Feminist Qualitative Research in the Postmodern Era: Integrating Theoretical Approaches

(Carmel Seibold, Australian Catholic University, Victoria.

email: c.seibold@patrick.acu.edu.au)

 

Abstract

 

This paper explores methodological issues in analysing data about single women’s experiences of midlife and menopause, and the processes of managing health and shaping identity at this significant life stage. It addresses questions such as, to what extent the research met the criteria of feminist research. As well it describes a process of arriving at a method of data analysis that took into account both the role of discourse and the phenomenology of lived experience in the construction of identity.

 

Introduction

Developing a methodology for certain types of qualitative research is an ongoing process. The nature of the problem to be investigated is fluid, only incompletely determined at the beginning of the study, and subject to change as the study progresses. The design cannot therefore be fully specified in advance, but rather emerges over time. The process followed, while by no means linear, involves making decisions about what Higgs (1998) has described as the research frame and research strategy. Within the research frame Higgs includes the internal context relating to the researcher’s personal frame of reference or philosophical and ethical stance and the external context involving the research question or topic and theoretical framework or paradigm. The research strategy involves making decisions about the research goals or purpose, the research approach and data collection and analysis methods.

 

This paper focuses on the research frame and the research strategy and describes a process in which I developed methodology and techniques as I interacted with the data and the literature in a study of single midlife women. I am using the term methodology as defined by Harding (1987, p 3), that is “a theory and analysis of how research does or should proceed” as opposed to method which refers to “a technique for (or way of proceeding in) gathering evidence”. While most qualitative researchers develop methodology as the study progresses not all adequately describe the process leaving the reader often with a sense that everything progressed in an orderly and sequential way. Social researchers, however, are increasingly becoming aware of how the researcher's self is actually part of the research process (Ellis, 1991; Kleinman & Copp, 1993), and feminist researchers emphasise exploring both the social construction of the research encounter and the research process as a lived experience (Oakley, 1981; Reinharz, 1992). Hence the emphasis in feminist writings on the personal pronoun.

 

Deciding on a Topic

My initial interest in midlife women, and menopause in particular, began in the mid 1980s when my older sister asked me, as a nurse and a sociologist, what I knew about menopause. My response was "very little" but I resolved to learn more. Over the next few years, despite engaging with an increasing body of medical literature, I became aware of the dearth of recent studies that were contextual in nature and went beyond assessing the safety or efficacy of various hormone treatments. During this period of increased interest I experienced a relatively early menopause at the age of 41. This event added a personal dimension to my interest in menopause. After further review of the literature it appeared that what was needed was a qualitative study which sought to provide an understanding of how women in Western societies negotiated midlife and menopause. I found after reviewing the available literature that I was particularly interested in studying single midlife women. The reasons were that they constitute a significant number of Australian women in midlife (one fifth of the population between 44 and 54, ABS, 1993) and much of the writing on menopause to date has assumed that all middle aged women are heterosexual and have partners. I was a single, divorced midlife woman myself, and although my experience was similar in some respects to that of my partnered friends it was also qualitatively different.

 

The Research Process

Framing the Study

Early reading in the area of menopause indicated two particular problems. Studies provided only a static picture rather than capturing a process of change, and researchers tended to ask participants questions pertaining only to the researcher's interest. I considered it important in any study of midlife to follow the women's progress for a period of time. I decided to interview the participants twice, twelve months apart, and ask them to keep a diary in the interim. Interviews and diaries were seen as providing the most interesting and rich data about the way in which midlife women negotiated midlife and menopause. An open ended interview schedule was designed since I wanted to ensure that the information I obtained was as untainted by my opinions and experience as possible. The request for journal keeping reflected the same rationale. The goals of the study were to provide research data on single women's experiences of midlife and menopause and to make those experiences available to other women.

Recruiting the Women

I set out to interview women who were single, that is had either never married, were separated, widowed or divorced and not currently in a partnership, defined as living with a partner, were in the midlife age group (40 to 55), and were menopausal. Midlife is popularly considered to begin at age 40 and statistics indicate that normal, as opposed to premature menopause, occurs between the ages of 39 and 59 with the average age being around 50 years (McKinlay, Biafano, & McKinlay, 1985; Walsh, 1978). Taking HRT or having had a hysterectomy was not a bar to participating in the study. Here, as at several stages, I learned from my data. After interviewing two women I realised I had inadvertently set up a situation whereby only menopausal women experiencing problems might be interviewed. The criteria for selection were changed accordingly, to seek women who were self assessed as midlife and subsequently identified at interview as pre menopausal, perimenopausal or menopausal.

 

The participants eventually recruited were aged between 44 and 55 at first interview. All the women were Australian born and English speaking. Because the recruitment was primarily via networking there was an unintended, although not unexpected bias towards woman with Anglo- Celtic middle class backgrounds. Eleven women had never married and were childless, one woman was widowed and eight women were separated or divorced. Eight of the previously married women had children.

Making Data

My initial research questions were concerned with identifying the impact of menopausal symptoms, and the debate surrounding HRT, on the lives and decision making processes of single midlife women, I therefore sought to establish the discourses of menopause women were exposed to in the academic, medical and popular literature. Magazines and popular texts for the period of the study were reviewed, as were mainstream medical sources.

 

This process of review and analysis of the representation of menopause in the popular media occurred concurrently with interviewing the women. Initially the interviews were very open ended but became more semi structured as I developed a fuller sense of relevant issues. My initial interview question was a broad one seeking to identify perceptions of midlife and menopause held by single midlife women. All the interviews commenced with the one open-ended question "Tell me what it means to you to be a midlife woman." Probes were included to ensure that certain areas were covered in the interview, such as whether or not the participant was taking HRT. More probes were added following subsequent interviews. For example, the first woman interviewed talked at length of needing to come to terms in midlife with the anger she felt toward her mother who had been dead thirty years. As a result of this interview a probe "relationship with mother", was added. In subsequent interviews relationships with sisters and family members were seen as significant and probes covering family relationships were included.

 

In making the request that the women keep a diary for 12 months I wondered how willing participants would be to do so considering the nature of the highly personal experience. I decided that I would make it as non-threatening as possible by not including complicated instructions on diary keeping. I provided a diary with simple instructions pasted in the front cover which were as follows

 

The weekly entry in your diary should only require about 5 to 10 minutes of your time. You are of course welcome to make entries at other times. Entries in the diary are to enable you to record your symptoms, feelings and health practices (including ways of relaxing and enjoying life!), and particular concerns or comments. I am also interested in dreams that you may see as relevant to your life stage.

 

The diaries sought to trace the women's experience of midlife and menopause over a period of twelve months, including the influence of the various discourses surrounding menopause and midlife health. The request to record their dreams was the result of a midlife friend reporting an increase in dreaming while menopausal, and also the Jungian literature on dream analysis for making sense of life experiences, including midlife. Seventeen women kept a diary.

 

The practice and ethics of making data

Interviews tended to be relaxed and in the form of conversations, although I monitored and limited my input in order not to dominate the process. The first interviews varied between 1-2 hours and the second, between 30 minutes to one hour. All women were given a written explanation of the study and signed a consent form at the first interview. Part of the explanation included the usual assurances that tapes and diaries would be kept in a locked drawer. Diaries would be destroyed, or returned to participants after transcription if they expressed a wish to have them back. Pseudonyms would be used when transcribing data, thus assuring anonymity and confidentiality. I reiterated this explanation at the second interview and the women were given the opportunity to mark anything on the first transcript they did not want included in the analysis. 

 

After conducting the first five interviews issues of an ethical nature beyond the mere formalities of normal ethical requirements began to emerge. The first of these was the realisation that the empathetic relationship which was set up during the interviewing process meant that participants revealed highly personal aspects of their lives far and beyond my expectations, and quite possibly theirs too. On two occasions the interview precipitated what could only be termed a cathartic experience as a result of a highly emotional revelation. I have written elsewhere of issues surrounding data collection in qualitative research studies from a feminist perspective (Seibold et al. (1994). A feminist research approach, which advocates interaction between the researcher and the participant and a willingness, by the interviewer, to invest personal identity in the relationship (Oakley, 1981; Pence & Shepard 1988) can be as problematic as the so called objective approach. Whose interests are being served and who holds the balance of power? It appeared that in the situation as established here the balance of power was with the interviewee during the making of the data, but I as the interviewer/researcher had the power of analysis afterwards. While the interviewer/researcher and the interviewee participate in a power relationship during the interview process, after data collection that power in terms of "apparatus of truth" referred to by Foucault (1980, p. 132) becomes problematic, since as the researcher I construct my interpretation of events.

 

The highly personal information revealed in several interviews raised the vexed question of informed consent. Despite clearance from the university ethics committee I found myself asking how realistic was it for me as a researcher to think that a signed consent was really consent to, in a sense, invade people's lives? Also to what degree is consent prior to an interview really informed consent? A strategy that I developed was to have the participants read the relevant form prior to the interview and sign the form after the interview had taken place. A further strategy (in keeping with a feminist approach to research) was to have the participants read the transcripts of the first interview at the second interview. Fortunately no one withdrew consent to use the data, a concern that I had since at this stage who owns the data? I took the position that since the women were free to withdraw consent at any time during the research, withdrawal of consent at this stage meant that data would not be used. I also considered sharing some of the conclusions reached and the way in which the data was to be presented in the final report. I found, however that apart from a few generalisations this was not possible. Acker, Barry and Esseveld (1983) suggest that a study based on feminist principles is adequate if the active voices of women are heard in the research account. As well as the women's voices I also saw the inclusion of contextual elements in the final report as important such as reasons for choosing HRT. For example, the choice of one woman to take HRT “because I was desperate and would have done anything” contrasting with decisions made by another woman who considered taking it as a form of prophylaxis in order not to "end up like the woman in the milk ad.

 

On the two occasions in which the interviews proved cathartic they took on elements of a therapeutic relationship. One woman revealed that she had been a victim of child abuse and another became very distressed when speaking of her relationship with her mother. After the interviews I thought about when and why an interview might or should be terminated. If you terminate an interview are you retreating from a relationship established? What responsibility do you have as a researcher to suggest or facilitate counselling? While in the two cases referred to it did not reach this stage, there was cause to consider the possibility, and whether this would constitute further invasion of privacy. As a feminist researcher where is the line between acting as a concerned woman and taking on a therapeutic relationship. Is there such a thing as the need for a "specialist" in this context? As a nurse I was not unfamiliar with the role of counselor but did not feel that it was appropriate in this context to assume this role. I did, however, use a degree of professional judgement in assessing the degree of distress experienced. I also ascertained that the women who had been a victim of child abuse had been seeing a professional counselor.

 

Ethical and the methodological issues alike raise the questions of knowledge construction and control. I monitored the power dynamics in order to provide a basis for drawing adequate conclusions about the findings. I found that knowledge construction occurred within the interview, since a number of women saw it as a way of gaining information and dealing with their fear and anxiety. Whenever information was shared I was always careful to emphasise what was my opinion, differentiating opinion from specific information where a source could be cited. Providing information was always a balancing act. If there was a request for specific information, and I considered I had information of value, it was usually provided at the end of the interview

 

The 40 interviews were transcribed in full but only sections of the diaries were transcribed. I decided not to transcribe the diaries in full but rather carry out initial analysis and later transcribe sections of the diaries relative to relationships of contents to categories identified in the interviews. Any new insights were also noted and categorised. I had a large body of data of 2000 transcribed pages, and 17 diaries varying in length between 2000 and 15000 words. The diary of 15000 words fortunately was on computer disc. The approaches to diary keeping varied, reflecting individual commitment and organisation.

 

It is a widely held rule of qualitative research that data collection and analysis must occur  simultaneously ( eg. Glaser & Strauss, 1967). But like many rules pertaining to method this fits poorly with reality. I found data built up and many months would elapse between periods of concentrated analysis. However, I always made it a point to listen to the tapes within 24 hours of the interview and read the transcripts immediately they were transcribed. I also did some preliminary coding directly on the transcripts at this time.

 

To develop a data storage and retrieval system I used a qualitative analysis computer program QSR NUD.IST (Non-Numerical Unstructured Data, Indexing, Searching and Theorising) (Version 4). QSR NUD.IST supports both coding and retrieval and theory building processes (Richards and Richards, 1991, 1994). Because long periods of time often elapsed between opportunities for concentrated analysis, keeping track of analysis using a manual system would have been a nightmare. But I also found that the computer could not solve all my problems. Returning to analysis after several months I found it necessary, not only to familiarise myself with the index system, but also to re-read the transcripts in order to contextualise the data. The effect of time in providing distance from the data was both negative and positive. It was negative in terms of the time it took each time to pick up the threads and it was positive in terms of being able to view the data with new insights. For example the data relative to the pre-midlife and pre-menopausal body at first appeared extraneous but with time and fresh insight it was seen as highly relevant to the way midlife women experienced their bodies, and constructed identity.

 

In initial analysis I drew on grounded theory techniques as described by Glaser and Strauss (1967), Strauss, (1987), and Strauss & Corbin (1990), as well as descriptions of inductive data analysis by Lincoln and Guba (1985). I did a line by line analysis of each transcript, a form of coding described as open coding, where concepts are identified in terms of their properties and dimensions and similar concepts are then grouped to form categories (Strauss and Corbin, 1990,  p. 65). The categories I created in the early stages were mainly descriptive and the early interviews determined the most obvious categories such as physical symptoms of menopause. The codes derived from each interview transcript were compared, one with another, as part of what Glaser and Strauss (1967, pp.101 -116) refer to as "the constant comparative method of analysis." An early index system included the following categories, self assessed menopausal stage, physical and psychological symptoms of menopause, body perception, body out of control, body changes, women's knowledge base and anticipation of menopause, seeking help, sources of information, perceptions of risk, HRT reaction, awareness of transition. Later categories I constructed in the process of direct questioning of the data and review of relevant literature. For example, after initial coding I asked the question "why are most of the women talking at length about experiences of the body prior to menopause". This led to an exploration of the literature around reflexivity, identity construction and embodiment, most particularly Giddens’ (1991) theory of the reflexive project of self “whereby self identity is constituted by the reflexive ordering of life narratives” (p. 244.). Giddens sees integrating bodily experiences, both conscious and unconscious as essential to this process. Having constructed a category " past bodily experiences" I then linked "past bodily experiences with "factors influencing identity construction" in order to further explore reflection on past body experiences as influential in identity construction in midlife.

 

As analysis proceeded, I saw the need to place more emphasis on the way the women constructed or talked about the self, including the embodied self. A type of analysis identified by Potter and Wetherall (1987, pp. 95-106) as fitting under the broad rubric of discourse analysis seemed appropriate, that is one that gives added emphasis to gendered subjectivity, as well as the influence on the women of social and scientific discourses in circulation. With this in mind categories were refined and further developed. I was able to collapse codes and categories, rename them or shift them to other parts of the index system. When I reached the stage of identifying broad topic codes or major categories with sub categories being attached below, these were then ordered into chapter headings. The data relating to midlife women and work was initially one category, but was later split into two to become a major category in terms of the impact of menopause on the working lives of women, and a category relating to negotiating work as part of a midlife identity. Following Lincoln and Guba (1985, p. 341) I saw this process as a construction rather than a theory.

 

Writing was a fragmented activity and I experienced a sense of frustration that, owing to time constraints, I could not engage more consistently with the data. I did, however, try to record any insights in memos. As the data firmed I sought patterns by constructing matrices summarising such things as menopausal symptoms and response to symptoms.  Huberman and Miles (1994) describe the interaction between display and emerging written text thus:

 

The display helps the writer see patterns; the first text makes sense of the display and suggests new analytic moves in the displayed data; a revised or extended display points to new relationships and explanations, leading to more differentiated and integrated text and so on (p. 433).

 

This process continued once writing began in earnest. The first part of the analysis relating to managing health was relatively easy to conceptualise. However analysis of subsequent chapters consisted of writing numerous drafts. Writing helped to make links between categories and see new ways of conceptualising topics. Richardson (1994, p. 516) in referring to this activity says "by writing in different ways , we discover new aspects of our topic and our relationship to it. Form and content are inseparable." Writing drafts was also aided by my ongoing engagement with a range of literature and encouraged me to ask further questions of the data. Returning to it to confirm new directions meant a continued interaction with the literature and the data. I was also fortunate in having one of the study participants prepared to read and comment on the construction of the thesis, most particularly the analysis chapters. This form of member check proved encouraging at my darkest moments, as well as going some way to assuring me that I had done justice to the women's stories. As a feminist researcher I saw this as particularly important.

 

Feminist Method in the Postmodern Era: A Research Journey

Is there a feminist method? This question took on significance as I began to grapple not just with the ethical issues already outlined, but also with methodological issues in research from a feminist perspective. Much writing on feminist method authoritatively asserts what is required for feminist research. A typical check list of what characterises feminist research is as follows: the principal investigator is a woman; the purpose is to study women and the focus of the research is women's experiences; the research must have the potential to help the subjects as well as the researcher; it is characterised by interaction between researcher and subject, non hierarchical relations and expression of feelings and concern for values (one or all may be incorporated); the word feminist or feminism is used in the report; non sexist language is used; the bibliography includes feminist literature (Duffy, 1985).

 

At first glance the requirements sound easy to fulfil and as a feminist I thought how can I do otherwise. As already discussed the requirements became less self evident as I embarked on a project. As the challenges of the project were faced, the contradictions became evident. Check lists such as the one identified appear to present a single methodological dogma, the only way of doing feminist research. Acknowledging that feminist research seeks to uncover the pervasiveness of gendered thinking, which uncritically assumes a necessary bond between being a woman and occupying certain social roles, does not necessarily help to uncover the ways in which women negotiate the world and the wisdom inherent in such a negotiation. Rather than a check list approach to feminist research and theory, I sought a more general interpretation. Lengermann & Niebrugge - Brantley, (1988) provided this by setting only three broad goals. Firstly, women's experiences are the major object of investigation; secondly the researchers always attempt to see the world from the vantage point of a particular group of women; and thirdly they are critical and activist in efforts to improve the lot of women and all persons.

 

While qualitative methods that acknowledge women's frame of reference as contextual and relational may often be the most appropriate research approach, what is crucial is that the method chosen must be the one most likely to yield fruitful answers (Du Bois, 1983; Reinhartz, 1992). The philosophical orientation of the researcher then becomes the most significant factor, and quantitative methods may be equally appropriate and serve to deconstruct empirical research within a patriarchal mode (Harding, 1986). I considered that a qualitative approach was the most appropriate method for research on single midlife women since the object of the research was to construct a picture of how women in the late 20th century experience midlife and menopause. Previous research into menopause has often been within fairly restrictive guidelines and the research design has yielded a narrow perspective. Early medical research, particularly, oversimplified women's experiences by examining only those aspects of women's lives that corresponded directly to the norms of men's development and experience (see Davis, 1980). Single women were virtually ignored. The aim of my research was to redress this imbalance by focusing on women's experiences, specifically single women's experiences.

 

The project met two of the criteria for feminist research as outlined by Lengermann and Niebrugge-Brantley (1988). Women's experiences were the major object of the investigation and the goal was to capture the reality of the experience from the vantage point of a particular group of women with emphasis on the subjective and contextual orientation. More problematic was the third criterion, critical activism and whether or not studies such as this will improve the lot of women. Klein (1983, pp 90-92) says, that it is in recognising  "an individual woman's personal problem as similar to many women's personal problems" and seeking to provide the insiders' perspective that converts a study "on women” to a study "for women”. The study also needs to be read by the women it concerns. Publishing, particularly in the popular media, is one way this may be achieved.

Method Making and Data exploration: a Complementary Process

Unlike the apparent sequential ordered model of a research process previously referred to, method making and data exploration happened together. As I grappled with questions such as what constituted feminist method, and carried out analysis I was exploring contemporary social theory, including feminist theory, and debating the extent to which a feminist poststructural approach to theorising should direct my analysis. After completing two thirds of the first round of interviews I was becoming aware of the part played by discourse in the construction of the subject, and the value of studying social texts from the point of view of how those texts (the participants' interviews) were constructed and how they reflected their sense of self. Since social texts not only mirror but also actively construct a version of things, the diversity of experience of the 20 women interviewed appeared increasingly to be related to age and experience and the choices the women had made. These choices appeared in turn to be influenced by discourse.

 

A significant factor influencing life choices and decisions made was the ways in which the women understood sex, gender, femininity and midlife and menopause. These derived from a range of sources and forms of knowledge production and were far from coherent. The data increasingly appeared to require analysis in terms of discourse and a poststructuralist feminist approach to analysis offered valuable insight. Always suspicious of "the flavour of the month" I made a point of reading and attending conference presentations where the authors identified their position as poststructural feminists. At times I found the analysis abstruse, and I was not sure that this approach met my need as a feminist researcher to be relevant and accessible. Weedon (1987), and others (Bordo, 1993; Hollway, 1984) offered an approach which I found to be appropriate for what was emerging in the data, that is, an enhancement of Foucault's interpretative analytics which looks at the relationship between subjectivity and meaning, meaning and social value, the range of normal subject positions open to women and how power and powerlessness may be invested in these. In the analysis it was becoming apparent that when women resist a particular subject position, they do so as a result of exposure to an alternative discourse, such as that offered by feminist discourse. After further reading and reflection the theoretical framework chosen incorporated Giddens’ (1991) notion of the “reflexive self” which emphasised agency in decision making, as well as drawing on feminist poststructuralist interpretations of the discursive and embodied construction of self. 

 

My philosophical approach to data collection and analysis was influenced by writers in feminist methodology (Du Bois, 1983; Harding, 1986, 1987; Klein, 1983; Oakley, 1981; Reinharz, 1992; Stanley & Wise, 1983, Stanley, 1990). I needed a method of analysis that acknowledged discourse in the construction of subjectivity and was therefore attracted to a form of discourse analysis described by Potter and Wetherall (1987, pp. 95-106) that concerned the language of the self or subject. They argue that any socio-psychological image of the self is linked to linguistic practices used in everyday life, and analysis of participant interviews must pay attention to how the individual constructs or talks about the self. Potter and Wetherall view the methods of conceptualising the self in different linguistic practices as having vital consequences for the positioning of people in society. These methods are not neutral or without impact, but rather produce senses of self that may be negative, destructive, oppressive, as well as senses which might change or liberate (Potter & Wetherell, 1987, p.110). The main object in this form of analysis is to displace attention from the self as entity, and focus on methods of constructing the self. In the past, a specific theory of self, such as trait theory, role theory, or theories from the humanistic tradition which emphasise the quest for self-fulfillment and self actualisation, usually directed the analysis. Rather than one approach directing the analysis of discourse, all become relevant, but only in terms of a number of possible methods used by participants for making sense, or creating a life narrative.

 

As I applied these principles to analysis, that is, directed attention to the way the women constructed a sense of self, I became uneasy about a tendency to sideline descriptions of phenomenological experience. With further analysis the significant role of embodied experience, including the lived experience of menopause , as well as reflexivity in identity construction became apparent. The writings of Dorothy Smith (1987, 1991) and Barry Turner 1992) proved valuable in my attempts to integrate two apparently diverse approaches. Smith in particular helped me deal with the tensions inherent in fore grounding language, while also wanting to do justice to embodied experience, as well as identifying how discourse analysis from a feminist perspective differs. In Texts, facts and femininity  (1991, p. 1) Smith contends that bringing an understanding of contemporary history and society, or sociocultural influences and women's experiences "into coherent relation to one another" is essential to engaging properly with either. She views sociology to date as creating "a construct of society that is specifically discontinuous from the world known, lived, experienced and acted in" (p.2). She stressed the need to develop a sociology that emphasises a mode of inquiry that begins "where people are" rather than where we think they might be. Turner (1992), while not specifically addressing the need to integrate approaches within one mode of inquiry, was useful in arguing for a sociology of the body that takes into account both social constructionism, post structuralism and the phenomenology of lived experience. He sees value in integrating apparently disparate theoretical approaches, and argues for a degree of theoretical reintegration and rapprochement that will "permit and promote a diverse tradition of social theory within a common framework of interest in the body" (Turner, 1992, p. 49). Feminists' interpretation of Foucault's theory of power/knowledge and the construction of subjectivity, along with insights from Smith and Turner, then, shaped analysis, particularly with respect to the interplay of women's experiences and sociocultural influences or discourse.

 

Both Smith (1991) and Potter and Wetherall (1987) view the text as the means by which the analyst gains insight into the world of the participant, and interviews are analysed for their textual form and representation of participation in social relations. The researcher as analyst recognises that the participant is the active writer of the text and is drawing on her/his experience to construct the text, just as the analyst is drawing on her/his experience and knowledge in reading and interpreting the text. The researcher as analyst then is interested in constructing the way the participant co-orders experience and brings into being the social self. Smith (1991, p.10)  like Weedon, also makes the point that from a feminist perspective the research aims, without a totalising theory, to explore "the relations of ruling", organising peoples’ lives. This term is used to identify any number of theoretical positions, including that of Foucault, that emphasise the historical and sociocultural influences on participants.

 

Smith further identifies ways discourse analysis from a feminist perspective is different to other forms: First, the emphasis placed on the researcher working from the insider's standpoint; second, that the objectives of the research aim to explore the actualities of experience, rather than those of a generalising science, and; third, the analysis of the microsocial becomes the means of accessing extended or macro - relations organising society. As already referred to, Oakley’s (1981) discussion was particularly valuable in articulating the "insider" quality as central to feminist research methodology. The insider's perspective comes from the shared experience of being a woman, that is, "a feminist interviewing women is by definition both 'inside' and participating in that which she is observing" (p. 57). As a single Anglo-Celtic midlife woman I shared, gender, culture and economic status with the majority of women interviewed. At the same time I was conscious of the need to create distance in order to ensure that my own experience was not directing the process. One of the ways this was achieved was by asking a set of questions identified by Christman (cited in Hall & Stevens, 1991). They are: How is this woman like me? How is she not like me? How are these similarities and differences being played out in out interaction? How is that interaction affecting the course of the research? How is it illuminating or obscuring the research problem?

 

Denzin and Lincoln (1994, p.11) see what they call the "Fifth Moment" of qualitative research, with an orientation towards more action and activist oriented research, more social criticism and more social critique, shaped by critical theories, feminism and postmodernism as having arrived. Qualitative researchers can no longer purport to directly capture the lived experience of individuals, but need to see their representation in terms of a social text or "tales of the field" (Van Maanen, 1988). Denzin (1992) considers that a more self consciously "interpretive" approach to analysis can be achieved by adopting insights from poststructural philosophy within the postmodern. He sees this as ensuring that the study of meaning making in social interaction will be connected, not just to discourses in circulation, but also to the communication industry and the way in which the "interacting individuals connect their lived experiences to the cultural representations of those experiences" (p. 96). My study therefore attempted a critical interpretivism by drawing on insights from feminism, and poststructural currents within postmodernism. The theoretical framework incorporated Giddens’ notion of the “reflexive self” emphasising agency in decision making, as well as drawing on feminist poststructuralist interpretations of the discursive and embodied construction of self.

 

In the first section of  the analysis I was concerned to explore the impact of menopause on the lives of the women, while also paying attention to the discourses influencing constructions of menopause and midlife health. In the second section I was particularly interested in the way in which bodily experience and sociocultural discourse interacted in the construction of self. In the final analysis section, the microsocial was seen as a way of both reflecting the way in which theories of self are used by participants in constructing identity in midlife, as well as interpreting from their accounts, the macro-relations organising society.

 

Conclusion

I have demonstrated the way in which developing a methodology for a study of single midlife women was an ongoing process. A qualitative method from a feminist perspective was seen as appropriate from the beginning, and the emphasis on a form of discourse analysis focusing on construction of subjectivity was fitting in light of what emerged in the data. Using grounded theory techniques helped make sense of the data initially, and reading about inductive interpretive approaches, as well as approaches to discourse analysis, helped with later analysis. The degree to which the nature of the problem was fluid is apparent in the way in which data analysis revealed, not just the expected processes of managing health, but also the way in which the women experienced their bodies and reshaped their identities in midlife. Developing a theoretical framework was therefore contingent on data analysis.

 

From initially taking for granted my philosophical and ethical stance, grappling with ethical and methodological dilemmas led me down a number of paths that aided me in identifying what it means to be a feminist researcher in the postmodern era

 

References

Acker, J. Barry, K. & Esseveld, J. (1983) Objectivity and truth: Problems in doing

     feminist research. Women's Studies International Forum, 6(4), 423-435.

Australian Bureau of Statistics (1993) No 411300, women in Australia, Canberra, ACT

Bordo, S. (1993). Unbearable weight: feminism, western culture and the body. University of

California Press, Berkely:

Davis, M.E. (1980) A doctor discusses menopause and estrogens. Budlong Press, Chicago.

Denzin, N.K. (1992) Symbolic interactionism and cultural studies. Blackwell, Cambridge.

Denzin, N.K. & Lincoln, Y.S. (Eds.), (1994). Handbook of qualitative research.

     Sage, Thousand Oaks.

Duffy, M. (1985) A critique of research: A feminist perspective. Health Care for Women

International, 6, 341-352.

Ellis, C. (1991) Sociological introspection and emotional experience. Symbolic Interaction,

14(1), 23-55.

Foucault, M. (1980) Power/Knowledge. Selected interviews and other writings 1972-1977.

Pantheon Books, London.

Giddens, A. (1991) Modernity and self identity: Self and society in the late modern age. Polity

Press, Cambridge.

Glaser, B. & Strauss, A. (1967) The discovery of grounded theory: Strategies for qualitative

 research. Aldine, Chicago.

Hall, J.M. & Stevens, P.E. (1991) Rigor in feminist research. Advances in Nursing Science,

13(3), 16-29.

Harding, S. (1986) The science question in feminism. Cornell University Press, Ithaca

Harding, S. (1987) Introduction: Is there a feminist method? In S. Harding (Ed.). Feminism and

 methodology. Indiana University Press, Bloomington,  pp. 1-14.

Hollway, W. (1984) Gender difference and the production of subjectivity. In Henriques, J et al.

 (Eds). Changing the subject: psychology, social regulation and subjectivity. Methuen,

 London.

Huberman, A.M. & Miles, M.B. (1994) Data management and analysis methods. In Denzin,

N.K. & Lincoln, Y.S. (Eds.). Handbook of Qualitative Research. Thousand Oaks, Sage,

 pp. 413-427.

Klein, R.D. (1983) How do we do what we want to do: Thoughts about feminist methodology. In

G. Bowles and R.D. Klein (Eds.). Theories of women’s studies. Routledge and Kegan

Paul, Boston, pp. 88-104.

Kleinmann, S. & Copp, M. (1993) Emotions and fieldwork. Sage, Newbury Park.

Lengermann, P.M. & Niebrugge- Brantley J. (1988) Contemporary feminist theory. In Ritzer, G.

      (Ed.),  Contemporary sociological theory. 2nd Ed. Knopf, New York, pp. 447-496.

Lincoln, Y.S. & Guba, E.G. (1985) Naturalistic Inquiry. Sage, Beverley Hills.

McKinley, S.M., Biafano, N.L. & McKinlay, J.B. (1985) Smoking and age at menopause in

 women. Annals of Internal Medicine, 103, 350-356.

Miles, M.B. & Huberman, A.M. (1994) Qualitative data analyis: An expanded source book, 2nd

Ed. Sage, Newbury Park.

Oakley, A. (1981) Interviewing women: a contradiction in terms. In Roberts, H. (Ed.) Doing

 feminist research. Routledge & Kegan Paul, London, pp. 30-61.

Pence, E. & Shephard, M. (1988) Integrating feminist theory and practice: the challenge of the

 battered women’s movement. In K. Yllo & M. Bograd (Eds.). Feminist perspectives on

 wife abuse. Sage, Newbury Park, pp. 282-298.

Potter, J. & Wetherall, M. (1987) Discourse and social psychology: beyond attitudes and

behaviour. Sage, London.

Reinhartz, (1992) Feminist methods in social research. Oxford University Press, New York.

Richards, T.J.& Richards, L. (1991) The NUDIST system. Qualitative Sociology, 14, 289-306.

Richards, T.J.& Richards, L. (1994) Using computers in qualitative research. In N.K. Denzin &

 Y.S. Lincoln (Eds.), Handbook of qualitative research. Sage, Thousand Oaks, pp. 445-

462.

Richardson, L. (1994). Writing : A method of inquiry.  In N.K Denzin and Y.S. Lincoln

     (Eds),  Handbook of qualitative research. Sage, Thousand Oaks, pp 516-529.

Seibold, C., Richards, L. & Simon, D. (1994) Feminist method and qualitative research about midlife. Journal of Advanced Nursing,19, 94-402. 

Smith, D.E. (1987) Women’s perspective on a radical critique of sociology. In S. Harding (Ed.)

, Feminism and methodology. Indiana University Press, Blooington, pp. 84-96.

Smith, D.E. (1991) Texts, facts and femininity: exploring the relations of ruling. Routledge,

 London.

Stanley, L & Wise, S. (1983) Breaking out: feminist consciousness and feminist research.

Routledge & Kegan Paul, London.

Stanley, L. (1990) Feminist praxis: Research, theory and epistemology in feminist sociology.

 Routledge, London.

Strauss, A.L. (1987) Qualitative analysis for social scientists. Cambridge University Press,

 Cambridge, MA.

Strauss, A.L. & Corbin, J. (1990) Basics of qualitative research, Grounded theory procedures and

techniques. Sage, Newbury Park.

Turner, B.S. (1992) Regulating bodies: Essays in medical sociology. London: Routledge.

Van Keep, P. A., Greenblatt, R. & Ableaux-Fernet, M. Eds.) (1976). Proceeding from the

 International Congress on Menopause. MTP Press, Lancaster, England.

Van Maanen, J. (1988) Tales of the field: On writing ethnography. University of Chicago Press,

 Chicago.

Walsh, R.J. (1978). The age of menopause of Australian women. Medical Journal of Australia.

 2(26), 181-182.

Weedon, C. (1987). Feminist practice and poststructural theory. Blackwell, New York.

World health Organization. (1980) Research on menopause (Tech. Rep. Series 670). Geneva, Switzerland.