Much of the domestic violence and abuse literature contains reports of quantitative research approaches that quantify the experience, identify those at risk, and recommend interventions. Although important, these approaches often fail to describe the experience and understanding from the perspective of and in the language used by women who have experienced abuse. This article reports a phenomenographic study of six women's understanding of their experiences of domestic violence during their childbearing years, the time period associated with pregnancy and the first twelve months after birth. The women described domestic violence as being experienced as a loss of self, being controlled and destruction. This report presents the women's views of domestic violence as a complex and damaging phenomenon that is experienced by them in a number of qualitatively different ways.
Key words: Domestic violence, language, women, control, loss, phenomenography, interview, childbearing, pregnancy, abuse
Domestic violence affects one in five Australian women during their lifetime (Raphael, Lawrence, O'Toole, & Roberts, 1990) and one in ten during pregnancy (Webster, Sweett, & Stolz, 1994). Over the past three decades, the prevalence of physical abuse of women in the United States has been reported to be between 0.9 percent and 20.1 percent. Later prevalence studies have reported similar statistics (Cokkinides & Coker, 1998; Glander, Moore, Michielutte, & Parsons, 1998). Variations in the prevalence rates have been attributed to the measures of abuse used, populations sampled, and the timing, method, and frequency of questionnaire administration. Studies of the effects of abuse on the outcomes of pregnancy found links to preterm labour, spontaneous and surgical abortion rates, and incidence of low birthweight infants. Most studies have focused on physical abuse, although abused women have reported emotional abuse to be more damaging to their health in the long term.
The factors associated with abuse are important and must be considered by health care providers. However, they do not specifically reflect the personal experiences of women. Research on the nature of the phenomenon, especially as defined by those experiencing it, has been minimally explored. Some researchers have made important contributions by describing women's experience of domestic violence and abuse, though not specifically in the context of childbearing. In many of these studies women have been asked to describe their experiences, but not to provide their own interpretation of the meaning of the experience. The researchers used the description and applied their own interpretation, explanation or meaning to the experience. This does not provide information about how the women make sense of their own experience.
Spender (1990) suggested that researchers have muted women's voices by limiting
...professionals who work with women suggest that women often fail to see themselves as subject to domestic violence or in an abusive relationship...
the language and words that they are "allowed" to use to describe experiences such as domestic violence. Their voices are defined, and definitions fall into acceptable categories, such as physical or sexual abuse. Spender argued "...once certain categories are constructed within language, we proceed to organise our world according to those categories" ( p.141). For example, Ferraro and Johnson (1983), in examining women's rationalisations for staying in or leaving violent relationships, have used categories generated by previous researchers in studies of deviant behavior. Using preconceived categories of thought may affect, and even limit, the effectiveness of health professionals. Comments from professionals who work with abused women suggest that the women often fail to see themselves as subject to domestic violence or in an abusive relationship, despite histories consistent with currently used research categories, such as physical, sexual or psychological abuse (Ferraro and Johnson, 1983; Towns and Adams, 2000).
The purpose of this study was to identify and describe a small group of childbearing women's understanding of their experience of domestic violence. This study explored the meaning that women used to make sense of the experience, within the context of their own lives. The interpretation and language used to construct the categories of description were derived directly by the women rather than the researcher.
During data collection, the women suggested that the use of pregnancy only as a time frame for examining their experience of abuse was not useful, as their description and understanding was based around their lives as a continuum in which abuse and their childbearing experiences coexisted. The study was designed to focus specifically on the women's experiences of abuse during the childbearing period, the time associated with pregnancy and including the first twelve months after birth. The use of this time frame was consistent with other previously described research that identified these specific time frames in women's lives and looked at abuse as affecting the outcomes (Gielen, O'Campo, Faden, Kass, and Xue, 1994).
The Phenomenographic Approach
Phenomenography uses qualitative data to identify variation and similarity in participants' understanding of phenomena they have experienced. It focuses on collective and common conceptions or ways of understanding the world. Developed at the University of Gothenberg thirty years ago, phenomenography is a method for describing the way people understand their experiences (Marton, 1981).
Categories of Description
The central concern of phenomenography is awareness of meaning or understanding of phenomena that are contextualised in the world around us. The context of experience is influenced by factors such as past experience, beliefs, values, faith and culture. For example, experiences of domestic violence described by women one hundred years ago might be different from those described by the women in this study. Beliefs, values, morals, culture and time determine the number, nature and boundaries of each category of description (Marton, 1981). They are the outcome of reflection and thinking about the relationship between a participant (subject) and a phenomenon experienced (object). According to early phenomenographic researchers, a limited number of categories of description reflect a participant's understanding of the phenomenon, despite the repeated study of a phenomenon.(Barnard, McCosker, & Gerber, 1999; Marton, 1986).
Categories of description have whole characteristics that represent the central meaning of conceptions, similarities and differences in meaning, and outline the number of qualitatively different ways a phenomenon can be described, analysed and understood. According to Marton (1988), categories of description have four primary characteristics. Relational categories deal with the intentional or subject-object relations comprising the conception. Experiential categories are based on the experience of participants in the study. Content-oriented categories focus on the meaning of the phenomenon being studied. Qualitative categories describe the phenomenon (Marton, 1981).
Categories of description are an abstract tool used to characterise understanding of the phenomenon (Dahlgren & Fallsberg, 1991). They are identified from an analysis of data abstracted from interview transcripts or from some other form of communication. They provide the reader with a formalised summary of description. Parts of data are abstracted from entire transcripts and condensed to portray meaning. Categories of description represent the conceptions that have been discovered or revealed to the researcher through the process of analysis. The descriptions are written in the form "x is seen as y", and are supported by quotations from the transcripts (Lybeck, Marton, Stromdahl, & Tullberg 1988, p.101). Common meanings are presented as categories that may be then be compared and grouped as an expression of understanding. No single category of description can be attributed to individuals; rather their collective understanding is described in the results of the study.
Thus, categories of description are conceptions of the object of study within the context of the reality defined by participants. The categories of description may not represent the entire range of possible conceptions of a phenomenon. However, consistent and rigorous application of the methodology should ensure that they do express the meaning of the phenomenon within the context as experienced by the participants (Barnard, McCosker, & Gerber, 1999). Throughout the analytic process the fundamental characteristic of the phenomenon originating from the experience of those interviewed is preserved, as is the essential meaning as expressed in the organisation of the content of experience. The significance of the categories of description is exposed within the similarities and differences described by the outcome space and are open for interpretation.
There are a number of possible components of a category of description that have both external and internal horizons. These components comprise the referential or meaning aspects (what) and the structural aspects (how). The referential aspects refer to the global meaning that is attributed to a phenomenon, or what forms the conception. Structural aspects refer to the way in which the phenomenon and its component parts are delimited and related to each other in the form of external and internal horizons of the phenomenon. Internal horizon refers to how component parts of the phenomenon are understood and are related to each other. External horizon refers to the way in which the phenomenon is delimited from and related to its context. The term delimit refers to the suspension of attributed meaning or definition from a preconditional perspective (Marton, 1988; Sandberg, 1994).
The Outcome Space
The outcome space is a diagrammatic representation of the logical relations between categories of description. Similarities and differences from the experience and understanding of a phenomenon can be viewed as a system of conceptual order. Marton (1986) referred to the system as our collective intellect that forms a structured pool of ideas, beliefs, and facts that underlie a reflection and construction of reality. The phenomenographic approach describes the collective intellect in the form of an outcome space, which is an empirical map of the "qualitatively different ways in which people experience, conceptualise, perceive, and understand various aspects of, and phenomena in, the world around them" (Marton, 1986, p. 31).
Participants were recruited from a course conducted at a community-based counseling organisation. The women, who were either self-referred or had been referred by a counselor, had completed a 12 week course entitled "Freedom from Violence." Six White women who identified themselves as having experienced domestic violence while pregnant or during the first twelve months after their child's birth agreed to participate in the study. One woman was still living with her abusive partner and the rest were either divorced or had been separated for at least 12 months. Each family had one to four children, the youngest of whom was between one and six years old. The participants' educational background ranged from high school to college graduate. Two of the women were employed, while the others relied on social security (money from the government).
Protection of Human Subjects
Due to the sensitivity of the phenomenon under investigation, and the potential for psychological risk, a face-to-face interview was deemed to be the most appropriate method of data collection. Following approval by the Human Research Ethics Committee of Queensland University of Technology, ethical clearance to proceed with the research was given by the community organisation through which the women were recruited on the proviso that appropriate counseling was available for each participant if required. Each woman was given a letter outlining the purpose of the study, the voluntary nature of her involvement, and an assurance that confidentiality would be maintained through the use of pseudonyms in the reporting process. Each participant then signed a written consent form that was retained by the researcher.
The first author (HM) conducted one in-depth, unstructured interview with each woman at a mutually convenient place and time. The interview, which was a discourse conducted in the form of a conversation, allowed the woman to become aware and verbalise her own meanings. A non-technical question, "What led you to the support group?" was used to open the interview. The intent was to preserve the view of the participant. The unstructured nature of the interview allowed each woman to redefine the question to suit herself and generate meanings through an exploration of her own thoughts and feelings. The purpose of the interview was to engage in a dialogue that raised the phenomenon from being unreflected to being consciously reflected upon by each participant (Bruce, 1997).
The interviewer was sensitive to the interaction between herself and each woman, and adopted a "deliberate conscious naiveté" (Kvale, 1983, p.178), allowing each woman to explore her own meanings without imposing the interviewer's understanding. The interviews were recorded to minimise selection bias by the researcher, and then transcribed verbatim.
The qualitative analysis of data included seven processes: familiarisation, condensation, comparison, grouping, articulating, labeling and contrasting. This analytic technique is not sequential, but is performed in an iterative manner, until the analysis is complete. The analysis took the form of comparison of the data obtained from the women in order to focus upon describing the phenomenon in terms of the essential meaning. The aim of the analysis was to identify and describe ways of understanding domestic violence and to present the research findings in the form of categories of description and an outcome space.
Analysis of data identified that domestic violence was experienced as loss of self, being controlled, and destruction.
Conceptions of domestic violence arose from the women's experiences and are expressed in terms of their understanding and reflection. The categories of description are not attributable to any one individual. They are a collective description of the qualitatively different ways that the six women understood and made sense of their experience of domestic violence during pregnancy and childbearing. Analysis of data identified that domestic violence was experienced as loss of self, being controlled, and destruction.
First Category of Description: Domestic Violence Is Experienced As a Loss of Self
This category of description was distinguished by the understanding of the experience of domestic violence as psychologic and emotional damage.
Women described their understanding of domestic violence as a way of life that led to changes in perception of their self concept and a loss of self esteem.
Women described their understanding of domestic violence as a way of life that led to changes in perception of their self concept and a loss of self-esteem. The changes were described as a gradual change in the nature of their relationships with their partners. The women felt they were no longer the persons they were when they entered the relationships. This awareness occurred at different points in time, prior to leaving the relationship or seeking counseling, or afterwards during a course about abuse.
The women spoke of a sense of "wounding" associated with their partners' behaviours, which they viewed as character assassinating, critical and leading to a sense of worthlessness. For example, the following quote illustrates the manner in which a woman described the realisation that the abuse she experienced had affected her self-esteem:
You know how they have those big long knives with the curve in them. Well it felt like that two of them had attacked me like this and it was so destroying, that it was chopped up and that's how I felt inside. I didn't want to live. I just wanted to die. My self-esteem was zilch.
Experiences took the form of playing "mind games." The partners' behaviour turned the women into objects and subsequently allowed the partners to perceive it acceptable to denigrate them. The partners also failed to treat the women as adults. For example, one woman recounted her experience of being treated as a child and denigrated by her partner: "And I'd just stand there, I could not move. I was totally reduced to a child, in the power play. There was no doubt about it. He told me I was totally useless."
Women in the study reported that the demand for perfection from their partners contributed to their sense of worthlessness. The women noted that the men could behave as judge and jury in the search for perfection: "They make you feel as if you're to blame. I know I used to do it. I'd say, 'I'm sorry I didn't mean to hurt you.' But I hadn't done anything and it's this rebellion and this twisting."
As partners withdrew their commitment to the relationship, the women developed a conscious awareness that their partners' behaviours were significantly affecting their self-esteem. The withdrawal of support appeared to be linked to a perception by some women that their partners had a fear of their ability to provide financially. Some partners responded to the announcement of the first pregnancy, or subsequent, unplanned pregnancies, with open hostility or withdrawal of support. One participant stated, "It's like nobody's at home. I have only realised that since he left. He wasn't there any which way. He couldn't support me. If I was emotionally upset, [he would say] 'that's your problem.'"
The understanding of domestic violence reflected in the first conception was often associated with a second conception. This was more focused on the physical and emotional environment in which the women lived rather than their response to that environment.
Second Category of Description: Domestic Violence Is Experienced As Being Controlled
...the sense of being controlled became more evident after the women had left the relationship.
Some of the women became aware retrospectively of the sense of being controlled as a dimension of the experience of domestic violence. Although physical abuse was obvious when it occurred during the relationship, the sense of being controlled became more evident after the women had left the relationship. Control could be physical, sexual, social, or economic. The experience included a sense of being controlled by person(s) and events they had no power to change. The following quote illustrates the experience of sexual control:
He was raping me because our sexual relationship had not been good and I had not been willing. I have sometimes been an acquiescent partner and I view that a little differently, but there were times when I said categorically, "No, I don't want to have sex now for x, y or z reasons," and he would just ignore what I said and go ahead anyway.
Some women also experienced economic control. One woman discussed the control of money in the family: "...he had control of all the money. He would say that I could have what I wanted which was, I suppose, sort of true, but I couldn't go and ask for it."
For some women "being controlled" included an attempt by the partner to isolate them from their children: "I suppose the one thing that sticks in my mind was he kept referring constantly, 'I can't wait 'til when the children have all left. When are they going to leave?'...and he kept referring to just himself [as] P [partner] and I."
The most pervasive characteristic of this control was the women's awareness of the presence of jealousy and possessiveness in their relationships. The woman's sense of being controlled by her partner was dependent on the nature of the relationships among herself, her partner and her family. The control also extended to isolating the women from family and friends. For some women the controlling influence was their partner's family, particularly his mother. The family exerted control in a number of ways such as inherited expectations of roles, and "idealised" behaviour combined with covert or overt criticism. Women reported being reduced to feeling like a child with no rights, both in their own homes and in those of parents and parents-in-law.
Third Category of Description: Domestic Violence Is Experienced As Destruction.
The final conception of domestic violence was one that is often perceived as the most recognisable "face" of domestic violence: a destructive force within women's lives.
This category of description was characterised by the women's experience of the physical aspect of domestic violence. Destruction was described as being sudden, episodic and variable, and engendered a constant state of anxiety and fear. Physical violence was directed towards the woman herself and at inanimate objects such as cupboards, doors and phones. One woman described the following:
He wasn't punching and kicking and violent all the time. He was very aggressive. He used to lock me in the house and pull the phone out of the wall if I'd ring my sister and things like that.
Whether the destruction was directed at inanimate objects or at the woman herself, each woman expressed the feeling that the intended target was always herself:
The way I felt and had trouble dealing with, is that even though I am not on the receiving end of the breakage, that it is directed at me and that I am at fault. I am not being the perfect wife, therefore he is having this terrible anger.
The women described a feeling of lack of safety...even when they were not directly subject to physical abuse.
The women described a feeling of lack of safety for themselves and their children as a result of the feeling that the destruction was always directed towards them, even when they were not directly subject to physical abuse. This fear was reported by the women as inhibiting their lifestyle. For example, they restricted their social interactions with friends to times when their partner was not around, and used the phone in a similar manner.
The Outcome Space
The outcome space is the diagrammatic representation of the relationship between the categories of description (Figure 1). The manner in which women described their understanding of the experience of domestic violence is portrayed best as a series of circles. The circles are related structurally, not hierarchically. They are joined by straight lines in order to indicate the directions of influence in both directions. The referential aspect of the outcome space relates to differences in perception, language used, and the focus of experience at any one time. The women's awareness of their experiences varied among loss of self, being controlled and destruction. Within each of these qualitatively different conceptions there was a variation in the nature and extent of the experience. For example, some women experienced domestic violence more as destruction than as being controlled, but the degree of variation was not quantified in this research.
Statements made by some women described how they had to put bits and pieces of experience together to form a conscious awareness that they were subject to domestic violence before they were able to take any form of action. This was based on their recognition that they gradually developed an awareness of the impact and nature of their partner's behaviour. Awareness of the experiences that constituted one conception did not constitute an awareness that they were subject to "domestic violence." This was based on the observation by the women that they did not consciously label the behaviour as domestic violence when it began, rather the labeling often occurred after they had left the relationship. The outcome space is important because the three categories of description together provide an understanding of the experience of domestic violence during the childbearing years for the six women interviewed.
Figure 1. Outcome Space Of Domestic Violence During Childbearing Years.
The findings of this study in relation to the physically destructive nature of the experience of domestic violence mirror findings in previous studies. Ferraro and Johnson (1983, p. 334) also reported women experiencing feelings of inferiority with an associated loss of self-confidence as a result of years of criticism and ridicule. They described women entering a period of "emotional dormancy" to survive within the marriage.
The women in this study experienced domestic violence in different ways that were not simply categorised as physical or sexual abuse.
The women's voices in this study suggest that when assessing women for health effects of domestic violence it is important to acknowledge the individual nature of their experience.
The women's voices in this study suggest that when assessing women for health effects of domestic violence it is important to acknowledge the individual nature of their experience. They may use different language to describe the experience than that used by the health professional. Assessment tools that use preconstructed language and are administered early in the relationship between a woman and her care provider may lead to less complete identification of abused women. For example, a woman who had never been hit or assaulted by her husband might answer no to a question about physical abuse. This statement might obscure the fact that her husband often uses other forms of physically destructive and intimidating behaviour to make her fearful. In this study, encouraging language constructed by the participants assisted the women in revealing their experience in their own terms. Questions framed around a woman's responses to, and outcomes arising from, their partner's abusive behaviours may be more helpful in identifying women who are experiencing domestic violence.
Current legal and health responses to domestic violence tend to reflect the view that people who experience domestic violence report it as destruction. The majority of approaches minimise the impact of describing the experience of domestic violence as a loss of self and being controlled. This is similar to a health professional's response of treating a presenting problem, such as a broken arm, without questioning an inconsistent story given to explain the cause of it. There are frequent reports of health professionals overlooking or ignoring inconsistent evidence. In other studies abused women who presented with a poor self-concept, suicide attempts, or "nerves" reported receiving inappropriate treatment. Understanding the experience of domestic violence in greater depth will facilitate better identification of and responses to women as individuals with different life experience.
This study focused on a small number of urban women who had access to a program that supported their attempts to resolve their domestic violence experiences. Further research with participants of different ethnic and cultural backgrounds is needed to determine if there are cultural differences in the recognition, definitions, and responses to violence against women.
Effective societal responses to domestic violence require that health professionals are informed and understand the nature of the phenomenon from the perspective of those experiencing it.
Effective societal responses to domestic violence require that health professionals are informed and understand the nature of the phenomenon from the perspective of those experiencing it. To provide women with information to enable informed decision-making, health professionals need insight into the experience of domestic violence. The manner of framing questions and the language used when asking women about domestic violence influences the women's responses. Education about domestic violence for health professionals needs to incorporate women's language and view.
Heather McCosker, RN, RM, MN
Heather McCosker holds the following qualifications - Registered Nurse, Endorsed Midwife, Bachelor of Applied Science (Distinction), Master of Nursing, Graduate Certificate in Education (Higher Education), Doctoral Candidate. She is currently a Lecturer, School of Nursing, Queensland University of Technology, Brisbane, Australia, where her teaching and research focuses on midwifery and women's health. She has been a midwifery clinician and teacher for the past twenty years. Her research and publication focus over the past 10 years has been on abuse as it is understood by women and midwives, predominantly using phenomenography.
Alan Barnard, RN, PhD
Alan Barnard, holds the following qualifications - Registered Nurse, Bachelor of Arts, Master of Arts and Doctor of Philosophy. He is currently a Senior Lecturer in the School of Nursing, Queensland University of Technology, Brisbane, Australia. His PhD examined surgical nurses understanding of technology and he has published extensively in the area of nursing and technology. He currently supervises Masters and PhD candidates who are using phenomenography as a research approach to examine a range of issues critical to the ongoing development of nursing practice.
Rod Gerber, PhD, MedSt
Rod Gerber is Professor and Executive Dean of the Faculty of Education, Health and Professional Studies at the University of New England in Armidale, New South Wales (Australia). He holds the following qualifications - Bachelor of Arts, Graduate Diploma in Education, Bachelor of Education, Master of Education Studies and doctor of Philosophy. He has taught and researched in areas of professional education, and has published in the areas of learning, professional development, health and education. He has an extensive record in qualitative research methodologies, especially most recently in the area of phenomenography where he has co-authored a range of journal articles. He has supervised many post-graduate research students using phenomenography, and has examined a large number of theses using qualitative research methodologies.
Article published November 21, 2003
Barnard, A., McCosker, H., & Gerber, R. (1999). Phenomenography: A qualitative research approach for exploring understanding in health care. Qualitative Health Research, 9, 212-226.
Bruce, C. (1997). The seven faces of information literacy. South Australia: Auslib Press.
Lybeck, L., Marton, F., Stromdahl, H., & Tullberg, A. (1988). The phenomenography of the "mole concept" in chemistry. In P. Ramsden (Ed.), Improving learning: New perspectives (pp. 81-108). London: Kogan Page.
Mackey, T. F. (1992). A psychological model for analysis of outcomes related to trauma. In C. Sampselle (Ed.), Violence against women: nursing research, education and practice issues (pp. 45-68). New York: Hemisphere.
Marton, F. (1988). Phenomenography: Exploring different conceptions of reality. In D. Fetterman (Ed.), Qualitative approaches to evaluation in education: The silent revolution (pp. 176-205). New York: Praeger.
Newberger, E. H., Barkan, S. E., Lieberman, E. S., McCormick, M. C., Yllo, K., Gary, L. T., & Schechter, S. (1992). Abuse of pregnant women and adverse birth outcome.Journal of the American Medical Association, 267, 2370-2372.
Pizzey, E. (1974). Scream quietly or the neighbours will hear. Middlesex, England: Penguin.
Raphael, B., Lawrence, J., O'Toole, B., & Roberts, G. (1990). Detection and prevention of domestic violence in emergency departments. Summary report. Queensland, Australia: University of Queensland, Department of Psychiatry.
Saljo, R. (1994). Minding action: Conceiving of the action versus participating in cultural practices. Nordisk Pedagogik, 2, 71-80.
Sampselle, C. M., Petersen, B. A., Murtland, T. L., & Oakley, D. J. (1992). Prevalence of abuse among pregnant women choosing certified nurse-midwife or physician providers. Journal of Nurse - Midwifery, 37, 269-273.
Sandberg, J. (1994). Human competence at work: An interpretive approach. Goteborg: Bas.
Scutt, J. A. (1991). The domestic paradigm; Violence, nurturing and stereotyping of the sexes. Women's Studies International Forum, 14(3), 163-172.
Shipley, S., & Sylvester, D. (1982). Professionals' attitudes towards violence in close relationships. Journal of Emergency Nursing, 8(2), 88-91.
Spender, D. (1990). Man made language (3rd ed.). London: Pandora.
Walker, L. (1979). The battered woman. New York: Harper and Row.
Webster, J., Sweett, S., & Stolz, T. (1994). Domestic violence in pregnancy: A prevalence study. Medical Journal of Australia, 161, 466-470.
Women's Policy Unit, Queensland Government. (1992). Women's experience of crimes of personal violence. Brisbane, Australia: GoPrint.
Wyatt, G., & Peters, S. (1986). Issues in the definition of child abuse in prevalence literature. Child Abuse and Neglect, 10, 231-240.