The Quantitative/Qualitative Debate and Feminist Research: A Subjective View of Objectivity

Abstract: Research methods are "technique(s) for ... gathering data" (HARDING 1986) and are generally dichotomised into being either quantitative or qualitative. It has been argued that methodology has been gendered (OAKLEY 1997; 1998), with quantitative methods traditionally being associated with words such as positivism, scientific, objectivity, statistics and masculinity. In contrast, qualitative methods have generally been associated with interpretivism, non-scientific, subjectivity and femininity. These associations have led some feminist researchers to criticise (REINHARZ 1979; GRAHAM 1983; PUGH 1990) or even reject (GRAHAM & RAWLINGS 1980) the quantitative approach, arguing that it is in direct conflict with the aims of feminist research (GRAHAM 1983; MIES 1983). It has been argued that qualitative methods are more appropriate for feminist research by allowing subjective knowledge (DEPNER 1981; DUELLI KLEIN 1983), and a more equal relationship between the researcher and the researched (OAKLEY 1974; JAYARATNE 1983; STANLEY & WISE 1990).

This paper considers the quantitative/qualitative divide and the epistemological reasoning behind the debate before investigating two research methods, the statistical survey and the semi-structured interview, in respect of their use to feminist researchers. It concludes by arguing that different feminist issues need different research methods, and that as long as they are applied from a feminist perspective there is no need for the dichotomous "us against them", "quantitative against qualitative" debates.

This paper is designed primarily to summarise the quantitative/qualitative debate from a feminist perspective. However I must first note that there is no one feminist perspective, and hence no one feminist methodology. As Caroline RAMAZANOGLU (1992) highlights, "What one means by feminist methodology depends in part on which authors one takes as examples" (p.208) and hence it is important while reading this paper to recognise the divisions within and between feminisms, as well as the divisions between feminist and non-feminist researchers. [1]

In order to fully understand the debate between quantitative and qualitative research methods and its relevance to feminist research it is necessary to consider the underpinning epistemological issues. Feminist researchers have criticised quantitative positivistic methods for ignoring and excluding women (e.g. OAKLEY 1974) and "adding" women to male knowledge, whereby the findings from research on men are generalised to women (STANLEY & WISE 1993), or "malestream methods" are used to research the experiences of women (MIES 1983). Jessie BERNARD (1975) questioned why research is conducted in certain fields of study but not others and how objectives, methodological and ideological stances are determined, and concluded that they all mirror maleness. This was also highlighted by Dorothy SMITH (1974, p.7) who argued "sociology ... has been based on and built up within the male social universe". [2]

Second wave feminism developed in the 1960's and questioned not only how knowledge is produced, but also who produces it and how it is used. Barbara DUBOIS (1983) highlights that what has been named "universal" knowledge is actually male knowledge, derived from male scholarship and therefore fundamentally flawed. She emphasises the androcentric basis of the social sciences and explains that

the "person" has been considered to be male, and the female, the woman, has been defined in terms, not of what she is, but of what she is not ... The androcentric perspective in social science has rendered women not only unknown, but virtually unknowable. (p.107, italics in original) [3]

Due to this androcentricity and the muting of women's voices within the social sciences, Shulamit REINHARZ (1992) suggests that rather than concentrating on the "sociology of knowledge" we should actually be investigating the "sociology of the lack of knowledge". She argues that this perspective "examines how and why knowledge is not produced, is obliterated, or is not incorporated into a canon" (p.248). It is this questioning of knowledge that forms the basis for feminist epistemological issues. [4]

In the sixth century BC the Pythagorean school of thought developed a table of opposites based on the primary contrast between form (good) and formlessness, or matter (bad, inferior). In this relationship male (form) is set up as dichotomous to female (formlessness). These associations formed the arguments by Greek philosophers such as PLATO (427-347BC) who suggested that women originated from the souls of men who lacked reason (see LLOYD 1984, p.5). These ideas were still prevalent in the early seventeenth century when Francis BACON (1561-1626) related the concepts of form and matter to knowledge (male), and nature (female). BACON claimed that nature is an object of knowledge, with men being the "knowers" and women the "knowable". An analysis by Genevieve LLOYD (1984) led her to conclude that "the maleness of the Man of Reason ... is no superficial linguistic bias. It lies deep in our philosophical tradition" (p.ix). [5]

Knowledge has traditionally been measured by how objective it is deemed to be, in the belief that if the reliability, objectivity and validity "rules" are followed "the truth" will be discovered. If research does not follow the "rules" it is often criticised and dismissed as methodologically flawed and hence "untrue". An example of this can be found in an introductory research methods textbook for psychology in which the author writes; "a majority of psychologists would agree that research should be scientific, and at the very least that it should be objective, controlled and checkable" (COOLICAN 1994, p.4). This statement is problematic in that it is not only saying objective research is desirable, but also assumes total objectivity is possible. Angela MCROBBIE (1982) argues that "representations are interpretations ... they employ a whole set of selective devices such as highlighting, editing, cutting, transcribing and inflecting" (p.51). This highlights the idea that quantitative data, like qualitative data, is interpreted and often manipulated by the researcher and therefore incorporates subjective acts within a supposedly pure objective analysis. Additionally, the striving for objectivity may result in the downplaying of validity if participants feel uncomfortable with the researcher. [6]

Even if the research methods employed are "hard" quantitative ones, they can never be purely objective. Humans, be they female or male, are not computers, and are unable to process information without some degree of subjective interpretation. This starts with the first stage of research: identifying the topic to be studied invariably involves subjectivity. As the process continues this is highlighted further, indeed, the introduction, or literature review, at the beginning of a report is actually a review of the literature that the researcher has deemed to be relevant. This has lead Caroline RAMAZANOGLU (1992) to argue that "it is more logical to accept our subjectivity, our emotions and our socially grounded positions than to assume some of us can rise above them" (p.211). Feminists have broadly rejected the idea of methods premised on the idea of "objectivity" being used to measure social knowledge, and have described such approaches as "an excuse for a power relationship" (STANLEY & WISE 1993, p.167). This rejection of pure objectivity is not limited to feminist researchers, and many other sociologists have questioned and rejected the notion, preferring to make knowledge claims based on findings being corroborated by other research. [7]

Feminist standpoint theorists such as Nancy HARTSTOCK and Dorothy SMITH argue within a Marxist framework that women can actually produce better knowledge rather than men due to their sex-class position. Research by feminist standpoint theorists is held to produce more complete, less distorted knowledge (HARDING 1986), and is based on MARX's concept of the "proletarian standpoint". Nancy HARTSTOCK (1983) argues that those in domination can only ever have or produce partial knowledge. Feminist standpoint theory has been rejected by Donna HARAWAY (1991), who argues that neither women nor men can ever have total knowledge, as all knowledge is partial. Post-modern feminists have made similar criticisms, claiming there is no one "truth" and although all standpoints are conflicting, none are privileged. These theories expand upon the work of earlier sociologists such as C.W. MILLS (1959), who argued that social laws are always historically specific. However feminists have added gender to the "hat" and shaken it up again. [8]

The problem with rejecting the notion of objectivity is that there remains a need for a measure by which to judge knowledge. HARAWAY (1991) suggests that the notion of complete objectivity should be redefined and replaced by situated knowledge, in which the researcher recognises that knowledge can never be regarded as universal. She writes "situated knowledges require that the object of knowledge be pictured as an actor and agent" (p.198) and many feminists have started to include an "intellectual biography" (STANLEY & WISE 1983; 1990) in their work, hence acknowledging both the situation the knowledge was produced in, and the located knowledge of the researcher. [9]

Others have argued that the measure by which to judge research should be the effect it has on improving women's lives (REINHARZ 1983) and the role it has in aiding the emancipation of women. Maria MIES (1983) suggests that "the "truth" of a theory is not dependent on the application of certain methodologies and rules but on its potential to orient the processes of praxis towards progressive emancipation and humanisation" (p.124). Experience has also been suggested as a measure of knowledge, and an important contribution is made by DUBOIS (1983) in arguing that a rejection of absolute standards based on notions of objectivity in favour of relativist standards based on subjective experience in no way makes the research less critical, rigorous or accurate. The use of experience as an index of the adequacy of research is, however, still contested by some researchers (most strongly by HAMMERSLEY 1992). [10]

In order to demonstrate how these theoretical aspects of feminist methodology(s) are operationalised in empirical social research I will now discuss two methods, the semi/unstructured interview and the statistical survey, in view of their uses and their limitations in contributing to feminist knowledge. [11]

Surveys are generally used to obtain responses from a sample that can be coded with variable labels and statistically analysed, with the results being generalised to a wider population. Due to the nature of the questions asked and the process of analysis (for example, frequency counts, calculation of the mean, between-group comparisons, in short, the kinds of operations associated with use of a statistical computer package, such as SPSS), the survey is generally defined as a quantitative method, and is utilised to examine widespread social issues whereby the results of a sample can be generalised upon to reflect society as a whole. The use of surveys to collect statistics has been criticised by many feminist researchers. Criticisms often focus on the crudeness of survey questions and data, which are arguably too simplistic to examine the complexity of the social issues being addressed. Denise FARRAN (1990) argues that rather than statistics being a representation of social reality, they are actually a construction of reality. She argues that statistics are "divorced from the context of their construction and thus lose the meanings they had for the people involved" (p.101). This has also been highlighted by Anne PUGH (1990), who argues that statistics need "chaperoning" (p.109), as they are often used out of context and generalised upon. [12]

Hilary GRAHAM (1983) criticises many aspects of the survey method, arguing that it "reflects the ideology of the nineteenth-century world in which it was developed" (p.132). She concludes that "the survey may well frustrate, from its inception, a feminist programme" (p.133). She argues that the survey method treats all individuals as being equal units and therefore does not reflect the patriarchal society in which the data are gathered. GRAHAM names her article "Do her answers fit his questions?", highlighting the subjectivity involved in composing questions for a survey. This is emphasised by Liz KELLY et al., who argue that "asking 'have you ever been raped?' will produce different responses from asking 'have you ever been forced to have sex?'" (1992, p.109), indicating that women do not always label forced sex as rape. Similarly, in a study investigating violence against female and male taxi drivers (WESTMARLAND & ANDERSON 2001) I wished to discover the prevalence of sexual abuse, but used the term "sexual harassment" rather than "abuse" in the questionnaire distributed. Although I believe that sexual harassment is a form of sexual abuse, the term "abuse" in contemporary society is generally used to describe abuse against children. I therefore feel a higher number of participants gave a response to the question "have you ever experienced sexual harassment at work?" than would have if I had asked "have you ever experienced sexual abuse at work?", despite my view that it is the same experience. Additionally, one male participant refused to answer any further questions when he came to this question, claiming "this questionnaire is designed for women". Interestingly the research showed that male taxi drivers are sexually harassed at work, albeit not to the same extent as women. [13]

The questions not asked can influence the research findings as much as the questions asked, which was also highlighted in my taxi driver research. The question about sexual abuse/harassment was a question not asked in previous research about violence towards taxi drivers. If a phenomenon is assumed not to affect a population there will generally be no relevant question included, hence suppressing and nullifying the experiences of the population studied. The "naming" of women's issues by feminist researchers has made (and is still making) an important contribution to the women's liberation movement as a whole. The significance of naming is described by Barbara DUBOIS (1983), who argues that

naming defines the quality and value to that which is named—and it also denies reality and value to that which is named, never uttered. That which has no name, that for which we have no words or concepts, is rendered mute and invisible: powerless to inform or transform our consciousness of our experience, our understanding, our vision; powerless to claim its own existence (p.108). [14]

She further argues that the social sciences have maintained and perpetuated the muteness of women's voices. [15]

In a further study of taxi-drivers, I found that the female drivers rarely named their experiences as "violent" in the survey, but follow-up unstructured interviews revealed that the women frequently invalidated and normalised their experiences of violence (WESTMARLAND 2001). This emphasises differing definitions of violence, with the women refusing to define an act as violent unless it included a physical attack, hence using a very different criterion to define violence than the one I was applying. It was acknowledged in advance that this was a potential flaw in the research design, however, possible solutions were thought not to be suitable. For example, one solution might be to define the phenomenon before asking each question. However, this would also raise questions of researcher objectivity and subjectivity. Whose definition would I be using? If I used my definition of violence am I implying that my definition of violence is more accurate (more "true"?) than the taxi drivers' own definitions? Would I then be labelling their experiences for them? A further alternative would have been to extend the questionnaire to ask about specific acts. For example instead of asking "have you ever experienced physical violence at work?" it would be possible to break this question into interrelated questions, such as "have you ever been hit at work?" and "have you ever been hit with an object while at work?" This, however, raises further issues: what exactly is the definition of the word "hit"? Does a "slap" count as a "hit"? What about a "head-butt"? Additionally, if a taxi driver in this area of the UK talked about being "hit" they would not be referring to a physical attack, but to a car crash in which they had been involved. To refer to a physical attack, phrases such as "got scragged", "took a beating", or "got done in" would be used by taxi drivers in the town in which the research was conducted. A further problem with this approach is that to extensively use definitions and add further interrelated questions would make the questionnaire substantially longer than the three pages of my survey. Length is an issue in obtaining a sound response rate. Past researchers have reported problems with response rates to their questionnaires by taxi drivers, for example a response rate of 7.4% was reported by Ian RADBONE (1997) whose survey contained 68 questions and was 14 pages long. This poor response rate is likely to have been related to the length of the questionnaire. It was therefore decided early in the research process that the definition of the various acts of violence should be determined by the respondent. Therefore, if the taxi driver had perceived an act as being physically abusive then it was recorded as such. This was the same for acts of verbal abuse and sexual harassment/abuse, hence the researchers own definitions were abandoned in favour of those of the participants. Although this is not ideal it was felt to be the best strategy, and it should be noted that, in reporting the research, no claims of objectivity, reliability or validity were made. [16]

Not all feminists have argued against the use of quantitative methods within feminist research, for example Toby JAYARATNE (1983) warns feminists against a total rejection of quantitative methods, and O'LEARY (1977) argues that to link feminist research with qualitative methods simply reinforces traditional dichotomies that may not be in the best interests of feminist research. These arguments in favour of quantitative methods are strengthened by the many examples of their effective use. For example, Betty FRIEDAN (1963) used survey data to develop her analysis of the "problem that has no name"; a book which greatly influenced second wave feminism, and Shulamit REINHARZ (1992) highlights that survey based data can be useful in looking at the prevalence and distribution of particular social problems. An example of this use of survey data can be found in the work of ANDERSON, BROWN and CAMPBELL (1993) who investigated sexual harassment within the police force. They found that nine out of ten women police officers had experienced sexual harassment at work, and that one in ten had considered resigning from the police force due to this harassment. Figures such as these would be extremely time-consuming, expensive and difficult to obtain on a national level using qualitative methods such as interviews. [17]

Helen ROBERTS (1981) suggests that the reason that relatively few women are involved in quantitative research may be explained by the "inadequacy of certain statistical procedures in looking at sex differences" (p.23) whereby crude and simplistic data-labelling does not reflect the complexity of women's experiences. Similarly Lorraine GELSTHORPE (1990) suggests that "the problem is perhaps not quantification itself but insensitive quantification" (p.91). Needless to say, if quantification is crudely done it is invalid. GELSTHORPE and MORRIS (1990), writing about domestic violence, argue that although positivist quantitative methods are generally abandoned by feminists in favour of qualitative approaches, the value of quantitative methods in this research field depends greatly on the questions being asked. They argue that quantitative methods can prove useful in producing background data, and concluded that "feminist researchers use a multiplicity of methods to explore wife-battering" (p.86). Survey data has also added to feminists' understanding of rape and has been used to identify attitudes about rape (e.g. READ & MILLER 1993), and to examine the prevalence of unreported rape against women by their husbands (e.g. PAINTER & FARRINGTON 1998). [18]

It is also important for researchers to "speak the same language" as those to whom the research will be presented. Audrey HUNT (1986) highlights the need for feminist research to produce statistics in order to formulate legislation, and Mary MAYNARD (1994) emphasises the role that quantitative methods have played in identifying the feminisation of poverty, arguing "the political potential of such work should not be underestimated" (p.13, italics in original). Similarly, Shulamit REINHARZ (1992) points to the use of survey data in the formulation of laws and policy making, highlighting that "statistical information about sexual harassment ... contributed to its reification in ways that encouraged the establishment of sexual harassment committees in universities and ... eventually provided legal redress for individuals." (p.80) [19]

In attempting to make a government take a feminist issue seriously, it is essential to present research in the (masculine?) language such a research audience expects. Governments are less concerned with the concerns of individuals per se but rather with the wider picture and, it may be argued, they are more likely to take issues seriously if they are presented in this way, and in their language. [20]

In contrast to the quantitative paradigm, qualitative researchers are generally more concerned with validity, rather than objectivity and reliability, and put less emphasis on finding "the truth". Semi and unstructured interviews are methods widely used in feminist research as they are claimed to "convey a deeper feeling for or more emotional closeness to the persons studied" (JAYARATNE 1983, p.145). Feminist researchers, greatly influenced by the work of Ann OAKLEY, make every effort to conduct interviews in a way that does not further oppress the participant. They attempt to actively involve the participant in the research process as much as possible. They reject the use of the word "subject" that implies the participant is an insensate object to be experimented on and observed like an animal in a zoo. Although a more equal relationship between the researcher and participant is often cited as increasing the validity of the research, this is not the primary reason feminist researchers insist upon this relationship. Feminist researchers are working within the wider women's liberation movement and are working towards the overall aim of all women being free from oppression. It is hence clearly not acceptable for researchers to further oppress women in the name of academic research. [21]

Historically malestream [sic] textbooks have documented the way an interview should be conducted, for example recommending distance between the interviewer and interviewee, not revealing the feelings or standpoint of the researcher, and not sharing knowledge. These guidelines were questioned by interactionist sociologists such as Howard BECKER (1971), who suggested that interviews should be more conversational in nature, and feminists such as Ann OAKLEY have argued that this is particularly salient when interviewing women. She argues that traditional guidelines contradict the aims of feminist research (OAKLEY 1981) and that for a feminist interviewing women, the "use of prescribed interviewing practice is morally indefensible (and) general and irreconcilable contradictions at the heart of the textbook are exposed" (p.41; see OAKLEY 1981 for examples of textbooks). [22]

Interview techniques have been adapted by interpretivist sociologists and feminist researchers to be more participant-friendly and these guidelines have been integrated into many mainstream contemporary textbooks (for example BURGESS 1984). [23]

Traditional research methods textbooks also advise against conducting research in which you are emotionally involved in some way, in the guise that this will minimise the supposed objectivity of the study. It has conversely been argued that a close and equal relationship to the researched can actually lead to an acquisition of more fruitful and significant data (FINCH 1984; OAKLEY 1981). Clara GREED (1990) writes of similar issues when she discusses her experience as a feminist surveyor while researching the position of women in surveying. She sees research as a two-way interaction, and writes

So I am studying a world of which I myself am part, with all the emotional involvement and accusations of subjectivity that this creates. I do not attempt to keep my surveyors at arm's length and do research "on" them as my subjects whilst maintaining a dominant position, as is common in much traditional "objective" research (p.145). [24]

This was the perspective that I adopted when interviewing female taxi drivers utilising my personal experience as a night shift taxi driver (WESTMARLAND 2001). I found that rather than hindering the research process this downplayed the researcher's new academic status, resulting in a more relaxed environment for both the interviewer and interviewee. The interviewees were also encouraged to invent their own pseudonyms in an attempt to further balance power relations. [25]

The interview can therefore be complementary, rather than oppositional, to survey research. Rather than the "us against them" relationship, interviews can give a deeper, more complex knowledge of the issues named by survey research. For example if we are faced with a chocolate, we can see from the outside that it is a chocolate, however we must delve deeper to discover whether it is hard or soft, has a hazelnut or an orange centre, and so on. It is this inner knowledge that is gained by interviews. They also allow us to validate to some degree what we have found in related quantitative research. The chocolate that looked like a hazelnut from the outside may turn out to have a soft orange centre when more closely examined. As described in the previous section where survey research was used to inform governments of the prevalence of women's issues, interviews allow us to delve deeper and more fully explain these issues. It is therefore not enough to simply know, for example, that women are more likely to be raped by acquaintances (45% of rapes; HARRIS & GRACE 1999) or intimates (43% of rapes) than by strangers (12% of rapes), and that only 6% of rape cases reported to the police result in a conviction (HARRIS & GRACE 1999), we need to know how this affects the lives of women. Feminism is primarily a movement for social change and only by delving deeper than the surface can we find out not only what needs to be changed, but also how it can be changed. [26]

This paper has described not feminist research methods, but rather research methods adapted for feminist use. What has traditionally been seen as a strength of quantitative research, namely objectivity, has been shown to reflect the subjective knowledge of the researcher and hence reveals the false dichotomization of objectivity and subjectivity, and of quantitative and qualitative methods. Without this unnecessary opposition the usefulness of mixed method research can be realised and feminist perspectives on research can be acknowledged simply as "good" research. I have demonstrated the usefulness of quantitative methods in the naming of women's oppression and also the usefulness of qualitative methods for delving further and using feminist research for change within the women's liberation movement. Although a survey may be the best way to discover the prevalence of problems, interviews are needed to fully understand women's experiences and theorise these experiences with a view towards social change. For example, a survey can tell us that women working outside the home generally get paid less than men, but does not explain how this makes women feel and how it affects their lives as a whole. [27]

To conclude, and in acknowledgement of discussion about other research methods, I support the stance taken by KELLY, REGAN and BURTON (1992) when they argue "what makes feminist research feminist is less the method used, and more how it is used and what it is used for" (p.150). Different feminist issues need different research methods, and as long as they are applied from a feminist perspective there is no need for the dichotomous "us against them", "quantitative against qualitative" debates. Neither method is "hard" nor "soft"—they are methods, and their success depends solely upon the researcher employing them. Feminists need broad-based knowledge as much as they need individual women's experiences. [28]

My thanks go to Dianne RAILTON (University of Teesside) for comments on an earlier draft of this paper, to the editors; Nigel FIELDING and Margrit SCHREIER for their constructive criticism and to Jan ANDERSON (University of Teesside) for inspiration, enthusiasm and friendship.

Anderson, Rhona; Brown, Jennifer & Campbell, Elizabeth (1993). Sexual harassment within the Police Force. London: H.M.S.O.

Burgess, Robert (1984). In the field: An introduction to field research. London: Routledge.

Coolican, Hugh (1994). Research methods and statistics in psychology (2nd ed.). London: Hodder & Stoughton.

Depner, Charlene (1981). Towards the further development of feminist psychology. Paper presented at the mid-winter conference of the Association for Women in Psychology, Boston.

Duelli Klein, Renate (1983). How to do what we want to do: thoughts about feminist methodology. In Gloria Bowles & Renate Duelli Klein (Eds.), Theories of women's studies (pp.88-105). London: Routledge and Kegan Paul.

DuBois, Barbara (1983). Passionate scholarship: notes on values, knowing and method in feminist social sciences. In Gloria Bowles & Renate Duelli Klein (Eds.), Theories of women's studies (pp.105-117). London: Routledge and Kegan Paul.

Farran, Denise (1990). Seeking Susan. Producing statistical information on young peoples leisure. In Liz Stanley (Ed.), Feminist praxis (pp.91-103). London: Routledge.

Finch, Janet (1984). It s great to have someone to talk to: The ethics and politics of interviewing women In Colin Bell & Helen Roberts (Eds.), Social researching: politics, problems, practice (pp.70-88). London: Routledge & Kegan Paul.

Gelsthorpe, Lorraine (1990). Feminist methodologies in criminology: a new approach or old wine in new bottles? In Lorraine Gelsthorpe & Alison Morris (Eds.), Feminist perspectives in criminology (pp.89-106). Buckingham: Open University Press.

Gelsthorpe, Lorraine & Morris, Alison (1990). Feminist perspectives in criminology. Buckingham: Open University Press.

Graham, Dee & Rawlings, Edna (1980). Feminist research methodology: comparisons, guidelines and ethics. Paper presented at the annual meeting of the American Psychological Association.

Graham, Hilary (1983). Do her answers fit his questions? Women and the survey method. In Eva Gamarnikow; David Morgan; Jane Purvis & Daphne Taylorson (Eds.), The Public and the private (pp.132-147). London: Heinemann.

Greed, Clara (1990). The professional and the personal. A study of women quantity surveyors. In Liz Stanley (Ed.), Feminist praxis (pp.145-155). London: Routledge.

Haraway, Donna (1991). Simians, cyborgs and women: the reinvention of nature. London: Free Association Books.

Harding, Sandra (1986). Feminism and methodology. Bloomington: Indiana University Press.

Harris, Jessica & Grace, Sharon (1999). A question of evidence? Investigating and prosecuting rape in the 1990s, Home office research study 196. London: Home Office.

Hartstock, Nancy (1983). The feminist standpoint: Developing the ground for a specifically feminist historical materialism. In Sandra Harding & Merrill Hintikka (Eds.), Discovering reality (pp.283-311). Dordrecht, Holland: Reidel Publishing Company.

Hunt, Audrey (1986). Use of quantitative methods in researching issues which affect women. Methodological Issues in Gender Research, 10, 12-19.

Jayaratne, Toby (1983). The value of quantitative methodology for feminist research. In Gloria Bowles & Renate Duelli Klein (Eds.), Theories of women's studies (pp.140-162). London: Routledge and Kegan Paul.

Kelly, Liz; Regan, Linda & Burton, Sheila (1992). Defending the Indefensible? Quantitative methods and feminist research. In Hilary Hinds, Ann Phoenix & Jackie Stacey (Eds.), Working out: New directions in women's studies (pp.149-161). Lewes: The Falmer Press.

Lloyd, Genevieve (1984). The man of reason: "male" and "female" in Western Philosophy. London: Methuen & Co. Ltd.

Maynard, Mary (1994). Methods, practice and epistemology: the debate about feminism and research. In Mary Maynard & Jane Purvis (Eds.), Researching women's lives from a feminist perspective (pp.10-27). London: Taylor and Francis.

McRobbie, Angela (1982). The politics of feminist research: between talk, text and action. Feminist Review, 12, 46-57.

Mies, Maria (1983). Towards a methodology for feminist research. In Gloria Bowles & Renate Duelli Klein (Eds.), Theories of women's studies (pp.117-140). London: Routledge and Kegan Paul.

Mills, C. Wright (1959). The sociological imagination. New York: Oxford University Press.

Oakley, Ann (1981). Interviewing women: a contradiction in terms. In Helen Roberts (Ed.), Doing feminist research (pp.30-62). London: Routledge & Kegan Paul.

Oakley, Ann (1997). The gendering of methodology: An experiment in knowing. Seminar to Swedish Collegium for Advanced Study in the Social Sciences, Uppsala, Sweden, 10th April.

Oakley, Ann (1998). Science, gender, and women's liberation: An argument against postmodernism. Women's Studies International Forum, 21(2), 133-146.

Painter, Kate & Farrington, David (1998). Marital violence and its relationship to marital and non-marital rape. International Review of Victimology, 5(3/4), 257-276.

Pugh, Ann (1990). My statistics and feminism. In Liz Stanley (Ed.), Feminist praxis (pp.103-113). London: Routledge.

Radbone, Ian (1997). Taxi safety initiatives study. Report presented to the Passenger Transport Board, South Australia.

Ramazanoglu, Caroline (1992). On feminist methodology: male reason versus female empowerment. Sociology, 26(2), 207-212.

Read, Stephen & Miller, Lynn (1993). Rapist or regular guy: Explanatory coherence in the construction of mental models of others. Personality and Social Psychology Bulletin, 19(5), 526-540.

Reinharz, Shulamit (1979). On becoming a social scientist. San Francisco: Jossey-Bass.

Reinharz, Shulamit (1983). Experiential analysis: a contribution to feminist research. In Gloria Bowles & Renate Duelli (Eds.), Klein Theories of Women's Studies (pp. 162-192). London: Routledge and Kegan Paul.

Reinharz, Shulamit (1992). Feminist methods in social research. Oxford: Oxford University Press.

Smith, Dorothy (1974). Women's perspective as a radical critique of sociology. Sociological Inquiry, 44, 7-13.

Stanley, Liz & Wise, Sue (1983). Back into the personal or: our attempt to construct feminist research. In Gloria Bowles & Renate Duelli Klein (Eds.), Theories of women's studies (pp.20-60). London: Routledge & Kegan Paul.

Stanley, Liz & Wise, Sue (1990). Method, methodology and epistemology in feminist research processes. In Liz Stanley (Ed.), Feminist praxis (pp.20-60). London: Routledge.

Westmarland, Nicole (2000). Doesn't your lad mind you doing this luv? Occupational violence and taxi drivers: A suitable job for a woman? Unpublished BSc dissertation, University of Teesside.

Westmarland, Nicole & Anderson, Jan (2001, in press). Safe at the wheel? Security issues for female taxi drivers: Visions of a feminine future. The Security Journal, 14(2).

Nicole WESTMARLAND has carried out research on violence against taxi drivers and women on probation. She writes from the standpoint of a white, working class feminist.

Westmarland, Nicole (2001). The Quantitative/Qualitative Debate and Feminist Research: A Subjective View of Objectivity [28 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 2(1), Art. 13, http://nbn-resolving.de/urn:nbn:de:0114-fqs0101135.

Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (FQS)

ISSN 1438-5627

Creative Commons Attribution 4.0 International License