Chapter Three: The Tease – Feminist Research Methodologies
Knowing and Doing: Epistemology, Methodology, and Methods
According to Sprague (2005), discussions of research methods are often overly-technical and boring because they tend to focus on “trivial” procedural details instead of addressing the epistemological assumptions that underlie a researcher’s choice of method (Sprague, 2005, pp. 4-5). Since critical and feminist researchers acknowledge that their social power and biases cause them to become aware of some aspects of the truth but to overlook others, the process of uncovering one’s epistemological assumptions through reflection is a crucial aspect of the research process (Sprague, 2005, p. 3; Baxter & Babbie, 2004, pp. 62-3). For critical researchers, reflecting on the research process in general, as well as on one’s choice of methods, is at least as important as developing an outline of the details of the methodological procedure itself. Since both “critical theory” and “feminism” are fairly broad categories of thought, I will specify how I understand these terms in the following section. I will also relate the epistemological assumptions embedded in my use of these theories to my choices of methods, using Harding’s (2005) distinctions between “epistemology,” methodology” and “methods” as a guide (as cited by Sprague, 2005, pp. 4-6).
According to Harding (2005), one’s “epistemology” is a theory about knowledge (as cited by Sprague, 2005, p. 5). It includes ideas about who can know, and under what circumstances knowledge can be developed (as cited by Sprague, 2005, p. 5). As stated earlier, my thought and analysis fit broadly within the critical/feminist paradigms; I also share assumptions with interpretivist and social constructionist scholars, and I qualify my feminist views with sex radicalism (hesitantly) (Baxter and Babbie, 2004, pp. 58-62; Queen, 2001, p. 94). These perspectives all question the positivist assumption that objective knowledge of the empirical world is possible and valuble, and they share the assumption that subjective understandings have ontological value (Baxter and Babbie, 2004, p. 33, p. 36; 60-63; Queen, 2001, p. 94; Sprague, 2005, p. 57). Although some feminists, interpretivists, and “strong” social constructionists (also known as post-modernists) believe that objective knowledge is impossible, many critical theorists and feminists assume that partial or culturally-mediated knowledge of the objective world is possible (Sprague, 2005, p. 39, p. 57). My willingness to engage in research at all implies that I take the latter view (Sprague, 2005, p. 39, p. 57).
These epistemological assumptions influence one’s choices of research methods (Sprague, 2005, p. 5). Harding (1987) refers to decisions about how to use research methods as one’s “methodology” (as cited by Sprague, 2005, p. 5). Research methods themselves are simply techniques for gathering and analyzing information, and include the familiar distinctions between quantitative and qualitative strategies (Harding 1987, as cited by Sprague, 2005, p. 5). Therefore, scholars who primarily value subjective understandings are likely to choose research methods that capture idiosyncratic, subjective understandings, and typically use qualitative methods to that end (Baxter and Babbie, 2004, pp. 58-9, 62; Sprague, 2005, p. 57). These “subjectivist” researchers may use multiple research methods as well (by combining in-depth interviews with a discourse analysis, for example). However, scholars who value both “subjective” and “objective” truths might combine quantitative and qualitative strategies in order to capture the “objective” factors that they assume influence subjective understandings, or to enhance objective analyses with subjective interpretations (Baxter and Babbie, 2004, p. 65). Alternatively (or in addition to combining methods), critical scholars in particular may attempt to bridge the “knowledge gap” between the researcher and the participant, by using collaborative strategies and engaging research participants who embody various perspectives (Sprague, 2005, pp. 74-9). Interrogating the “philosophical” assumptions that underlie one’s choices of methods can uncover a great deal about how a researcher sees the world, and it can allow us to figure out how their own biases influence their understanding of research “findings.”
Critical Assumptions: Research Methodologies
My decisions in regards to methodology in particular were informed by interpretivist and feminist assumptions. From an interpretive perspective, I assumed that the resulting data would depict a story of informants’ lives that reflected both my own and the informants’ subjectivities; although ultimately I believe that both the interviews and this research report are inevitably weighted towards my own interpretations and biases (taking a critical view) (Baxter & Babbie, 2004, pp. 326-7; Sprague, 2005, pp. 60-1). I understood my methodology to be “feminist” inasmuch as I assumed the existence of power dynamics between myself and informants (Sprague, 2005, p. 57). Given this dynamic, I believed it was important to acknowledge and validate women’s experiences and subjectivities (since women’s views and experiences contribute to social dialogues on femininity, gender, and bodies in general) (Kirsch, 2005, p. 4; Pitts, 2003, p. 47; Miller, 2009, p. 34). However, unlike researchers who make “stereotypical” assumptions about feminist methods, I did not believe it was possible or desirable to account for this power dynamic by transferring authority to research participants entirely (Sprague, 2005, pp. 57-8). At the very best, I felt I could account for this dynamic by remaining reflexive in regards to my own understandings (which is an ever-evolving process), seeking to reduce harm to participants by following the University’s standard ethics review process, and by remaining accountable to the informants by re-constructing their stories in ways that reflect their complexities, instead of enforcing gendered and sexual stereotypes (Miller, 2009, p. 34). In the following sections, I will provide a more detailed accounting of my research and sampling methods, discuss the aforementioned ethical considerations as well as the population I studied, and discuss the reporting process.
Since I believed that qualitative methods would lend themselves to an account of informants’ subjectivities, I decided to engage in semi-structured interviews with six study participants to gather data about the local burlesque community (Baxter and Babbie, 2004, pp. 329 – 330). The interviews generally took two hours, and sometimes strayed from the planned topics of discussion (but usually in productive ways). In the interest of confidentiality, I met with informants in a private location; in the interests of their convenience, they choose the location. We agreed on times that fit into both parties’ schedules.
I followed an interview guide (Appendix 1) when I spoke with participants, which included reminders regarding the procedure, the topic areas I wanted to discuss, and questions related to each topic area. Since I wanted to gain an understanding of the informants’ perspectives rather than anticipate their responses, I attempted to allow conversation to develop naturally throughout the interview. I asked descriptive questions in order to gain a general sense of informants’ self-perceptions, their experiences within the community and as performers, and their motivations for participating in burlesque (for example, “how do you experience femininity in everyday life?”) (Baxter & Babbie, 2004, pp. 334-5). In order to explore issues raised within the literature or during the interview process, I asked verification questions as well (Baxter & Babbie, 2004, pp. 334-5). I found that I could motivate storytelling by simply asking the questions, but many of the topics I had planned to discuss emerged in conversation before I’d had an opportunity to ask specific questions. Informants related past experiences with their families, within professional relationships, and within intimate relationships; we discussed feminist ideologies and politics in general; and informants’ self-understandings, particularly as it related to their sexual identity development and experiences as performers. Although I imposed a degree of structure on the interviews, in my mind they felt like casual conversations. For the most part, informants and I established rapport quickly. Upon completion, informants often stated that they appreciated the opportunity to discuss their involvement in burlesque; this seems to suggest a degree of familiarity and comfort that resonated with my own experiences of the interviews.
Ethical Considerations: General Obligations as a Researcher
In accordance with standard practices within social science research, I respected basic ethical imperatives intended to maintain the health and dignity of research participants (Baxter and Babbie, 2004, pp. 85-6). I ensured voluntary participation by soliciting participants using the most passive recruitment methods possible. To avoid coercion, I posted messages (Appendix 2) in public forums (where more deliberately “persuasive” messages are common), and didn’t explicitly ask any of my contacts to participate in the study. I didn’t offer any rewards or enticements for participation, nor did I deliberately target a group that may be particularly vulnerable to enticement or coercion (Baxter and Babbie, 2004, p. 86). In addition, I offered the a right to withdraw from the study (with limitations as specified in an informed consent document; Appendix 3)(Baxter and Babbie, 2004, p. 86). Before participating, informants were required to read and sign the document to indicate understanding and acceptance of the terms.
Since the sample population consisted of individuals who had somewhat “public” personas, and were solicited from a friend network, I felt it was highly likely that I would personally know the identity of any particular volunteer (Baxter & Babbie, 2004, p. 88). Therefore I could not ensure anonymity, but I could ensure confidentiality (Baxter & Babbie, 2004, p. 88). At the time of my study the risks of exposing participants’ identities seemed particularly salient, since burlesque performers were being subjected to regulations that restricted the extent to which they could use nudity on stage. During the ethics review process, it was uncertain how these regulations might apply to performers in an interview situation; I consulted with a representative of the relevant quasi-judicial body and determined that there were no risks of sanctions in that particular context. Taking a more cautious route anyways, I gave informants the option of using a pseudonym; many felt the risks of revealing their identities was low, and opted to use stage names. For my part, I agreed to destroy any “keys” linking the pseudonyms to participants’ given names at the end of this project.
Population: Sampling Methods
I used purposive, judgmental, and snowball sampling techniques in order to gather my sample (Baxter and Babbie, 2004, p. 135). My methods were ‘purposive’ because the identities of participants were related to the goals of my study (they all self-identified as women, and had participated in at least one burlesque performance in Calgary) (Baxter and Babbie, 2004, p. 135). It was “judgmental” because I used my knowledge of the population to identify individuals who have influence within the community and could help me locate potential study participants (Baxter and Babbie, 2004, p. 135). To this end, I contacted prominent community members and performance organizers, Danika Challand and “Betty Galore.” I asked Ms. Challand if she could forward the recruitment notice to her contacts, and I asked Ms. Galore if I could post the recruitment notice to her Facebook Group (“The Alberta Burlesque Alliance”). Although Ms. Challand did not reply to my request, Ms. Galore consented to the posting. I posted recruitment notices (Appendix 2) to my own and the Calgary Burlesque Alliance’s Facebook page, which facilitated access to both my own and broader social networks. Finally, at least one participant was recruited through a “snowball” procedure (Baxter & Babbie, 2004, p. 313). A mutual friend had noticed the recruitment notice on my Facebook page, informed her about the study, and the informant contacted me (Baxter and Babbie, 2004, p. 135).
Since I used non-probability sampling techniques, my conclusions have low external validity and I cannot generalize my findings to a broader population (Pitts, 2003, p. 19; 2006). This means that I can only make modest claims about this particular group, and I cannot make any broad causal or systemic explanations (Baxter & Babbie, 2004, p. 298). However, such explanations are outside the scope of research that starts from interpretivist assumptions (although such claims are not necessarily outside the scope of some critical/feminist assumptions, they are outside of my own). Instead of using quantitative standards for validity and reliability, many qualitative researchers believe that non-probability samples can be used to generate credible, dependable, confirmable, and transferable findings (Baxter & Babbie, 2004, pp. 297-8). This means that my findings can be considered “trustworthy” inasmuch as they resonate with informants, if a reader can follow my interpretive process, my conclusions are logically connected to the data, and my description of the group is “thick” enough that readers can draw their own conclusions about whether my findings might apply to another context or group (Baxter & Babbie, 2004, p. 298).
However, for the most part, this sample reflects my own (and participants’) observations of the demographic composition of the local community of burlesque performers. Although the community is more ethnically and racially diverse than my sample group, it tends to be dominated by more Caucasian participants (or those that appear, at first glance, to be white). My observations suggest that the class backgrounds of those who become involved in the community are somewhat more diverse, although audience members and performers are generally working or middle class. It is unlikely that those living in extreme poverty are likely to participate, since genres that involve feminine performances of gender require capitalist resources (Miller, 2009, p. 69). However, the community employs strategies that facilitate a degree of access for those with fewer resources, such as “collective” costume-making endeavors. Members also exchange goods and services through a “barter” economy. In terms of sexual identities, heterosexual relationships appear to be normative in the community, but it is generally a gay/queer friendly, pro-sex group (Pitts, 2003, p. 20). Since many of the performances take place in bars, involvement is limited to adults.
Population: Description of Sample
I limited my sample group to women aged 18-35, who had been involved in at least one burlesque show in their lifetimes as a performer or in a supporting role. In general, my sample consisted of white, educated, middle class women, ranging in age from late adolescence to their mid-thirties. Since I did not think that sexual identity was related to involvement in burlesque at the outset of this research, I did not explicitly ask informants’ their sexual identity. However, sexual practices became apparent throughout the course of the interviews when I asked about relationships in general. All of the participants but one discussed heterosexual relationships; another represented herself as a “queer ally.” Therefore, I can say that most of the sample group discussed heterosexual practices at the time of the interview, but their views suggest that they were all gay or queer-friendly, and pro-sex (Pitts, 2003, p. 20). Most of the informants had some degree of post-secondary education; the youngest participant had a high-school education.
Transcribing and Coding
I transcribed the interviews myself, using qualitative data analysis software (Nvivo, Versions 9 and 10). I began to create “working” coding categories as I performed the literature review, as I conducted the interviews and refined the interview procedure, and throughout the transcription process. Through processes of inductive reasoning, I identified patterns in participants’ responses that I compared to the theoretical literature and used to refine the interview design (Baxter and Babbie, 2004, pp. 366-7). Looking back on this process, I can see that my interview design and coding categories emerged together. During the interviews, I would notice that some issues and themes seemed particularly salient to participants (such as body image, for example); I also found that some of the topics I had set out to discuss would emerge “spontaneously” before I had a chance to even ask the associated questions (such as topics relating to the overall theme I later referred to as, “feminist views on sexuality, agency, and the body”). I would note these themes/topics in my research journal as a possible “coding category idea”, and I would also make a point of discussing the particular issue with the subsequent informants. By the time I had finished the interviews, I had produced a number of tentative coding categories through this iterative interpretive process. As I reviewed the transcripts, further coding categories emerged.
Reporting: Entering the Feminist Conversation on Research Reporting
In response to their critiques of conventional academic discourses, feminist researchers have developed a number of reporting strategies that are designed to make their own perspectives visible, avoid objectifying informants, make research accessible to the communities studied, and engage the reader on intellectual and emotional levels (Sprague, 2005, p. 167). The most obvious difference between feminist research reporting and conventional academic discourses is the use of “familiar” language and personal pronouns and use of the active voice; this strategy is designed to make the researcher’s subjectivity apparent to the reader (Sprague, 2005, p. 167). Feminist researchers’ also call attention to the role their own biography and interests play in shaping the research question and design, in order to reveal their biases and blind spots (Sprague, 2005, p. 167). Finally, some researchers advocate presenting the research process in a narrative format, in order to make draw attention to the role a researchers’ subjectivity has in the process of knowledge construction (Sprague, 2005, p. 167).
My social location and perspective are complex (Sprague, 2005, pp. 71-2, p. 74). On the one hand, I am a member of the academy and have access to certain privileges and discourses as a result; on the other, I occupy a gendered and class location and have an obligation to speak to, and from, those positions as well (Sprague, 2005, pp. 71-2, p. 74). To retain a degree of legibility (and credibility) as an academic, I have chosen to incorporate standard reporting conventions, most notably by using written rather than oral discourse, structuring the report into logically-organized chapters, and engaging in conversation with the scholarly literature. At the same time, I have become so embedded and personally invested in the research process that I feel it would be disingenuous to present my analysis and findings in an “omnipotent” or “objective” authorial voice. Therefore I have tended to use personal pronouns liberally, highlighted my own processes of sense-making, and drawn attention to my own subjectivity and interests where this seems relevant to the discussion.
I admit that strategies that take this report away from conventional academic discourses have limitations. While the use of personal pronouns reveals the boundaries of my own knowledge, some may read a lack of authority or confidence into this text (particularly if their own interpretive sensibilities lean more towards positivism than interpretivism or postmodernism). I may also run the risk of coming across as self-indulgent. From a feminist perspective, it is more important that I do not focus so much on my own voice and experiences as a researcher that I marginalize the voices of informants (Sprague, 2005, p. 169). Although I cannot entirely control how this text is read, I can attempt to manipulate the conventions to strike some balance between legibility in the academic world, generate an accessible discussion, while maintaining focus (Sprague, 2005, p. 169).
At the very least, it is apparent that I must balance a “personalized” writing style with voices from the scholarly community, as well as informants’ (Sprague, 2005, p. 171). Depending on the topic under discussion, it may not be possible (ironically) to attain “substantive” equality of representation in this regard, particularly if I am going to engage with the conventions of standard academic discourse at all. Therefore, I have made every attempt to highlight my own voice throughout the text, since I did shape it through conscious and unconscious sense-making strategies. Although academic discourses are likely inseparable from my own subjectivity at this point in my academic career, I highlight the voices of other scholars where it is logical to do so (in the literature review and analysis). Finally, participants’ voices come out primarily during the analysis. Although text tends to ossify knowledge, I trust that the reader will accept that these words reflect an on-going conversation between myself, the scholarly community, and informants.
/* Style Definitions */
mso-padding-alt:0cm 5.4pt 0cm 5.4pt;
mso-fareast-font-family:”Times New Roman”;
mso-bidi-font-family:”Times New Roman”;
Baxter L., & Babbie E. (2004). The Basics of Communication Research. Toronto: Thompson Learning.
Kirsch, G.E. (1999). Ethical Dilemmas in Feminist Research: The Politics of Location, Interpretation, and Publication. New York: State University of New York Press.
Miller, M. (2008). Branding Miss G_: Third Wave Feminists and the Media. Toronto: Sumach Press
Pitts, V. (2003). In the Flesh: The Cultural Politics of Body Modification. New York: Palgrave MacMillan.
Queen, C. (2010). Sex Radical Politics, Sex-Positive Feminist Thought, and Whore Stigma. In R. Cochrane (Ed.), WMST 331 Book of Readings Winter 2010 (pp. 130-135). Calgary: University of Calgary Bookstore.
Sprague, J. (2005). Feminist Methodologies for Critical Researchers: Bridging Differences. Walnut Creek, CA: Altamira Press.