2013. Vol.4, No.7, 559-565
Published Online July 2013 in SciRes (
Copyright © 2013 SciRes. 559
Violence against Women: Methodological and Ethical Issues
Kaltrina Kelmendi1,2
1Department of Psychology, University of Prishtina, Prishtine, Kosovo
2Faculty of Social Studies, Masaryk University, Brno, Czech Republic
Received April 18th, 2013; revised May 22 nd, 2013; accepted June 20th, 2013
Copyright © 2013 Kaltrina Kelmendi. This is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the
original work is properly cited.
Research on violence against women has improved and provides important information on patterns,
prevalence, risk and consequences of this major threat to female well-being. Since the identification of
violence against women as a problem worthy of study in 1970, evident progress has been made in under-
standing physical, psychological and sexual violence against women. However, while methodological
improvements appear in later studies, the literature review shows many limitations and restrictions when
conducting research on violence against women. The objective of this paper is to review the methodo-
logical issues that arise when studying violence against women. The paper focuses first on the history of
research on violence against women, by elaborating on each perspective. Second, the paper identifies and
describes methodological difficulties when researching violence against women such as methodology,
operational definitions of violence, sampling frame and risk factors related to violence. The paper also
elaborates on major ethical principles that should be considered and respected when researching violence
against women. Finally, the paper recommends certain changes that should be made in order to improve
future research on the subject.
Keywords: Intimate Partner Violence; Methodological Issues; Ethical Issues
Violence against women, in its multiple forms, is increas-
ingly recognized by individuals and states as a global problem
and a serious violation of women’s rights (Krug, Dahelberg,
Mercy, Zwi, & Lozano, 2002; Garcia-Moreno, Jansen, Ellsberg,
Heise, & Watts, 2006; UN Study, 2005; Ellsberg, 2006). Vio-
lence affects women in different ways; in particular on their
mental and physical health. It leads to stress, depression, loss of
self-esteem, reluctance to join the wider world, and more severe
psychological and physical problems. The World Health Or-
ganization, which defines health as a state of complete physic al,
mental and social well-being, describes domestic violence by
male partners as the most common health risk in the world for
women (World Health Organization, 2002).
Intimate partner violence is the most common, widespread
and predominant form of violence against women. In the WHO
multi-country study conducted with over 24,000 women from
15 sites in 10 countries representing diverse cultural settings:
Bangladesh, Brazil, Ethiopia, Japan, Namibia, Peru, Samoa,
Serbia and Montenegro, Thailand, and the United Republic of
Tanzania, it is estimated that the range of lifetime prevalence of
physical violence by intimate partner was between 13% (Japan
city) and 61% (Peru province), with most sites falling between
23% and 49 %. Between 4% and 49% of ever-partnered women
reported severe physical violence. The range of lifetime preva-
lence of sexual violence by intimate partner was between 6%
(Japan city and Serbia and Montenegro) and 59% (Ethiopia
province), with most sites falling between 10% and 50 %. The
range of lifetime prevalence of physical or sexual abuse, or
both, by intimate partner was between 15% (Japan city) and
71% (Ethiopia province) (Garcia-Moreno, Jansen, Ellsberg,
Heise, & Watts, 2006: p. 1264). As could be noticed, research-
ers find considerable variation in the prevalence of partner vio-
lence from country to country and among studies within a
country. Therefore, it is hard for researchers to compare inter-
national prevalence data on violence because different studies
use different study populations, methods and definitions of
This paper attempts to describe the history of research in the
area of violence against women, through analyzing and exam-
ining the focus of research through the years. In addition, the
paper aims to review methodological issues that arise when
studying violence against women, focusing on: methods of data
collection, operational definitions of violence, measures and
risk factors. Furthermore, it tries to provide public opinion with
ethical considerations and implications for conducting such a
specific research. Finally, this review will focus on intimate
partner violence, specifically on violence against women by
male partners. For the purpose of this paper we define intimate
partner violence as: “Any behavior within an intimate relation-
ship that causes physical, psychological, or sexual harm to a
partner in the relationship. Such behaviors include acts of
physical aggression, psychological abuse, forced intercourse
and other forms of sexual coercion, various controlling behav-
iors such as isolating a person from family and friends, moni-
toring her movements, and restricting access to information or
assistance” (Heise & Garcia-Moreno, 2002: p. 89).
Historical Background of Violence
against Women
Violence is a dynamic and complex phenomenon that tran-
scends racial, ethnic, gender and age boundaries. Since family
structure exists in all cultures, it is present in all cultures. Vio-
lence in families was not considered a problematic issue until
the 1960s, due to cultural norms that refused to consider it a
problem with consequences (Hotaling, Straus, & Lincoln,
1990). Hence, intimate partner violence as a social or legal
phenomenon was acknowledged only recently. The second
wave of the feminist movement that emerged from the United
States during the 1970s, and later from Western Europe, pro-
vided major contributions in this direction. The feminist move-
ment originated from women who shared their life experiences
of violent acts such as physical violence, rape and incest
(Tjaden, 2005: p.1). As a result, women began to organize con-
sciousness raising campaigns, write books, open shelters, and
advocate reforms in laws regarding child abuse, domestic vio-
lence and sexual assault. Similarly, most of the scholars at that
time focused on changing judicial systems to accomplish re-
form in the criminal codes defining sexual assault, domestic
violence, and child abuse.
In the beginning of the 1980s violence against women began
to be addressed as a public health issue. The public health per-
spective identified the prevalence and types of violence; it de-
veloped models that explained the violence and developed in-
tervention programs for the treatment of victims (Carlson,
Worden, Ryn, & Bachman, 2000). However, most researchers
doubted the usefulness of this approach due to the health sec-
tor’s historic aim of diagnosis and treatment of individuals
without raising social levels (Carlson, Worden, Ryn, & Bach-
man, 2000). Despite the above mentioned doubts about the
public health approach, the World Health Organization’s report
on violence against women was considered an important ad-
vancement among scholars in identifying types of interpersonal
violence, such as: 1) physical violence, 2) sexual violence, 3)
psychological violence, and 4) deprivation/neglect (Krug, Da-
helberg, Mercy, Zwi, & Lozano, 2002). Moreover, according to
Klipatrick (2004), three important aspects were emphasized in
the public health definition of violence: 1) intentional use of
force or power (which refers to the tendency to harm the part-
ner); 2) the intentional use of power or force (refers to the ten-
dency to show power in a relationship, including: threats, in-
timidation and omission); and 3) these intentional acts are not
required to produce injury, harm or deprivation in order to be
defined as violence (p. 1214). Recently, violence against
women has come to be recognized as a legitimate human rights
issue and a significant threat to women’s health and well-being
(Ellsberg & Heise, 2005). The identification of the link between
violence against women and discrimination was a key to begin
to consider the violence against women as human rights issue
rather than criminal or public health issue (UN Study on Vio-
lence against Women, 2005). The acknowledgment of violence
against women as a human rights issue was supported by evi-
dence based research on the prevalence and incidence of vio-
lence and could be considered a major advancement. Multiple
forms of violence against women indicated that violence
against women is global and is based on the inequality between
men and women. In this regard the human rights perspective
sets out a number of measures, including those addressing tra-
ditional or religious practices, by which states might prevent
and eliminate violence (UN Study on Violence against Women,
2005: p. 15).
Methodological Issues on Researching Intimate
Partner Violence
Most of the studies done before the 1970s and at the start of
the 1970’s were conducted in clinical samples and used psy-
chopathological models in their attempt to explain violence as a
psychological disorder (Gelles, 1980). The first representative
study, based upon the large sample of adult population, was
conducted in the United States during 1975 and 1980. The aim
of these studies was to obtain valid and reliable data on preva-
lence and causal analyses relying on statistical techniques. The
results of this study (Gelles, 1980) and other studies (Straus,
1977, 1978; Steintmetz, 1977, 1978; Straus, Gelles & Stein-
metz, 1980) suggested that women are as violent as men toward
their partners. Since then, researchers in the field of intimate
partner violence have argued for gender symmetry or asymme -
try of intimate partner violence. One side of the debate argued
that men’s use of violence against their partners differs from
women’s. Based on findings from national crime surveys (po-
lice, hospitals, shelters, courts) they considered that women are
more likely than men to be victims of violence and to be injured
during the intimate partner violence acts. According to these
researchers, the violence is rooted in gender inequality and
male dominance and there is no gender symmetry between the
violent acts of men and those of women (Dobash R. E. & Do-
bash R. P., 1977, 1978; Bograd, 1984; Dobash R. E., Dobash R.
P., Wilson & Daly, 1992).
The debate’s influence on gender symmetry vs. asymmetry
prevalence studies on intimate partner violence expanded
greatly in the second half of the 1990s. The major contribution
of research in the 90s was considered to be the distinctions of
different types of violence (Johnson, 1995; Dobash R. E. &
Dobash R. P., 1992; Lloyd & Emery, 2000; Riggs & O’Leary,
1996; Tjadenn & Thoennes, 1999) and the emphasis on “con-
trol”, (Johnson, 1995; Dobash & Dobash, 1992; Pence & Pay-
mar, 1993) seen mainly in the feminist literature, which argues
that intimate partner violence is a problem when men use vio-
lence to control their wives (as cited in Johnson & Ferraro,
Recently the research in intimate partner violence, especially
male violence against females, has increased. According to
Garcia- Moreno & Watts (2011), after the publication of find-
ings from the WHO multi-country study in 2005, the research
on intimate partner violence increased fourfold, from 80 to 300
in 2008. However many countries, as in the Middle East and
West Africa, still lack reliable data (Garcia-Moreno & Watts,
2011) and much of the existing information cannot be mean-
ingfully compar ed .
A population-based survey is considered the most reliable
method for obtaining information on violence against women in
a general population. Since it uses randomly selected samples,
its results are representative of the larger population and it ac-
curately provides estimates of the prevalence of violence
against women. The population based surveys that most coun-
tries use are of two types: dedicated surveys designed to gather
detailed information on different forms of violence against
women, and larger scale surveys (generic surveys) designed to
gather data on broader issues such as poverty, crime or repro-
duction, but also have questions or a module of questions on
Copyright © 2013 SciRes.
violence against women. According to Walby and Myhill
(2001), many national representative surveys on violence
against women and domestic violence have been conducted all
over the world during the 1990s, particularly in Europe and
North America, such as The National Violence against Women
survey conducted in Australia, France, Germany, Sweden and
the United States of America. Based on the findings it seems
that generic or dedicated surveys have a strength and a weak-
ness (Walby & Myhill, 2001). The strength of the generic sur-
vey is that it covers a lot of people for a certain period of time,
but there is a dilemma regarding the use of shorthand and
screener quest ions, since it might hinder the victim’s disclosure
if she has not been previously identified as a victim. Dedicated
surveys were shown to be more effective in collecting data on
violence against women, since their methodology had been
previously adapted to the issue, but their disadvantage is related
to lack of funding and resources.
Furthermore, the interviewing process in surveys is consid-
ered an important step toward enhancing the disclosure of vio-
lence by sharing the story of abuse and recovery. In this regard,
Walby and Myhill (2001) identified three important aspects:
privacy, interviewing skills and gender of the interviewer. It has
been proven that when the interview is done with nobody pre-
sent; when the interviewer is female and specially trained, the
disclosure rate is higher. For instance, in the British Crime
Survey, when the woman’s partner was involved in completing
the questionnaire the rate of reporting lifetime violence dropped
to less than half the rate reported when no one else was present,
from 23% to 10% (Walby, 2000). Similarly, Sorenson et al.
(1978) found that those interviewed about sexual assault were
1.27 times more likely to reveal a sexual assault if they were
interviewed by a woman than by a man.
Standard methodology for the implementation of surveys on
violence against women within the framework of official statis-
tics has not yet been developed at the international or suprana-
tional level (UN Study on Violence against Women, 2005). For
example, many of the prevalence estimates for intimate partner
violence are not comparable because of methodological differ-
ences in how violence has been defined and measured. In this
regard Tjaden (2005) mentions three factors related to lack of
accurate data in these surveys: narrow focus, differing time
frame and inadequate measurement. It should be emphasized
that there is an ongoing effort of international organizations to
support internationally comparative surveys. In this regard, two
important studies should be mentioned: the International Vio-
lence against Women Surveys (IVAWS) coordinated by the
European Institute for Crime Prevention and Control and the
WHO Multi-Country Study on Domestic Violence and
Women’s Health. The IVWAS has been conducted in ten coun-
tries (Australia, Denmark, Canada, England, Wales, Finland,
Italy, Netherlands, Sweden and Switzerland). It is a compara-
tive study specifically designed to target’s men’s violence
against women, including physical and sexual violence (John-
son, Ollus, & Nevala, 2008). The WHO Multi-Country Study
has been conducted in 15 cities and 11 countries (Bangladesh,
Brazil, Ethiopia, Japan, Namibia, Peru, Samoa, Serbia and
Montenegro, Thailand, and the United Republic of Tanzania)
and its aim was to collect data on intimate partner violence,
sexual assault, child abuse and the consequences of violence
(Garcia- Moreno, Jansen, Ellsberg, Heise, & Watts, 2006).
In addition to population studies, service based studies are
considered another source of information on violence against
women. Service based studies include data about violence
against women gathered through public and private agencies,
supporting women who have experienced violence, including:
health centers, police stations, shelters, legal and advocacy
organizations and lawyers’ associations. Although these data
could not be used for measuring prevalence of the phenomena,
they contribute to understanding risk factors that influence vio-
lence and help in designing adequate policies to protect women
on the supranational level (UN Study on Violence against
Women, 2005).
Qualitative studies are considered another important and
necessary method of data collection on violence against women.
Mostly, the qualitative studies are based on in-depth interviews,
focus groups or participatory research (Michau, 2002; Kilonza
et al., 2003; Sagat, 2000; Zimmerman, 1995 cited in Ellsberg &
Heise, 2005: pp. 74-81). Taking into account that qualitative
research is considered useful for describing complex phenom-
ena, its major strengths in understanding intimate partner vio-
lence are: collection of data on women’s experiences of vio-
lence in their words; their thoughts on the causes of violence;
rich details describing the cultural and local context in which
violence occurs; understanding of contextual and setting factors
as they relate to the intimate partner violence; a naturalistic
setting of data collection and the women’s interpretation of
intimate partner violence. Findings from qualitative research
are mainly presented through narratives, case studies, descrip-
tions and quotes. In addition, qualitative studies are of impor-
tant relevance in designing preventive campaigns, evaluating
and monitoring interventions and engaging the community in
fighting violence against women (UN Study on Violence
against Women, 2005).
Sampling Frame
There are great variations from year to year in the population
studies used for researching violence against women. In the
beginning of the 70’s most research on violence against women
was based on non-representative clinical samples. Only in the
late 70’s the first study was done with representative samples
(Gelles, 1980). According to the findings of this study (Straus,
1977, 1978) and studies conducted by Strauss, Gelles,
Steinmetz Gelles (1980) and Steinmentz (1977, 1978) the gen-
der neutral view of intimate partner violence became popular,
which suggested that females are equally violent as males in
relationships. Similarly, feminist scholars at that time con-
ducted studies on the specific factors related to male violence
against women. Feminist theoerists suggested that there is no
gender symmetry in perpetration of violence and only men are
socialized to believe that perpetrating violent behavior is ac-
ceptable in romantic relationships (Johnson, 1995). Further-
more, according to the feminist approach, male violence against
females is situated within the history of family and larger con-
text of gender inequalities that gives a man the right to domi-
nate the home and subordinate his partner (Yllo & Bograd,
1988 cited in Johnson, 1995: p. 283). The feminist approach
relied on research designs that collected data with those agen-
cies that came in contact with hospitals, shelters and other rele-
vant stakeholders.
However, even though the methodology used in population-
based surveys is considered the most reliable for collecting
accurate data on violence against women, these studies have
been subject to many critics, especially feminist scholars
Copyright © 2013 SciRes. 561
(Johnson, 1995; Johnson & Ferraro, 2000; Johnson & Leone,
2005; Johnson, 2006). Although the survey aimed to gather
information from target samples, according to feminist scholars
non-respondents’ responses may have differed drastically from
the respondents’ responses (Johnson, 1995). Additionally, the
accuracy of the gathered data depends on the response rate. A
man who assaults his wife probably will not participate in the
survey, whereas the battered wife will be scared to answer the
questions. Conversely, according to Ruiz-Perez et al. (2007),
the weakness of clinical based samples (hospitals, shelters) lies
in the data gathered from reported cases that experienced the
most severe type of violence, and the effect of trauma will ob-
struct their responses as well. Accurate data on violence are
lacking, since the less severe cases of violence are excluded.
Furthermore, most of the national surveys tend to exclude
certain populations from the studies such as temporary residents
of shelters and hostels. These particular population groups are
more likely to be subject to violence than those who are in-
cluded in the sampling frame. Additionally, these women are
more likely to experience violence in its aftermath than the
general population. Therefore, the exclusion of these categories
under-represents the accurate rate of violence in national sam-
ples (Walby & Myhill, 2001).
Operational Definition of Violence
The most difficult and controversial issue in studying vio-
lence against wome n i s re lated to its operational definition. This
happens because most common terminology comes from di-
verse theoretical perspectives and has different meanings in
different regions (Walby & Myhill, 2001). However, despite
the acknowledgment of scholars on the need for uniformity in
definitions and measurements of violence against women, there
is no formal mechanism for arriving at that desired uniformity
(Saltzman, 2004).
In the beginning of 1970, due to increased interest in the is-
sue of violence and its varieties, many studies have been con-
ducted and the main aim of these studies was to refute the gen-
eral conception that violence happens due to psychological
disorders, and to provide opinion with valid and reliable data.
Therefore, the researchers tried to improve the methodological
techniques on data collection and tried to expand knowledge on
the explanation of violence. However, despite their struggle to
provide reliable and valid data on domestic violence, research
(Kempe et al., 1962; Straus et al., 1980; Giovannoni & Becerra,
1979) in the 70’s was facing many problems regarding the
conceptualization of violence and its measurement (as cited in
Gelles, 1980).
Additionally, according to Klipatrick (2004), each perspec-
tive given in definition of violence against women historically
contains some problematic issues, especially regarding the nar-
rowness of focus. For instance, the criminal justice perspective
is considered a narrow definition since it includes only violent
acts that are defined legally as crime, ignoring other violent acts
such as psychological or emotional abuse and deprivation. Be-
cause public health perspective includes non-violent acts such
as emotional/psychological abuse, deprivation and neglect, it is
considered broader compared to criminal justice perspectives,
but the tendency to focus on sexual abuse rather than other
types of violence is considered narrow and problematic. Fina lly,
the human rights perspective is considered broader since it
includes all types of violent crimes against women: psycho-
logical abuse, harmful traditional practices such as genital cut-
ting, forced marriages and state tolerated discrimination against
Feminist scholars suggest that even broader definitions of
violence against women should be used, since narrow definition
lowers incidence and prevalence of phenomena (DeKeseredy,
2000 as cited in Kilpatrick, 2004: p. 11). According to feminist
scholars, we have to be very cautious when measuring intimate
partner violence since we have to differentiate between two
types of intimate partner violence: common violence that refers
to conflicts in the relationship and patriarchal terrorism that
uses violence in order to show power and control over the fe-
male (Johnson & Ferraro, 2000, Johnson, 2011 ). Failure to
make such distinctions could lead to incorrect estimates re-
garding the frequency of common couple violence against men
and the conclusion that there is a widespread “battered hus-
band” syndrome (Johnson, 1995: p. 292). Recent papers (John-
son & Ferraro, 2000; Johnson & Leone, 2005; Johnson, 2006)
make further distinctions between intimate partner violence
such as Intimate Terrorism, Violent Resistance, Mutual Vio-
lence Control and Situational Couple Violence. Intimate Ter-
rorism refers to a relationship where only one of the partners is
violent and controlling, and the other partner is either nonvio-
lent or violent but not controlling. Violent Resistance refers to
cases when one of the partners is violent but not controlling,
while the other partner is violent and controlling. Situational
Couple Violence refers to non-controlling violence that occurs
in a relationship between partners. Mutual Violence Control
refers to controlling violence between partners whereas both
partners are violent and controlling. Furthermore, findings from
the study conducted by Graham-Kevan & Archer (2003) on a
mixed population and a shelter sample, using a CTS style ques-
tionnaire, support earlier findings from the study conducted by
Johnson & Ferraro (2000). Based on these findings (87%) of
intimate partner violence was perpetrated by males, while
common couple violence was almost symmetrical, perpetrated
by (45%) males and (55%) females. As was expected, the shel-
ter population experienced 70% of all intimate partner violence,
while 94% of common couple violence was found in the com-
munity sample. Finally, over the past decade, there is a growing
body of research that demonstrates the existence of different
types of intimate partner violence such as (Graham-Kevan &
Archer, 2003; Johnson, 1995; Johnson & Leone, 2005; Johnson
& Ferraro, 2000; Johnson, 2006, Kelly & Johnson, 2008).
Similarly, Ruiz- Perez et al. (2007) and Walby (2004), refer-
ring to the UN Declaration on Elimination of Violence Against
Women, consider that, although the broader definition of vio-
lence is complex and includes most of the women’s violent
experiences, the terms used within the definition tend to lose
their power and m e a ning.
Measuring Vi o l e nce ag a i n st Women
There are many instruments used to measure violence against
women such as: Conflict Tactic Scale; Psychological Mal-
treatment of Women Index; and Measure of Wife Abuse and
Index of Spouse Abuse (Hegarty, Sheehan, & Schonfeld, 1999).
The Conflict Tactic Scale developed by Strauss and Gelles
(1979) is considered the most widely used measure for assess-
ing intimate partner violence and is the only standardized in-
strument (Dietz & Jasinski, 2007). The original CTS scale
measured acts used in conflicts, verbal and physical aggression
and it focused on the severity and frequency of such acts. In
Copyright © 2013 SciRes.
addition, this scale is used to measure husband to wife violence,
wife to husband violence and parent to child violence. The
revised version, which is called CTS2, includes also some
questions on sexual violence (Ellsberg & Heise, 2005). Besides
its wide usage, the CTS scale was the object of many criticisms.
First it focuses mainly on physical abuse, with some items on
verbal aggression, while it ignores other forms of violence,
especially psychological violence (Hegarty, Sheehan, &
Schonfeld, 1999). In addition, controversial findings obtained
from larger scale surveys concluded that women are as violent
as men and suggested so-called gender symmetry. Feminist
scholars stated that these findings are opposite to the results
gained from the studies done in shelters and social services,
since the scale is not able to distinguish among types of inter-
personal violence (Johnson, 1995). Furthermore, most of the
research suggests that female violence is rather related to self-
defense than a tendency to injure somebody (Ellsberg & Heise,
2005). CTS has also been criticized for not taking into consid-
eration the context, meaning and motive of the violent acts
(Schwartz, 2000) and their impact, since the same acts may
cause different injuries and have different meanings for men as
for women (Walby, 2001).
Nevertheless, other instruments were also predisposed to-
ward criticism. For instance, Psychological Maltreatment of
Women mainly focuses on emotional abuse and ignores other
forms of violence. The Index of Spouse Abuse and Measure of
Wife Abuse is a 30-item self-report scale that focuses on
broader aspects of partner violence, respectively in physical and
non-physical violence. However, both of them were validated
in small samples (Hegarty, Sheehan, & Schonfeld, 1999). Re-
cently, a WHO questionnaire has been designed primarily for
measuring violence against women based on the previous re-
search experiences. This questionnaire gathers data on physical,
sexual and emotional forms of violence. In addition, it provides
us with the frequency of violence, duration of violence, health
consequences and response to abus e (Ells b e r g & H e i s e , 2005).
Finally another problematic aspect, of instruments used to
measure violence against women, is related to different time
frames in which violence has been measured (Tjaden, 2005;
Wal by & Myhill, 2001). In addition, accordi ng to Wal by ( 2001)
the prevalence and incidence should be defined separately.
Some of the instruments ask about the violence experienced
during the person’s lifetime while others ask about just the last
year. As a result, comparisons between the findings would be
impossible. Furthermore, the recall bias in surveys asking about
violence experienced during one’s lifetime would lead to an
underestimation of violence, while the response rate for the last
year will give more accurate e stimates of viole nce .
Risk Factors Related to Violence against Women
There is no single, definitive cause of family violence, and
many people—regardless of gender, age, race, ethnicity, educa-
tion, cultural identity, socioeconomic status, occupation, relig-
ion, sexual orientation, physical and mental abilities or person-
ality—may be vulnerable to abuse at any stage of their lives.
There are numerous factors associated with violence and find-
ings from the study WHO multi country study on women’s
health and domestic violence found that many similarities
across sites in terms of risk factors f or in timat e part ner vio lenc e
(Abramsky, Watts, Garcia-Moreno, Devries, Kiss, Ellsberg,
Jansen & Heise, 2011; Heise, 2012). However, an ecological
model could be considered a most useful model to explain and
describe diverse factors related to each other, and could classify
factors as follows: socio-cultural risk factors, social structural
risk factors, family factors and individual factors for victims
and perpetrators. Many studies were conducted to identify indi-
vidual risk factors related to violence; however most of the
research on risk factors for violence is focused on male perpe-
trators (Tolan, Gorman-Smith, & Henry, 2006). According to
Kantor and Jasinski (1998) male perpetrators are characterized
by: low self-esteem and low impulse control; anxiety disorders,
depression, antisocial personality disorder and substance abuse,
poor social skills and insecurity (as cited in (Carlson, Worden,
Ryn, & Bachman, 2000; Tolan, Gorman-Smith, & Henry,
2006). In addition sexual abuse as a child has been identified as
a risk factor in males for sexual offending (Crowell & Burgess,
1996). Furthermore, the experience of violence in the family
and corporal punishment as a child are considered as risk fac-
tors for intimate partner violence as an adult (Riggs & O’Leary,
1996; Abramsky et al., 2011). Age is identified as another indi-
vidual risk factor for victims and perpetrators; the risk is higher
for younger partners. Hence, when analyzing the risk factors for
victims there is always some doubt if these factors are conse-
quences of victimization or risk factors as such. This ambiguity
is especially related to social isolation and substance abuse
(Crowell & Burgess, 1996).
Family risk fa ctors are related to relationshi p factors and dy-
namics within the family or couple that lead to violence. There
is little research done in this regard. However some of the iden-
tified factors are: poor problem-solving and communication
skills, relationship status, and economic dependency of females
(Carlson, Worden, Ryn, & Bachman, 2000). Stress is consid-
ered another risk factor for violence against women, especially
when the family experiences poverty, immigration, discrimina-
tion or medical problems (Levy, 2008). Moreover, based on the
findings of a UN Study (2005) the following factors were iden-
tified: male control of wealth and decision-making authority
within the family; a history of marital conflict; and significant
interpersonal disparities in economic, educational or employ-
ment status.
Social structural risk factors are related to economic status
and community factors. Most research concludes that poverty;
low-income rates and unemployment of males could be consid-
ered risk factors. Community risk factors mainly related to
services offered for victims of violence, the stigma accompa-
nying these services, and the phenomenon of violence in gen-
eral (Carlson, Worden, Ryn, & Bachman, 2000). Furthermore,
findings from the UN Study (2005) identify similar and differ-
ent factors related to the community aspect: women’s isolation
and lack of social support; community attitudes that tolerate
and legitimize male violence; and high levels of social and
economic disempowerment.
Socio-cultural risk factors, according to the findings of most
studies, relate to cultural norms and traditions that promote the
so called culture of acceptance of violence, which could be
considered another risk factor for victimization. Some of the
studies find that patriarchal families, cultures accepting of vio-
lence and gender stereotypes are considered major risk factors
(Crowell & Burgess, 1996). The UN Study (2005) identifies
gender roles, male dominance and tolerance of violence for
conflict resolution as major risk factors related to socio-cultural
Copyright © 2013 SciRes. 563
Ethics in Researching Violence against Women
Research on violence against women is a very difficult and
challenging process due to the sensitive nature of the pheno-
menon. Despite the sensitivity, reliable information on violence
against women is a crucial element for prevention, treatment
and elimination of different forms of violence. In this regard
WHO, when conducting research on gender based violence in
1999, published guidelines for safety and ethics, addressing the
safety of both respondents and interviewers; ensuring the pri-
vacy and confidentiality of the interview; providing special
training to interviewers on gender equality issues and violence
against women; providing information and referrals for re-
spondents in situations of risk; and providing emotional and
technical support for interviewers (UN Study on Violence
against Women, 2005: p. 60).
According to Ellsberg and Heise (2002), some major ethical
principles should be taken into account when conducting re-
sea rch on gend er based violence: non- malfeasance, benef icen ce,
respect for individuals and justice (balancing risks and bene-
The major ethical concern in studies on violence against
women is related to the principle of non-malfeasance or mini-
mizing the harm. This principle is related to the fact that re-
spondents who participate in the study are more likely than
others to suffer physical harm if their partners find out the aim
of the study. WHO suggested some alternatives for minimizing
harm to respondents: interviewing only one woman per house-
hold; not informing the wider community that the study in-
cludes questions on violence; not conducting research with men
in the same clusters while women are interviewed; conducting
the interview in complete privacy; using dummy questionnaires
if somebody enters the room; usi ng candy and games to distract
children; providing information on services available to protect
victims of violence (Ellsberg & Heise, 2005: p. 39). Addition-
ally, taking into consideration that interviews about violence
may provoke intense emotions and painful feelings, the inter-
viewers should be trained to respond in such situations and end
the interview by e mphasizing the woman’s strengths.
The principle of beneficence for research on violence relates
to the maximizing of benefits for respondents that participate in
the study. Telling the story of violence is considered a trans-
forming experience for the women, since they did not have an
opportunity to talk about it before (Ellsberg & Heise, 2005: p.
43). Another aspect of beneficence relates to methodological
considerations that maximize disclosure, including: specific
wording and templates for questions; interviewers’ sex; attitude;
skills; empathy and trust. Finally, the major consideration of the
beneficence principle is associated with findings that lead to
social change. According to WHO guidelines, researchers and
donors have an ethical obligation to help ensure that their find-
ings are properly interpreted and used to advance policy and
intervention development (Ellsberg & Heise, 2002: p. 39). Ad-
ditionally, the principle of respect for individuals is considered
an important ethical principle that should be taken into account
when conducting research on violence against women. Respect
involves two important considerations: respect for autonomy
and protection of vulnerable people. These considerations relate
to individual informed consent procedures that ensure respon-
dents understand the purpose of the research and that their par-
ticipation is voluntary (Ellsberg & Heise, 2005: p. 36). Besides,
there is ongoing debate among researchers regarding the ade-
quate form of informed consent, since some of the researchers
believe it is important to inform participants regarding the
questions on violence, whereas other researchers think this will
hinder the disclosure and increase the respondents’ anxiety.
Finally, the principle of justice in research is related to balanc-
ing the risks and benefits to women who participate in the study.
In this regard, the risk of women participating in the study is
fairly large, but the risk of silence, ignorance and inaction to-
ward the violence is large as well. Therefore the researchers are
guided continuously to balance these two realities (Ellsberg &
Heise, 2005).
Improving Future Studies
Research on violence against women has provided important
information regarding patterns, prevalence, risk and conse-
quences of this threat to female well being. Since the identifica-
tion of violence against women as a problem worthy of study in
1970, progress has been made in understanding physical, psy-
chological and sexual violence against women. Additionally,
the research on violence against women in the past two decades
has improved and is considered a crucial element for designing
prevention and treatment programs for violence against women.
Although methodological improvements have been made in
later studies, a review of the literature shows many limitations
and restrictions that should be taken into account when re-
searching violence against women.
Taking into consideration the phenomenon’s complexity,
when conducting research on violence against women a mixed
methodology should be used, combining qualitative and quan-
titative methods. In this regard, quantitative methods will pro-
duce information that could be summarized in numbers and
percentages of women who experienced violence, while quali-
tative methods will enable detailed and in-depth information to
be gathered from the experiences of violence. Consequently the
combination of both methods would provide a truthful picture.
Recently researchers on violence against women encourage the
use of more than one method when analyzing the same issue.
Furthermore, considering that the research’s aim relies on un-
derstanding the phenomenon of violence against women and
designing effective preventive and awareness-raising programs,
I think that intimate partner violence should be studied more
carefully and as a product of the social processes, contexts and
interactions in which the phenomenon exists. Intimate partner
violence is a subjective and personal issue and the knowledge
that we gain through mono-method studies, particularly quanti-
tative studies, is rather vague and superficial. Therefore, I
would strongly recommend the analysis of intimate partner
violence through the assessment of social contexts, individual
differences and multiple realities based on social, political,
cultural, ethnic, gender and disability values. I think that
through such analysis a deeper and richer understanding will be
reached. Integrating qualitative methods in the study of intimate
partner violence will enhance understanding of the problems
and research findings, interpret inconsistent results and advo-
cate for social change.
Abramsky, T., Watts, Ch. H., Garcia-Moreno, C., Devries, K., Kiss, L.,
Ellsberg, M., Jansen, H., & Heise, L. (2011). What factors are asso-
ciate with recent intimate partner violence? Findings from the WHO
Copyright © 2013 SciRes.
Copyright © 2013 SciRes. 565
multi-country study on women’s health and domestic violence. BMC
Public Health, 11, 109. doi:10.1186/1471-2458-11-109
Bogard, M. (1984). Family systems approaches to wife battering: A
feminist critique. Americal Journal of Orthopsychiat ry, 54, 559- 568.
Carlson, B. E., Worden, A. P., Ryn, M., & Bachman, R. (2000). Vio-
lence against women: Synthesis of research for practitioners. Final
Report submitted to the National Justice, grant number 98-WT-
VX-K011. Washington, DC: US Department of Justice.
Crowell, N. A., & Burgess, A. W. (1996). Understanding violence
against women. Washington, DC: National Academies Press.
Dobash, R. E., & Dobash, R. P. (1977, 1978). Wives: The “appropri-
ate” victims of martial violence. Victimology: An International
Journal, 2, 426-442.
Dobash, R. E., Dobash, R. P., Wilson, M., & Daly, M. (1992). The
myth of sexual symmetry in marital violence. Social Problems, 39,
71-91. doi:10.2307/3096914
Dietz, T. L., & Jasinski, J. L. (2007). The evect of item order on partner
violence reporting: An examination of four versions of the revised
Conflict Tactics Scales. Social Science Research, 36, 353-373.
Ellsberg, M., & Heise, L. (2002). Bearing witness: Ethics in domestic
violence research. Violence against Women, 359, 1599-1604.
Ellsberg, M., & Heise, L. (2005). Researching violence against women:
A practical guide for researchers and activists. Washington, DC:
World Health Organization.
Ellsberg, M. (2006). Violence against women: A global public health
crisis. Scandinavian Journal of Publi c Health, 34, 1403-4948.
Garcia-Moreno, C., Jan sen, H. A., Ellsberg, M., Heise, L., & Watts, C.
H. (2006). Prevalence of intimate partner violence: Findings from
the WHO multi country study on women’s health and domestic vio-
lence. Geneva: World Health Organization, Department of Gender,
Women and Health.
Garcia-Moreno, C., & Watts, Ch. (2011). Violence against women: An
urgent public health priority. /1/10-085217.pdf
Gelles, R. J. (1980). Violence in the family: A review of research in the
seventies. Journal of Marriage and Family, 42, 873-885.
Heise, L., Ellsberg, M., & Gottemoeller, M. (1999). Ending violence
against women. Populatio n Information Programme, 27 .
Heise, L. (2012). Determinants of partner violence in low and middle
income countries: Exploring variation in individual and population
—Level risk. PhD Thesis, London: London School of Hygiene Tropi-
cal Medicine.
Hegarty, K., Bush, R., & Sheehan, M. (2005). The composite abuse
scale: Further development and assessment of reliability and validity
of a multidimensional partner abuse measure in clinical settings.
Violence and Victims, 20, 529-547.
Hotaling , G. T., Straus, M. A., & Lincoln, J. (1990). Intrafamily vio-
lence, and crime and violence outside the family. Family Violence,
Johnson, M. (1995). Patriarchal terrorism and common couple violence:
Two forms of violence against women. Journal of Marriage and
Family, 57, 283-294. doi:10.2307/353683
Johnson, M. (2006). Conflict and control: Gender symmetry and
asymmetry in domestic violence. Violence against Women, 12, 1003.
Johnson, M., & Ferraro, K. J. (2000). Research on domestic violence in
the 1990s: Making distinctions. Journal of Marriage and Family, 62,
948-963. doi:10.1111/j.1741-3737.2000.00948.x
Johnson, M., & Leone, J. M. (2005). The differential effects of intimate
terrorism and situational couple violence: Findings from the national
violence against women survey. Journal of Family Issues, 26, 322.
Johnson, H., Ollsun, N., & Nevala, S. (2008). Violence against women:
An international perspective. Berlin: Springer.
Johnson, M. P. (2011). Gender and types of intimate partner violence:
A response to an anti-feminist literature review. Aggression and Vio-
lent Behaviour, 16, 289-296. doi:10.1016/j.avb.2011.04.006
Kilpatrick, D. G. (2004). What is violence against women: Defining
and measuring the problem. Journal of Interpersonal Violence, 19,
1209. doi:10.1177/0886260504269679
Kelly, J. B., & Johnson, M. P. (2008). Differentiation among types of
intimate partner violence: Research update and implications for in-
terventions. Family Cour t Review, 46, 476-499.
Krug, E. G., Dahelberg, L. L., Mercy, J. A., Zwi, A., & Lozano, R.
(2002). World report on violence and health. Geneva: World Health
Riggs, D. S., & O’Leary, K. D. (1996). Aggression between heterosex-
ual dating partners: An examination of a courtship model of court-
ship aggression. Journal of Interpersonal Violenc e , 11, 519-540.
Ruiz-Perez, I., Plazaola-Castano, J., & Vives-Cases, C. (2007). Meth-
odological issues in the study of violence against women. Journal of
Epidemiol Community Health, 61, ii26-ii31.
Saltzman, L. E. (2004). Issues related to defining and measuring vio-
lence against women response to Kilpatrick. Journal of Interpersonal
Violence, 19, 1235. doi:10.1177/0886260504269680
Schwartz, M. D. (2000). Methodological issues in the use of survey
data for measuring and characterizing violence against women. Vio-
lence against Women, 6, 815- 838. doi:10.1177/10778010022182164
Steinmetz, S. K. (1977, 1978). The battered husband syndrome. Victi-
mology, 2, 499-509.
Straus, M. A. (1977, 1978). Wife beating: How common and why?
Victimology, 2, 443-458.
Straus, M. A., Gelles, R. J., & Steinmetz, S. K. (1980). Behind the
closed doors: Violence in the American family. Garden City, NY:
Sorenson, S. B., Stein, J. A., Siegel, J. M., Goldin g, J. M., & Burnham,
M. A. (1978). The prevalence of adult sexual assault: The Los An-
gelos epidemologic center catchment area project. America Journal
of Epidemiology, 125, 1154-1164.
Tjaden, P. (2005). Violence against women: A statistical overview,
challenges and gaps in data collection and methodology and ap-
proaches for overcoming them. Geneva: World Health Organization,
UN Division for the Advancement of Wom en.
Tjaden, P., & Thoennes, N. (1998). Prevalence, incidence, and conse-
quences of violence against women: Findings from the national vio-
lence against women survey. Research in Brief. Washington, DC: US
Department of Justice, National Institute of Justice, and US Depart-
ment of Health and Human Services, Centers for Disease Control and
Tolan, P., Gorman-Smith, D., & Henry, D. (2006). Family violence.
Annual Review of Psychology, 57, 557-583.
UN Study on Violence against Women (2005). In-depth study on all
forms of violence against women. 61st Session of the General As-
sembly Item 60(a) on Advanceme n t o f wo m e n.
United Nations General Assembly (1993). UN declaration on the
elimination of violence against women. New York: United Nations
General Assembly.
Walby, S. (2001). Improving the statistics on violence against women.
Statistical Journal of the Unit e d N a ti o n s , 22, 193-216.
Walby, S., & Myhill, A. (2001). New survey methodologies in re-
searching violence against women. British Journal of Criminology,
41, 502-522. doi:10.1093/bjc/41.3.502
World Health Organization (2002). Reducing risks promoting health
life. Geneva: World Heal t h Organization.