in activities of bending,
kneeling or stooping, and 3 (10.7%) respondents reported a lot of
limitation in activities of walking more than one kilometer. Details of the
respondents’ limitation of daily activities were presented in table 8
(Numerical values in the table denote number of respondent).
55
Table 8.
Degree of limitation in daily activities of 28 respondents
Daily activities Limited a
lot
Limited a
little
No
limitation
at all
Unknown
Vigorous activities 2 6 17 3
Moderate activities 0 1 27 0
Lifting/carrying groceries 0 2 25 1
Climbing several flights 0 6 21 1
Climbing one flight 0 0 28 0
Bending, kneeling/stooping 1 7 20 0
Walking more than 1 km 3 5 17 3
Walking several blocks 0 2 26 0
Walking one block 0 0 28 0
Bathing/dressing 0 0 28 0
3.3.2 The impact of physical health on daily activities: Figure 7 shows that
some respondents had problems with their work or other regular daily
activities as a result of their physical health during the past four weeks. 8
(28.6%) respondents reported that they had to cut down time on work.
56
9 (32.1%) respondents reported they had accomplished less than they
expected. 8 (28.6%) reported that they were limited in work or other
activities. 6 (21.4%) reported that they had difficulty performing work
or other activities.
28.60%
71.40%
32.10%
67.90%
28.60%
71.40%
21.40%
78.60%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Percentage of abused women
Cut down time
on work
Acomplished
less work
Limited in
kinds of work
difficulty in
performing
work
The impact on daily activities
no
yes
Figure 7. The impact of health problems on the abused women’s daily
activities
________________________________________________________________
3.3.3 The impact of emotional problems on daily activities: Figure 8 showed
that quite a number of the respondents’ emotional problems had affected
their daily activities during the past four weeks prior to the interview. 15
(53.6%) had cut down the time spent on work or other activities.15
57
(53.6%) had accomplished less than they wanted. 12 (42.9%) did not do
work or other activities as carefully as usual.
53.60%
46.40%
53.60%
46.40%
42.90%
57.10%
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00%
Percentage of abused women
Cut down time on
work
Acomplished less
work
Worked
carelessly
Result of the impact
no
yes
Figure 8. The impact of emotional problems on women’s daily activities
______________________________________________________________
3.3.4 The impact of physical and emotional problems on social activities:
3.3.4.1 Extent of the impact of physical and emotional problems on their normal
social activities with family or friends: During the past four weeks, 1
(3.6%) respondent reported that their physical and emotional problems
had extremely affected their social activities, 2 (7.1%) reported quite a bit,
58
6 (21.4%) reported moderately, 9 (32.1%) reported slightly, 9 (32.1%)
reported not at all, and 1 (3.6%) did not answer.
3.3.4.2 Time frequency of the impact of physical and emotional problems on
their social activities with friends or relatives: 1 (3.6%) reported that her
physical and emotional problems had affected her social activities most
of the time, such as visiting friends or relatives, 10 (35.7%) reported
some of the time, 6 (21.4%) reported a little of the time, 10 (35.7%)
reported none of the time, and 1 (3.6%) did not answer.
3.4 Bodily pain:
3.4.1 Severity: 8 (28.6%) respondents reported no bodily pain within last four
week. 5 (17.9%) reported very mild bodily pain. 6 (21.4%) reported mild
bodily pain. 7 (25%) reported moderate bodily pain. 1 (3.57%) reported
severe, and 1 (3.5%) reported very severe bodily pain.
3.4.2 Effect of bodily pain on work: 2 (7.1%) reported quite a bit of effect, 4
(14.3%) reported moderately effect, 5 (17.9%) reported a little effect, 16
(57.1%) reported no effect, 1 (3.5%) reported quite a bit effect.
3.5 Respondents’ self perception
3.5.1 Towards their life: Table 9 shows the respondents’ self perception
towards their life within the past four weeks (numerical values in the
59
table denote number of respondents).
Table 9.
Respondents’ self-perception towards their life within the last four weeks
Self-
Perception
All
the
time
Most
of the
time
A good
bit of
time
Sometime A little
of time
None
Feeling full of pep
1 2 7 11 7 0
Being a very
nervous person
2 1 4 10 9 1
Being a happy
person
4 6 5 10 3 0
Feeling calm and
peaceful
2 8 4 0 11 3
Having a lot of
energy
0 4 6 16 1 1
Feeling so down in
the dumps that
nothing cheered
them up
0 2 0 9 6 11
Feeling downhearted
and blue
1 1 1 9 10 6
Feeling worn out
1 4 12 0 8 3
Feeling tired
2 6 1 15 3 1
60
3.5.2 Towards their health: Some of them reported they were healthy at the
moment (Table 10).
Table 10.
Self-perception towards their health
Health condition
level
Definitely
true
Mostly
true
Don’t
know
Mostly
false
Definitely
false
Likely to get sick more
easier
2 1 4 10 11
Expecting health to get
worse
0 4 8 7 9
Feeling as healthy as
anyone else
9 13 5 1 0
Feeling personal health
excellent
4 13 5 5 1
4. Intimate partner abuse and their quality of life
Verbal abuse was the most frequently reported abuse in this study,
followed by minor physical violence. Therefore, verbal abuse and minor
physical violence were analyzed for their correlation with the Respondents’
quality of life subscales.
61
4.1. Correlation of verbal abuse with respondents’ quality of life
subscales
The analysis revealed no relationship between verbal abuse and the
respondent’s quality of life subscales, as shown in Table 11.
Table 11.
Correlation of verbal abuse with the respondents’ quality of life subscales
Correlation and
level of
significance
PF RP BP GH V SF RE MH
Correlation (rs)
0.114 0.082 -0.174 0.413 -0.063 0.132 0.059 0.023
Level of
significance (p)
0.562 0.679 0.375 0.161 0.751 0.503 0.767 0.907
Note: PF=physical functioning, RP=role physical, BP=bodily pain, GH=general
health, V=vitality, SF=social functioning, RE=role emotion, MH=mental health
4.1.2 Correlation of minor physical abuse with respondents’ quality of life
subscales: There was no relationship between minor physical violence
and the respondents’ quality of life subscales as shown in Table 12.
62
Table 12.
Correlation of minor physical abuse with respondents’ quality of life subscales
Correlation and
level of
significance
PF RP BP GH V SSF RE MH
Correlation (rs)
0.186 0.062 -0.085 0.158 0.108 -0.094 -0.030 0.164
Levels of
significance (p)
0.343 0.754 0.669 0.421 0.585 0.634 0.880 0.405
Note: PF=physical functioning, RP=role physical, BP=bodily pain, GH=general
health, V=vitality, SF=social functioning, RE=role emotion, MH=mental health
5. Respondents’ mental health and their quality of life
5.1 Correlation of respondents’ mental health subscale and their
demographic data
There was no relationship between mental health subscales and
demographic data as shown in Table 13.
63
Table 13.
Correlation of respondents’ mental health subscale and their demographic data
Correlation and
level of significance
Marital status No. of children Years of living in
Hong Kong
Correlation (rs) 0.299 0.331 0.346
Level of
significance (p)
0.123 0.085 0.071
5.2. Respondents’ mental health and their quality of life subscales
There was a moderate relationship between mental health and
physical function; between mental health and role physical and between
mental health and vitality, while a stronger relationship was detected
between mental health and general health as shown in Table 14.
Table. 14.
Respondents’ mental health and their quality of life subscales
Correlation and level of
significance
PF RP BP GH V SF RE
Correlation (rs)
0.425 0.428 0.203 0.642 0.579 -0.171 -0.244
Level of significance (p)
0.024 0.023 0.300 0.000 0.001 0.383 0.206
Note: PF=physical functioning, RP=role physical, BP=bodily pain, GH=general
health, V=vitality, SF=social functioning, RE=role emotion, MH=mental health
64
6. Respondents general health and their quality of life
6.1 Respondents’ general health and their quality of life subscales
A moderate relationship was detected between general health and
physical function, role physical and vitality as shown in Table 15.
Table 15.
Correlation of respondents’ general health and quality of life subscales
Correlation and level of
significance
PF RP BP V SF RE
Correlation (rs)
0.546 0.410 0.249 0.570 0.085 0.166
Level of significance (p) 0.003 0.030 0.201 0.002 0.667 0.400
Note: PF=physical functioning, RP=role physical, BP=bodily pain, GH=general
health, V=vitality, SF=social functioning, RE=role emotion, MH=mental health
6.2 Comparison of the abused women’s quality of life with the
normative value
The mean of the abused women’s quality of life was compared
with the normative value to assess the abused women’s quality of life.
With the exception of the general health subscale, the scores in the eight
subscales of the SF-36 were appeared to be lower than those in the
normative values of Hong Kong Chinese women in the 18 to 40 age
65
group. Although the mean of the physical functioning subscale of the
abused women appeared slightly higher than the normative values, the
difference was negligible (Table 16).
Table 16. Comparison of the mean of the women’s quality of life and the normative
value:
Mean Standard deviation (SD) Subscales
Study group Normality
value
Study group Normality
group
Physical functioning
92.57 91.83
9.62 12.89
Role physical
72.32 82.43 41.02 30.97
Bodily pain 68.89 83.98
22.67 21.89
General health 68.20 55.98
20.12 20.18
Vitality 52.18 60.27
19.05 18.65
Social functioning 68.13 91.19
26.40 16.49
Role-emotional 51.20 71.67
42.05 38.39
Mental health 62.79 72.79
19.73 16.57
66
CHAPTER 5
Discussion
1. Nature of abuse
The findings of this study are consistent with those other earlier local
research studies in identifying verbal abuse as the most frequently reported form
of abuse. For example, a study by Tang (1999) revealed that 67% of the 1132
women surveyed had at least one incident of verbal abuse from 11th August to 3rd
November 1998 and 10% of those women experienced at least one episode of
physical abuse by their intimate partner during antenatal. Verbal abuse was also
the key finding of a study by Leung, Leung, Lam and Ho (1999), in which 99
women (15.7%) of the 631 women surveyed reported to have been verbally
abused. It should be pointed out that the official statistics on domestic violence
revealed a very different picture. For example, in the 2001 statistics (C. Sin,
personal communication, April 16, 2002), physical abuse was identified as the
most common abuse reported (95.3%), followed by psychological abuse (2.2%);
sexual abuse (0.46%) was the least common. The discrepancy between the
research studies and official statistics may be due to the different definitions
67
employed to define and classify abuse. Further, while the research studies
focused on self-report of ‘healthy’ individual, the official statistics rely on
referrals from social services, the police and the judiciary system, which tend to
deal with the more severe cases often with physical injuries. Compared with
studies conducted outside Hong Kong, differences in the prevalence of the type
of abuse were noted in this study. Whereas verbal abuse was the most frequently
reported abuse here, physical abuse was the most common type of abuse in
Western countries (Cocker, Smith, McKeown & King, 2000; Mazza,
Dennerstein, Garamszegi & Dudley, 2001) or Japan (Yoshihama, M & Sorenson,
1994). In those countries, psychological or emotional abuse is usually a
precursor of physical abuse. Culture may explain the differences in the types of
abuse reported. Since Chinese people attach great importance to maintaining
social harmony (Bond, 1996), outright conflict involving physical violence is
avoided as far as possible. Thus, when two people degenerate into open
dispute, their vocal duels rarely lead to physical assault as they are usually
constrained by other people. So, even when the two opponents continue to abuse
each other verbally, physical assault rarely happens (Bond, 1991; Tang, Wong &
Cheung, 2002). This is referred to as “a civilized person uses his month, but not
his fists” (Bond, 1991, p.66). In addition, Chinese people are collectivist
68
orientated. In such a culture, social support is high; and deviant behaviours
such as sexual aggression and physical violence are perceived as an offense to
the individual and to the community and avoided as much as possible (Hall &
Barongan, 1997; Yick, 2000). This may account for the low rate of physical
violence reported by the respondents in this study.
The traditional Chinese culture that dictates family affairs (such as
domestic violence) should not be disclosed to outsiders may have been
responsible for the low incidence of physical violence reported (Tang, Wong &
Cheung, 2002; Xu, Campbell & Zhu, 2001). However, this does not explain why
the respondents may be prepared to report verbal abuse but not physical
violence.
Finally, as there is little consensus among researchers on the definition of
intimate partner abuse and the use of a screening instrument, it is impossible to
make an accurate comparison of the various study findings.
2. Risk Factors
Risk factors are characteristics associated with an increased likelihood
that a problem behaviour will occur (Jasinski & Williams, 1998). Risk factors
that correlate with intimate partner abuse have been explored in several studies,
including such factors as alcohol (Hedin & Janson, 2000; Johnson & Holly, 2001;
69
Richards, 1991); lower social class, lower status job, and lower educational level
of the abusive partners (Heise et al.’s 1994; Smith, 1990), and machismo attitude
toward women (Richards, 1991).
An interesting finding of this study is that the partners’ occupation
education, alcoholic habit, and the number of children were not associated with
abuse. While this is comparable with the findings of another local study (Tang,
1999), it contradicts the results of an earlier study on pregnant abused women
(Leung, Leung, Lam & Ho, 1999), in which lower social class and manual
workers were found to be correlated with abuse.
Surprisingly, over 75% of the intimate partners in this study were
reported to be towards equalitarian; and only 20% of them were thought to
favour male dominance. The findings did not appear to be consistent with the
Chinese traditional belief and value, which support a patriarchal system (Choi,
Cheung, Tang & Yik, 1993). The discrepancy may be explained by the small
sample size and convenience sampling used in this study. Further, as the
partner’s orientation toward gender equality is reported by the women, it may
not have been accurately portrayed.
3. Quality of Life
With the exception of the general health subscale, the scores in the eight
70
subscales of the SF-36, which denote the women’s quality of life, were lower
than those in the normative values of Hong Kong Chinese women in the 18 to 40
age group. The respondents’ self-report of their general health, which was found
to be higher than that of the normative values of Chinese women in Hong Kong,
raised an interesting question. Was their general health indeed better than women
in the general population or did they perceive their general health to be better?
The higher score in the general health subscale as reported by these women may
be due to the fact that they had got over the enduring tiredness and physical
discomfort experienced during the antenatal period. On the other hand, as these
were self-reports, they reflected the perceptions of the respondents only. As
perception is personal: it represents the meaning given by one to one’s
experience (Combs & Snygg, 1991), and may not be an accurate representation
of the reality. Also, it is known that abused women tend to delay help-seeking
for their health problems until these become serious, in order to avoid social
stigma and save face in the family (Sampselle et al., 1992). Thus, there is a
danger that the respondents in this study may not recognise the need to seek
professional help despite the poorer quality of life as reported in the role
physical, physical function, bodily pain, vitality, social functioning, role
emotional and mental health subscales.
71
The lower scores reported by the women in the vitality, role emotional,
role physical, bodily pain, social functioning and mental health subscales should
be noted. Since this group of women had just delivered babies within the past
few months, and had to cope with the extra demands of motherhood, their
reported problems in role physical, physical function, bodily pain, vitality, role
emotional and mental health could further compromise their self-care ability and
their responsibilities as mothers and wives. This compromised ability could well
induce further violence due to the woman’s inability to fulfil her role as
perceived by her abuser.
Studies have shown that psychological abuse is probably the most
damaging form of abuse (Dutton, 1995; Jasinski & Williams, 1998; O’Leary &
Maiuro, 2001; Ratner, 1993; Sackett & Saunders, 1999; Semple, 2001). As it is
directed at the women’s basic sense of self and causes a profound negative
impact on their self-concept. Anxiety, stress related symptoms and depression
may result (Hampton, 1999). Although Chinese women may use the traditional
beliefs of ren (endurance) and yuan (predetermined affinity) to help them cope
with the abuse (Tiwari, Wong & Ip, 2001), the detrimental effect of
psychological abuse may still leave a mark on these women. This was confirmed
in Leung, Kung, Lam, Leung and Ho’s study (2002), in which abused women
72
were found to be twice as likely to develop postnatal depression than non-abuse
women. Postpartum depression may last as long as a few days to one and a half
years after delivery. Baby negligence, child abuse or attempting suicide will be
the result if the condition becomes serious. Thus, the trauma caused by
psychological abuse should not be under-estimated.
73
CHAPTER 6
Summary
This study has investigated the incidence and nature of intimate partner abuse
by the use of the Chinese version of The Conflict Tactic Scale (Straus, 1979) and
explore the quality of life of abused women by comparison with the normative value
of Hong Kong Chinese women in the 18 to 40 age group. The result showed that
verbal abuse was the most frequently reported form of abuse and that the quality of
life of these abused women was lower than those the normative value group except
in the general health subscales. This health issue should not be neglected and poses
an implication to nursing practice.
1. Implication For Nursing Practice
As seen from this study, intimate partner abuse is not only a social
problem; it is also a health care issue. Inevitably it affects maternal mental health,
depresses emotions, impairs physical function and affects social activities,
eventually resulting in a lower quality of life. Attention should be paid to those
women so that delayed antenatal visits, postpartum depression, infanticide and
suicide are not overlooked. Screening for abuse during antenatal visits and
74
postnatal visits is warranted especially amongst pregnant women with high risk
factors. However, choosing a proper screening method, which is culturally
specific, is beneficial for early detection of intimate partner abuse in order to
give proper intervention and treatment.
With the increasing numbers of intimate partner abuse in recent years,
nurses will come into increased contact with abused women. Nurses should be
well prepared to care effectively for abused women; consequently, training will
be needed for nurses to aid the victims as well as a means provided for
reinforcing counseling skills. When helping abused women, several issues
should be addressed such as the abused women’s culture of help-seeking
behaviour, learned helplessness, low self-esteem, how to help them to break the
cycle of violence and the impact of abuse affecting their mental health and
lowering their quality of life. Nurses should be aware of their need so as to
provide specific cultural care. Some women may be reluctant to expose the
abuse due to cultural factors, so nurses should be sensitive enough to address this
problem.
There are evidences that lawyers; policemen and nurses are not properly
prepared to manage women involved in intimate partner abuse (Chung, Wong, &
Yiu, Yiu, 1991). Nurses should recognize their own personal values, beliefs,
75
attitudes, and nursing practices when caring for these women. According to
Fishbein & Ajzen (1975), one’s value, belief, experience and knowledge will
influence one’s formation of specific attitudes, and in turn it will determine an
individual’s intention to act. The damaging effects of verbal abuse on
self-esteem and self-identity are well known. For this group of vulnerable
women, their self-concept and self-esteem are usually low (Orava, McLeod &
Sharpe, 1996). Nurses should be aware of these behaviour patterns when
providing care for these women to prevent further victimization. The abused
women may lack awareness of their human rights (Fry & Barker, 2001). For this
reason, there may be a need to offer choices and resources to assist them to cope
with their devastating situation.
Abused women may be reluctant to be referred to a place of refuge even
when the abuse is severe because of being concerned about their mother caring
role, as well as their self-care attitude (Campbell, 1989). Besides, they may be
afraid of further abuse by their partner or relaliation or having to testify against
their partner in court. In view of the abused women’s fears and feelings, nurses
would do well to consider themselves as client advocates during the nursing
healing process. If nurses approach abused women in a distant or patronizing
way, or become frustrated when they do not take their advice, it will cause
76
further victimization for these ones. The code elements of nursing are patience,
environment and nursing. In practice these forms a healing process. Therefore,
the best approach to care for these women is a holistic one.
The abused women in this study appeared to experience a poorer quality
of life than the general group; they may be due to psychological stress from a
variety of sources such as unsatisfactory marital relationships, abuse by their
intimate partner and decreased social activities. These factors will lead to
problems of ill health if their problems are not solved. Thus, health promotion by
health education for these women and the public on the concept of a
health-related quality of life is needed.
2. Strength
Although this is a pilot study with a small sample size, it provides some
insight into the nature of intimate partner abuse and its effect on the quality of
life of the abused women. In addition, this is the first trial to address the quality
of life of abused women in Hong Kong.
3. Limitation
This study has a number of limitations. First, this is a pilot study and the
sample size is small therefore it is limited to generalization. Second, the people
recruited may not be representative due to convenience sampling. Third, the
77
measure of a male partners’ perception towards sexual equality is by an indirect
means, that is, from the victims. The measure of perceptions coming directly
from their partners may have different results. Fourth, The Conflict Tactics Scale
(CTS) has the limitation of social desirability (Straus, 1990). Straus (1990)
further explained that social desirability is a threat to the validity of these scales
because not every respondent will be willing to describe the true picture in
which the partner kicked or punched the respondents. Some respondents may not
tell the truth. Therefore, there may be over reporting or underreporting. Fifth,
The Conflict Tactic Scale (1979) is derived from Western culture. Although the
Chinese version of CTS showed satisfactory reliability (α ranged from .76
to .86) (Tang, 1994), there is no support of cultural appropriateness for Chinese
culture (Xiao, Campbell & Zhu, 2001) and this may be is a threat to the validity
of these scales
4. Recommendations
4.1 Recommendations for nursing research
4.1.1 Replication of this study is recommended with a larger randomized
sampling size to increase generalizability. In this study, the partners’ risk
factors are not conclusive due to the small sample size. They should be
further explored. Women’s risk factors should also be studied.
78
4.1.2. Research of the attitude, knowledge and belief of nurses are
recommended in order to detect their understanding and attitude to this
issue in order to provide proper training.
4.1.3. The experiences of the victims are studied extensively in Western
countries. There are no related studies in Hong Kong. Besides, the
context of intimate partner abuse is not well understood amongst Hong
Kong Chinese. In order to obtain the context of intimate partner abuse, an
in-depth study of the experiences of victims and the abusers is
recommended to obtain the conceptualization of intimate partner abuse
from the victims and abusers in order to facilitate intervention and
prevention efforts.
4.1.4. The Chinese translation of CTS is developed from Western culture.
Applicability of the instrument in Chinese is questioned. In future, testing
the sensitivity and validity of the instruments are recommended in order
to obtain a true picture of the issue.
4.2 Recommendations for nursing practice
4.2.1 Empowerment of abused women
Abused women usually are more powerless in an abusive
relationship due to a reluctance to seek help because of social stigma and
79
having to save family face. Besides, the abused women usually are
traditional, submissive and fear to provoke the anger of their husbands
(Tiwari, Wong & Yip, 2001). Thus, interventions are needed to focus on
empowering women to become self-sufficient and responsible for
themselves by offering resources, safety measures, a means to choices
and enhance decision-making. Empowerment is an intervention that has
been found to be effective for abused women. Research studies have
shown that abused women having received empowerment intervention
reported significantly less violence and took more actions to increase
their safety than those women in the comparison group (Parker,
McFarlane, Soeken, Silva &Reel, 1999). Abused women face many
decisions when deciding to stay or leave an abusive relationship. The
education of having choices is another form of empowerment. Many
abused women believe that they have no way out of the abusive
relationship because they do not want to be divorced. This may be related
to the traditional patriarchal views that support the social role of women
as child bearers, nurturers and supporters of the family. Many abused
women may not think that they have choices. In order to have proper
control of one’s life, Dutton (1992) states that there are three perquisites
80
in effective decision- making. First, one must believe the right to make
personal choices. Second, one must be aware that there are different
choices. Third, one must have the resources to engage in decisionmaking.
Therefore, providing resources for these abused women to help
them to cope with the vulnerable situation is essential.
4.2.2 Empowerment of nurse on education
Bevis (1989) stated that a nurse has to be empowered in order to
be compassionate, well educated, creative, capable of independent
judgment and action, and courageous. Nursing education is not adequate
to educate nursing student about this societal problem because there is no
such training in basic nursing curriculum. A study was taken in the
Accident and Emergency Department in one of the Hong Kong Hospital
Authority Hospitals and reported that nurses are not well prepared to
encounter intimate partner abuse (Chung, Wong & Yiu, 1996). Recently,
Hong Kong Harmony House (Non Government Office) has offered
training for policeman for handling abused women (Harmony House,
2001). However, there is still no proper training for nurses in hospitals to
counter intimate partner abuse.
81
4.2.3 Proper screening and intervention
Adopting a cultural specific and sensitive screening method to
screen any abuse is needed for the abused women to obtain proper
assistance and advice without blame and shame. Holistic intervention
will facilitate the process. Thus, setting up intervention and protocol for
offering help and a referral service for these abused women is warranted.
For this group of abused women in this study, counseling, follow up or
referral service are also recommended to assess their progress and to
offer proactive intervention before their problems become serious.
4.2.4 Public education
A multi-disciplinary approach is recommended to help to prevent
and control intimate partner abuse. It can be done by proactive public
health education by mass media to increase public concern. Perhaps
education of children during their formative years to shape the social
norm of gender equality would be an advantage in counteracting future
abuse.
5. Conclusion
Because this was a pilot study, it therefore cannot be generalized to fit the
overall population. It reveals the critical need for additional research using large
82
randomized samples of victims and abusers. It also provides some hints that the
major focus of the incidence of abuse amongst Hong Kong Chinese people is
verbal abuse. It also provides some hints of the poorer quality of life of this
population. It serves to infuse some insight into the health problem issue faced
by these abused women. Finally, the results of this study highlight the fact that
health care professionals may be overlooking a life-threatening problem by not
screening pregnant women for intimate partner abuse.