--- breaking the silence ---Toward a Theory of Women's Doctoral Persistence
© Roberta-Anne Kerlin, 1997
CHAPTER THREE
METHODOLOGY
Research is the "production of a publicly scrutinizable analysis of a phenomenon with the intent of clarification" (Reinharz, 1992, p. 9).The purpose of this study was to examine the nature of women doctoral students' experiences, the meanings women attach to these experiences, and through this knowledge to develop a better understanding of the factors that contribute to the persistence of women who pursue the doctorate.
Design of the Study
We can, and I think must, look upon human life as chiefly a vast interpretative process in which people, singly and collectively, guide themselves by defining the objects, events, and situations which they encounter .... Any scheme designed to analyze human group life in its general character has to fit this process of interpretation (Blumer, 1956, p. 686).
It was essential to the design of the study that those who might volunteer to participate met certain criteria. The study was designed to include only those who, at minimum, had attained candidacy in their respective doctoral programs. It was also important to strike a relative balance between those who were doctoral candidates and those who had already completed their degrees. The criteria for including Ph.D. graduates in the study was that they must have completed their degrees within the previous year. The intent was to ensure that recollections of their doctoral experiences remained relatively fresh in their minds. It was also critical that all participants in the study indicate their willingness to engage in active self-reflection and self-disclosure about their doctoral experiences. Criteria for Selecting Study Participants
In December, 1994, I posted a request to the Women's Studies electronic discussion list inviting doctoral students and their partners who might be interested in writing to me about their own doctoral experiences, or those of their partners', to participate in the study. In all, 51 people responded to the request, of whom 46 were women. The 5 male respondents included 4 partners of women doctoral students, but only one d Invitation to Participate in the Study
ral student. Due to the gender imbalance of 1
and 46 women I chose to focus the study on women's doctoral experiences.
In my initial contacts with the 46 women we discussed the nature of the study and their participation as well as processes for maintaining the confidentiality of their identities and the privacy of their communications. Following these preliminary formalities I shared with each of them five broadly focused questions designed to use storytelling as a form of narrative inquiry to simulate an open-ended interview process. In developing these initial questions it was my intent to encourage the participants to write about their experiences with increasing detail: Storytelling/Narrative as a Method of Inquiry
Following an initial period of correspondence, five women expressed interest and curiosity about the study but declined to participate further due to other commitments and/or a lack of time. Six other women who initially stated an intention to write in detail describing their experiences did not continue to correspond. Of the 35 remaining women, 27 wrote describing isolated events or turning points in their doctoral experiences, but for various reasons, did not choose to correspond in further detail. The 8 remaining women agreed to formalize their long term commitment to participate in the study by signing letters of informed consent; ultimately, only 7 of the women followed through with this commitment.
- I would like to know what motivated you to do a doctorate ... why you wanted to do the degree in the first place.
- I'd like to know a little about your background ... how you came to be oriented to your field of study.
- Regarding your expectations of the doctoral experience, if you think back to the days before you entered your doctoral program, can you describe what the degree represented to you then and what you thought the process would be like?
- Now that you have progressed to the point you are in your program, in what ways have your views about the degree and the degree process changed?
- Please describe, in as much detail as possible, the story of your doctoral experience, giving particular attention to the critical events and challenges you have faced and the way in which these events have influenced your academic, professional and personal development.
As each of the women's stories progressed I responded to their notes, asking questions to probe for additional information, either in the details of the events they described or in their reflections about their experiences, for the purpose of seeking further clarification of my understandings. These exchanges were a continuing process. I anticipated that some women, particularly in the initial stages, might wish to write about their experiences without having their stories interrupted or distracted by conversation. In this regard, it was my intent to be a good 'listener' - that is, I made every effort not to 'talk' or 'write' over the participants' voices; as much as possible I encouraged participants' to tell what was meaningful to them about their experiences without introducing issues I thought might have been important; I used their own language in my responses rather than terminology of my own. I anticipated that as the storytelling progressed some participants might wish to assume a more conversational mode of interaction. In balancing my roles as listener and conversationalist I made every effort to be guided by the women's individual preferences.
In the year between December, 1994 and December, 1995, these seven women corresponded regularly with me about their doctoral experiences using electronic mail as the primary method of communication. The individual transcripts of our communication ranged between 100 and 200 or more, single-space typed pages. Duration of the Study
Five of the 7 women were quite comfortable communicating by electronic mail. Three women expressed some concern about the privacy of the electronic medium. One woman took the initiative early in our correspondence to verify my connection with the university by telephoning and leaving a message for me in my department office; she was satisfied that I was who I claimed when I sent her a response by electronic mail asking about the nature of her inquiry. Two women had concerns about the privacy of electronic mail with regard to some aspects of their doctoral experiences and while both used electronic mail for some of our correspondence, they chose to send some of their written communications via regular surface mail; one sent hand-written accounts of some of her experiences; the other sent some of her reflections about her experiences to me on a floppy disk. Women's Responses to the Use of Electronic Mail
The use of semi-structured interviews has become the principal means by which feminists have sought to achieve the active involvement of their respondents in the construction of data about their lives (Bologh, 1984, p. 388). Rationale of Storytelling/Narrative as a Method of Inquiry
Whereas structured interview formats aim to capture "precise data of a codable nature in order to explain behavior within preestablished categories ... [the unstructured interview] is used in an attempt to understand the complex behavior of members of society without imposing any a priori categorization that may limit the field of inquiry" (Fontana and Frey, 1994, p. 366). Unstructured interviews have the advantage of situating any prior conceptions held by the researcher in the background and giving priority to the participants' own conceptions of their experiences. The disadvantage of an unstructured interview format is that lack of a specific focus may tend to produce a great deal of material that may not be closely connected with the research. When time is at a premium the unstructured interview may not make the best use of this limited resource.
In using storytelling/narrative as a method of inquiry it was my intention to guide participants, with a minimum of direction, in the telling of their stories, and to encourage deeper levels of reflection and analysis without limiting or restricting their focus. The open-ended or semi-structured interview format that I developed is situated between the two extremes of the structured and unstructured interview and although this approach may require a greater length of time than a structured interview, it has the advantage of allowing the participants to raise new issues and concerns that I, as a researcher, had not conceptualized as being pertinent.
This method of inquiry shares some of the characteristics of other research techniques such as oral history and interpretive interactionism. It has been suggested that "often oral history is a way to reach groups and individuals who have been ignored, oppressed, and/or forgotten .... to understand and bring forth the history of women in a culture that has traditionally relied on a masculine interpretation" (Fontana and Frey, 1994, p. 368). Interpretive Interactionism is said to follow from more open-ended styles of interviewing, but in addition, it adds a new element borrowed from religious usage by James Joyce, that of epiphanies (Fontana and Frey, 1994). Epiphanies are described by Denzin (1989, p. 15) as "those interactional moments that leave marks on people's lives [and] have the potential for creating transformational experiences for the person." Through open-ended questioning and gentle probes for deeper reflection, this method of inquiry is intended to create an atmosphere of trust which will allow such interactional moments of insight to emerge from the experience.
Phenomenological interviewing is described by some feminists as an "interviewee-guided investigation of a lived experience that asks almost no prepared questions .... Feminist phenomenological interviewing requires interviewer skills of restraint and listening as well as interviewees who are verbal and reflective" (Reinharz, 1992, p. 21)The method I developed can be thought of as a feminist phenomenological process insofar as it begins with women's experiences and relies heavily on the participants' willingness to write extensively about their experiences. Interviews began with open-ended questions that were intended to provide a broad structure for the participants' stories. As I receives their stories, I developed further questions using the participants' own language, to garner additional detail and clarification of their meanings and intents.
The method of interviewing was to follow the language and the logic of the person's thought, with the interviewer asking further questions in order to clarify the meaning of a particular response (Gilligan, 1982, in Reinharz, 1992, p. 42).In most instances the questions I posed arose directly from the information presented to me by the participants. In some instances, particularly in the latter stages of inquiry, my probes, which I embedded in the context of their stories and our written exchanges, were based on a series of questions which I formulated during the pilot study (see Appendix E, Intermediary questions 6-21). For this reason it was necessary to engage the participants in repeated interviews, or ongoing conversations, about their experiences. I encouraged them to digress in directions of their own choosing and to ignore questions of mine which they felt were not important to their experiences. Using this approach it was my intention to maximize each participant's control over her own story and not direct the participants' stories in ways they might otherwise not intend or wish.
Multiple interviews are likely to be more accurate than single interviews because of the opportunity to ask additional questions and to get corrective feedback on previously obtained information .... Multiple open-ended interviews are well suited to understanding how a woman develops her ideas. They can be done, however, only among interviewees who have time to invest in the process (Reinharz, 1992, p. 37-38).The benefits to both researchers and study participants in using electronic mail to conduct research and, in particular, to construct personal narratives are mixed. Electronic mail facilitates communication between individuals who are separated in both time and place and thus allows me as a researcher to communicate with study participants who, otherwise, I would have been less likely to meet face-to-face. This distance between the researcher and the researched can affect participants differently. For some, control over their participation in the study is maximized by remaining at a distance from the researcher in the familiarity of their own milieu, by having complete control over what they choose to communicate and by having the opportunity to participate at their own convenience or even withdraw from the study at a time of their own choosing. In contrast, those who might prefer the intimacy of face-to-face contact with the researcher can find the absence of social cues in electronic communication to be unsettling and may be less inclined to participate in such a study. The use of electronic mail to conduct this kind of research creates a new kind of relationship between the researcher and the researched in which both become bound by the emotional intimacy that emerges from sharing very personal experiences while at the same time we remain relative strangers according to the traditional conventions of interpersonal relationships.
The speed of electronic mail, though sometimes slowed by busy networks, typically facilitates exchanges that seem to be instantaneous. When two individuals communicate synchronously the interaction can resemble the conversational turn-taking of verbal exchanges. Asynchronous exchanges, on the other hand, provide individuals with the opportunity to reflect on their thoughts to a greater degree than they are able to in face-to-face communication.
This method of inquiry has a particular advantage over telephone and face-to-face interview methods. As Jacks, Chubin, Porter and Connolly (1983) discovered when using telephone interviews, the initial responses given by interviewees were not necessarily the most significant from the respondents' perspectives. The extended duration of the current study and the repeated exchanges with the participants over a one year period allowed time to earn the participants' trust and to probe for additional meanings as well as to obtain clarification of my own understandings.
A further advantage of electronic mail is the resulting transcript which provides an artifact of the conversation that remains in its original form and is accessible to both sender and receiver for continued reflection and elaboration. For researchers this transcript eliminates the time and cost of transcribing taped conversations. However, the transcript is limited to 128 text characters and reflects a narrow spectrum of communication. Absent are the normal cues associated with face-to-face communication including non-verbal gestures, pauses, hesitations and the tonal qualities of voice. Experienced users of electronic mail develop strategies to compensate for this reduced availability of social cues. Emoticons such as the horizontal smiley face, [ :-) ] and the wink [ ;-) ] are often used to communicate mood or emotion. Some will interject bracketted expressions of intent to clarify meaning as in the following example:
> I liked your comment about wanting to prove that you were "smart enough (or bull-headed enough)" to complete the degree! (GRIN)None of these strategies compensate completely for the reduced availability of social cues when using electronic mail. However, the opportunity for participants to review and reflect on what has been written does serve as an important trade-off to this limitation.> Sometimes the asterisk [ * ] or the underscore symbol [ _ ] is used to add emphasis to particular words or phrases:
> Nobody else has mentioned this and I'm *sure* every one of us has gone through it...
> Needless to say, when you follow an act like that, as I did all through my school years, you are constantly compared (by teachers and parents) to your older (and, by inference, _smarter_) older brothers.
Multiple interviews characterize much feminist research perhaps because multiple interview research helps form the strong interviewer-interviewee bonds some people define as characteristic of feminist research (Reinharz, 1992, p. 36). The Role of the Researcher
In structured interviews, particularly survey interviews, the traditional role of the researcher as interviewer has been one of an interested, but affectively detached observer who plays "a neutral role ... on the one hand, casual and friendly but, on the other hand directive and impersonal" (Fontana and Frey, 1994, p. 364, 367). However, in recent years feminist researchers have questioned whether such objectivity is even possible. Feminists also have questioned the underlying assumptions, ethics and authenticity of an interview paradigm that reflects masculine traits like detachment to the exclusion of feminine traits such as sensitivity and emotionality (Fontana and Frey, 1994, p. 367; 370). LeCompte (1993) has suggested that "positivistic science imposes a false distance between researchers and the researched by mandating that the researcher maintain an artificially impersonal stance toward the people studied" and that this detached perspective results "in data that present a partial and therefore false, and an elitist and therefore biased, reality" (p. 11-12). LeCompte suggests that authenticity is achieved, not merely by attributing a sense of genuineness to the quality of the narrative, but that authenticity is reflected in the relationship that exists between the researcher and the researched. She argues further that authenticity cannot be achieved when those who are researched are placed in a position that is subordinate to that of the researcher.
The poststructural remedy to the positivistic canon of conventional science is to overturn old dichotomies between research/practice, author/text, subject/object, knower/known, method/procedure, and theory/practice (LeCompte, 1993, p 12).LeCompte presents a new paradigmatic role in which the researcher serves as the mediator between two silences, the silence within and the silence without. In the latter case, the silence without, the researcher's role can best be understood as mediating between those in power and those who are silenced and oppressed by that power - as LeCompte (p. 10) describes it, those who "have been deprived of voice without their consent." This has been the traditional perspective of critical and feminist researchers as mediators between the powerful and the powerless. It is the former case, mediating the silences within, which may be viewed as characteristic of a new research paradigm. These silences within occur at two levels as LeCompte describes them: among peers or research participants, and within themselves. From this perspective LeCompte views the researcher's role as that of a mediator who assists participants in giving voice to their own thoughts and understanding to the events and circumstances in the larger context of their own lives. She suggests that the power to name and describe this reality resides first and foremost with those who have been silenced and that this step is pre-requisite to the second phase which transcends awareness and empowers the oppressed to engage in activism: "someone in the equation, someone other than the researcher, has to want to change the situation, take action, and define the change as both possible and worthwhile" (p. 15).
I see this as my role as a researcher: to serve as a mediator in the 'outering' of these inner silences and to make visible these silences to those in positions of power who might otherwise not see or choose to know them. This I call,
--- breaking the silence ---
My interview style emphasizes the importance of developing rapport with the participants. It is reflective and, with appropriate cautions, self-revealing. I used the language of the participants so as to avoid naming their experiences for them. Through active listening - through written responses that actively engaged their issues and ideas - I made every effort to provide an atmosphere of engagement and trust which allowed participants to develop ideas and construct meaning, to share attitudes and feelings which typically are not quantifiable and usually are missed in survey and structured interview research.
British sociologist Ann Oakley posited a contradiction between 'scientific' interviewing requiring objectivity, and feminist research requiring openness, engagement, and the development of a potentially long-lasting relationship. She advocated a new model of feminist interviewing that strove for intimacy and included self-disclosure .... Guiding this new model was a proposed feminist ethic of commitment and egalitarianism in contrast with the scientific ethic of detachment and role differentiation between researcher and subject (Reinharz, 1992, p. 28).I anticipated that participants might wish to introduce new questions of their own and that they would request personal information about me or information about the progress of my study, either or both of which I openly shared with them. I also anticipated that the interview process would evolve in such a way that the narrative assumed a more conversational character. While my own self-disclosure served as a model of openness to aid in building trust with the participants, the degree to which I disclosed information about myself was not without risk to the participants. Regardless of my intent as a researcher, participants in a study may not always receive a researcher's self-disclosure in a constructive way. Many feminists argue that "researcher self-disclosure during interviews is good feminist practice .... [but] there is no single feminist perspective on researcher-interviewee relations and self-disclosure. Rather there is an openness to numerous possible meanings of these phenomena" (Reinharz, 1992, p. 32, 34). It has been my experience in online conversations that both the timing and the degree of self-disclosure is critical. It cannot occur too early in terms of each participant's readiness to receive it without running the risk of overshadowing or even interfering with the participant's story; nor can it be delayed much beyond the participant's own need to know without causing participants to feel as if they are being exploited or 'used' for research purposes.
Ethical questions are heightened in feminist interview research because feminists try hard to avoid perpetuating the exploitation of women (Reinharz, 1992, p. 27).The risk to participants regarding premature self-disclosure on the part of the researcher is well illustrated in the following passage by Bombyk, Bricker-Jenkins and Wedonoja (1985). As one interviewee wrote about the interviewer's self-disclosure:
Personal sharing on her part (where she was from, what she has done, some of her own views) was triggering off in me a self-censoring process. I began to notice myself stereotyping her and second guessing what she would want to hear and not want to hear based on my perception of the information about herself .... She was giving me ... personal information as a way of equalizing the relationship and revealing herself as I had been revealing myself, yet it seemed more out of her need to self-disclose rather than my need at that point to know about her. At that early stage of the interview, I felt like I first needed time to establish myself within the role of participant before moving towards more of an interactive sharing.Fortunately, the interviewee notified the interviewer:
Once I voiced my concerns ... I was able to influence the process and it contributed to my sense of safety and trust. The fact that [she] was responsive to my concerns and took time within the interview process for ongoing feedback made a big difference in creating an atmosphere that facilitated self-exploration and self-disclosure. As the interview progressed, I was delighted to learn more about Mary and to have a dialogue about some of the topics ... I began to see how it was unrealistic to think that such a relationship within a brand new situation like this would be able to be "instantly" created - as some of the guidelines of feminist research seem to suggest - without some form of a developmental process and adaptation to the unique needs and concerns of the individuals involved (Reinharz, 1992, p. 33).One of the interviewers, Bricker-Jenkins, wrote, "Thanks to feedback from respondents ... I have learned to 'pace' my interactions and look for cues from the participant as to readiness to know more about me" (Reinharz, 1992, p. 33).
Interpreting the Stories Consistent with the perspectives reflected in critical feminist theory, grounded theory requires an 'interpersonal interaction' on the part of the researcher with both the data and the participants in the study. The researcher must not only observe the participants but must observe self-behaviour and so make visible one's own preconceptions, values and beliefs (Hutchinson, 1988). In the context of grounded theory, this concept of juxtaposing one's own understandings is referred to as 'bracketing'.
About Grounded Theory The most critical aspect of grounded theory which differentiates it from other qualitative research methods is its emphasis upon theory development (Strauss and Corbin, 1994). Theory is said to be grounded when it emerges from and generates explanations of relationships and events that reflect the life experiences of those individuals, groups and processes we are attempting to understand. Denzin (1978) describes four functions that all data serve in contributing to theory development: research data initiate new theory or reformulate, refocus and clarify existing theory. Grounded theory is considered to be particularly appropriate when little is known about a topic and there are few existing theories to explain a particular phenomenon (Hutchinson, 1988). Hutchinson also indicates that grounded theory is to be understood as a form of social criticism. This characteristic of grounded theory parallels important characteristics of a critical feminist perspective as discussed earlier and for this reason and those described above grounded theory was thought to provide the most appropriate method for data analysis in this study.
Grounded Theory Method Grounded theory is qualitative in its philosophy of science, its data collection, its methods of analysis, and its final product offers a rich and complex explanatory schema of social phenomena .... [it] is a form of social criticism; it does make judgements about identified patterns of social interaction (Hutchinson, 1988, p. 126).The method of grounded theory involves specific procedures which, when applied appropriately and with vigilance will result in theory that is rigorous and well grounded in the data. Criteria for ensuring the quality of grounded theory will be discussed following a description of the procedures involved in developing grounded theory.
The first procedure, data recording, may be thought of as a pre-analytic step of the grounded theory method. The immediate recording of data is said to be essential to the successful generation of grounded theory (Hutchinson, 1988). According to this criterion, the methods for recording data in this study are particularly appropriate. Using written narrative in the form of electronic mail, the participants in this study actually recorded the data themselves. This leaves little room for error in the original data set.
Data coding represents the first phase of data analysis. Hutchinson (1988) describes three levels of coding in the grounded theory method. Glaser (1978, p. 57) describes a set of three questions that should guide the open coding (level one):
Open coding describes the action and behaviour of the participants.
- What is this [sic] data a study of?
- What category does this incident indicate?
- What is actually happening in the data? What is the basic social psychological problem(s) faced by the participants in the... scene?
The analyst compares incident to incident with the purpose of establishing the underlying uniformity and its varying conditions (Glaser, 1978, p. 49).Such a description of coding is applied readily to an 'observed' setting. In the case of written narrative where storytellers are describing their own experiences, this concept of coding must be also include expressed thoughts, beliefs, feelings and described events and relationships. Hutchinson suggests that open coding, the coding of each sentence and each incident, should be used to develop as many codes as possible to ensure full theoretical coverage and to prevent the researcher from imposing any preconceived impressions on the data. "These beginning codes, no matter how conceptually primitive, quickly start theoretical sampling and constant comparisons of incidents. How relevant these concepts are to the basic problem and basic social process becomes a question of further analysis" (Glaser, 1978, p. 45).
Data coding at levels two and three is intended to elevate the data to higher levels of abstraction (Hutchinson, 1988). Level two codes typically represent categories that describe the level one codes.
The analyst continues to code and ... compares the concept to more incidents (Glaser, 1978, p. 50).For example, in a message in this study one of the participants described herself as an overachiever. This occurrence of 'overachiever' was coded as level one. At a more abstract level, level two, this construct represents her academic self-concept. I then began searching the data for other occurrences and representations of academic self-concept. This particular construct developed in the following way.
Level three codes represent theoretical constructs derived from the data in combination with academic knowledge and knowledge acquired through praxis.
... while the first and second type of comparisons [continue] throughout the study, the analyst also compares thirdly concept to concept with the purpose of establishing the best fit of many choice of concepts to a set of indicators, the conceptual levels between concepts that refer to the same set of indicators and the integration into hypotheses between the concepts, which becomes the theory (Glaser, 1978, p. 51).The constant comparative method is central to the data analysis in generating grounded theory. Using this method all the sample codes generated at each of the three levels are compared repeatedly within and between each other until the basic properties of a category or construct are defined. "Comparative analysis forces the researcher to 'tease out' the emerging category by searching for its structure, temporality, cause, context, dimensions, consequences and its relationship to other categories" (Hutchinson, 1988, p. 135). Additionally, it is appropriate and desirable to compare the data categories and constructs that emerge between various groups of participants in the study. In this way the process of constant comparison is intended to generate a theory rich in detail.
Theoretical sampling is the process of data collection for generating theory whereby the analyst jointly collects, codes, and analyzes ... data and decides what data to collect next and where to find them, in order to develop ... theory as it emerges. This process of data collection is controlled by the emerging theory, whether substantive or formal (Glaser, 1978, p. 36).Theoretical sampling begins during the data collection phase of the study and involves searching the transcripts for emerging categories that characterize the narrative and seem significant. As constructs are derived from the data repeated theoretical sampling can be used to increase the depth of focus and to ensure consistency; that is, to ensure that data are gathered in a systematic way for each category (Strauss and Corbin, 1990). "Theoretical sampling is ... used as a way of checking on the emerging conceptual framework rather than being used for the verification of preconceived hypotheses" (Glaser, 1978, p. 39). Saturation is achieved when all the data fit into the established categories and no new categories emerge from the data.
The essential relationship between data and theory is a conceptual code .... There are basically two types of codes to generate: substantive and theoretical. Substantive codes conceptualize the empirical substance of the area of research. Theoretical codes conceptualize how the substantive codes may relate to each other as hypotheses to be integrated into the theory (Glaser, 1978, p. 55).Theoretical sampling and constant comparison reflect cyclical processes which are fluid and flexible, but at the same time they ensure that the analysis is planned, rather than haphazard, and well grounded in the data. Through this process a core variable or basic social psychological process is identified which explains most of the variation in the data. "The core variable has three essential characteristics: it recurs frequently in the data; it links the data together; and it explains much of the variation in the data" (Sherman and Webb, 1988, p. 133). By relating this core variable to the various levels of codes already identified, the critical factors emerge and provide the basis for writing about the theory.
Just as the grounded theory method of analysis is not applicable to experimental studies that seek to verify hypotheses, neither should the criteria for scientific rigor derived from positivistic origins be applied to the grounded theory method. Positivistic notions of validity, reliability and generalizability cannot be applied in the same way to qualitative research. Nonetheless, there must be some criteria by which the quality of grounded theory research can be evaluated. Sherman and Webb (1988) identify six such categories including the degree of fit, functionality, relevance, modifiability, density, and integration. Ensuring Rigor and Empirical Grounding of the Study
The degree of fit is described as resulting in codes and categories that are derived from the data and not forced. This lends credibility to the study in that the appropriateness of the fit can be easily understood by others not directly involved in the study. "Since most of the categories of grounded theory are generated directly from the data, the criteria of fit is automatically met and does not constitute an unsatisfactory struggle of half fits" (Glaser, 1978, p. 5). Glaser (p. 5) suggests that "it is important to constantly refit [categories] to the data as the research proceeds to be sure they do fit all the data they purport to indicate." Although Sherman and Webb do not use the term 'functionality' per se, this is their intended meaning for describing a theory that 'works.' As such, a functional theory explains variation in the data and the interrelationships among the constructs in a way that produces a predictive element to the theory. They further describe a quality theory as one that possesses relevance related to the identified core variable or basic social psychological process. Relevance evolves through the emergence of a core variable from the data in a way that is neither forced nor concocted and is a result of the researcher's theoretical sensitivity to the milieu. Relevance is verified through the immediate recognition by the participants in the study of the importance of the phenomenon - a form of recognition that sometimes has been described as the "ahhh haaa" phenomenon. The fourth criterion of a well-grounded theory is its ability to reflect and accommodate the fluctuating nature of the phenomenon being examined. As such the theory must be flexible and modifiable. The fifth criterion is density. A theory is said to be dense when it "possesses a few key theoretical constructs and a substantial number of properties and categories" (Sherman and Webb, 1988, p. 138). The last criterion described is that of integration. A systematic relationship between the constructs and propositions is thought to ensure an appropriate fit into a tight theoretical framework (Glaser and Strauss, 1967).
Strauss and Corbin (1990, p. 254-256) also put forth a series of questions which they view as appropriate criteria for examining the empirical grounding of a study. These questions are as follows:
The criteria described by Glaser and Strauss (1968), Sherman and Webb (1988) and Strauss and Corbin (1990) provide a sufficiently broad perspective from which to evaluate the quality of a grounded theory. Perhaps one additional criterion that would lend credibility to the theory is a measure borrowed from other qualitative research genres, that of auditability. The provision of an audit trail provides evidence for the way in which processes are carried out and decisions are made, thus making the process both visible and verifiable to others who might wish to closely scrutinize the theory.
- Are concepts generated?
- Are the concepts systematically related?
- Are there many conceptual linkages and are the categories well developed? Do they have conceptual density?
- Is much variation built into the theory?
- Are the broader conditions that affect the phenomenon under study built into its explanation?
- Has process been taken into account? (explanations that describe change must be linked to the conditions that caused it)
- Do the theoretical findings seem significant and to what extent?
Striving for and attaining narrative rigor requires a lens through which one views and understands a phenomenon and a method to ensure the completeness of the process. Both processes are central to the role of the researcher.
Ensuring Narrative Rigor LeCompte (1993) identifies two elements, double description and double consciousness, which she describes as providing a theoretical frame for conducting research on silence. She uses the analogy of binocular vision to describe the concept of double description. Monocular vision, she suggests, can produce only a unitary vision. In contrast, binocular vision "involves seeing a phenomenon with both eyes so that, in effect, the researcher has two or more descriptions rather than one" (p. 16). Double description involves a blurring of the boundaries between 'subject' and researcher, sacrificing the clarity of a unitary description and meaning for a vision that reflects greater depth and diversity. Double description is rooted in double consciousness - a collaboration with and embracing of the other
in ways that change researchers and those they study so that their destinies are inextricably linked and shared. Such consciousness must transcend the self-consciousness customarily called for in critical and collaborative research. It is required for the ethical presentation of personal narrative because the researcher's notes usually constitute the only data, construction of reality, text to be interpreted, or story to be told (LeCompte, 1993, p. 17).These concepts of double description and double consciousness, as described by LeCompte, share some of the same qualities as those concepts which Lincoln (1993) considers to contribute to narrative rigor: scrutiny, persuasiveness, isomorphism, and authenticity. Lincoln suggests that narratives (both sought and presented) that reflect fidelity (validity) and rigor must fulfill the criteria of these four elements. She argues that narratives must be persuasive, not only to research and policy communities, but most importantly, to those whose lives they represent and, therefore, that narratives, particularly co-created narratives, are subject to the scrutiny of those who are researched. The criterion of internal validity is reflected in the construct of isomorphism and suggests that narratives will reflect a closer one-to-one correspondence with the lives of the researched. Lastly, isomorphism, in combination with a narrative that conveys the appropriate 'feeling tone' of the represented lives, is thought to reflect authenticity. As such, the "range of mood, feeling, experience, variety and language" portrayed in the narrative should create for the reader, a heightened sensitivity to the lives of the researched.
On December 18, 1994, I posted a request to the Women's Studies electronic discussion list (WMST-L) inviting doctoral students and their partners who might be interested in writing to me about their educational experiences to participate in the study. In all, 55 people responded to the request, of whom 46 were women. The other nine respondents included partners of women doctoral students and male doctoral students. Five women withdrew in the early stages of the pilot study, primarily due to other commitments and lack of time and six others who intended to write more extensively also did not continue to write. In all, 35 women continued to write to me about their doctoral experiences between December 1994 and the end of February 1995. The Pilot Study
Selecting Participants for the Pilot Study
One point is worthy of mention. The letter inviting participation in this study (Appendix B) included the following statement: "It is my purpose to understand better, the many personal and professional issues that may impede completion of the doctorate in a timely fashion." This phrasing may have attracted participants who were particularly conscious of factors influencing their progress. However, as discussed earlier, because women do have longer times to degree than men and because women's completion rates are lower than men's, it is important to understand not only the factors that enhance student progress, but those factors that impede progress and ultimately may influence Ph.D. completion rates. As Lipschutz (1993, p. 70-71) points out completion rates in the range of 50% are difficult to justify from both an economic perspective and in terms of program effectiveness.
> Without addressing the question of how many Ph.D.s American universities should be producing, we can nonetheless insist that more than 55 percent of those admitted into doctoral programs should receive the degree simply on the basis of the commitment of personal and institutional resources that Ph.D. study implies ... Moreover, as faculty and graduate deans seek the financial resources from government, foundations and the private sector that graduate education will need to sustain its quality over time, they are being asked to justify their completion rates.Electronic mail provided the primary method of communication. Generally the women seemed to be comfortable communicating with electronic mail. Only three women expressed some concern about the privacy of the electronic medium and two of them chose to share some of their experiences through written communications via regular surface mail. One woman took the initiative to verify my connection with the university by telephoning and leaving a message for me in my department office; she was satisfied that I was who I claimed when I sent her a response by electronic mail asking about the nature of her inquiry. Printed, single-spaced, my correspondence with the 55 respondents fills four 3-inch binders.
In my initial contacts with each of the women we discussed the nature of the study and their participation as well as processes for maintaining the confidentiality of their identities and the privacy of their communications (Appendices D: Human Subjects Research Consent Form, and E: Protection of Human Subjects). Following these preliminary formalities I shared with them five broadly focused, open-ended questions as a way to help them begin describing their experiences (see Appendix C, questions 1-5). As the women wrote to me I read and responded to their notes, asking questions to probe for additional information, either in the details of the events they described or in their reflections about their experiences, and to seek further clarification of my understandings. Initial Exchanges with the Women
Very early in our correspondence some of the women expressed curiosity about who I was and what my personal interest was in conducting the study and I openly shared this information with them answering any questions they put forward. I attempted to match my responses to the degree of detail the women seemed to be requesting. Other women seemed less curious about me and my motivations and have been more intensely focused on telling their own stories.
Of the 46 women who responded initially, 10 had completed doctorates with six receiving their degrees in 1994. Thirty-five women were registered in doctoral programs in the United States and one was registered in Canada. All 35 had completed their candidacy exams, had defended successfully the preliminary oral examination, and were actively engaged in the dissertation writing phase of the doctorate. None of the women had pursued degrees in non-traditional fields of study; all pursued doctorates in the arts, humanities and social sciences. The respondents included single heterosexual women, single mothers, married women with children and grandchildren, married women without children, and two women who identified themselves as lesbians.
About the Women Respondents
In their stories the women described a range of experiences in connection with their doctorates; some experiences were self-affirming; others were undeniably injurious to the women's well being. Among the 10 women who wrote most extensively about their experiences, despite their concerns and self-doubts, as best I could tell, I was convinced that each one of them was a 'finisher.' The women were goal oriented and persistent, some to the point of stubbornness, and willing to sacrifice other important personal needs to achieve their goals. Generally they saw themselves as capable individuals willing to "risk oneself at something that is important."
Preliminary Findings
Preliminary themes emerging from this pilot study were presented at a paper symposium at the Annual Meeting of the American Educational Research Association in San Francisco on April 20, 1995. [See Hidden Rules, Secret Agendas: Challenges Facing Contemporary Women Doctoral Students] The theme of the paper focused on the challenges facing contemporary women doctoral students. In preparation for the conference several drafts of the paper were shared with the women participants to obtain their feedback. They made many helpful suggestions at various stages that improved the final paper in important ways and none of the women indicated that the ideas expressed were less than representative of their experiences. Three sample responses are included below.
Feedback from the Participants
Bobbi:Thanks for sharing the aera paper. It really is moving and frustrating and uplifting and horrific and ... well, too close to home!
Karen
Bobbi,
I read the paper and it is an impressive piece. Much of the historical info was not new to me, as I have had a couple of history of higher ed and history of women in higher ed courses, so the stats were not new. However, you summarized the material more succinctly (and usefully) than most writers. I assume that much of this will also form your lit review for the D?
The section where you started talking about the women's experiences, though ... that's where the whole piece comes alive. Up to that, it's a useful--but let's face it, typical-- academic piece. The background is necessary to understand the current situation. But it was amazing to me, as a participant and also as a reader, that the voices of your participants not only mirror and support each other so well, but also mirror the statistical and historical data. And do so in a much more interesting and intense way than any table, trend line, or standard deviation ever could!
I found myself reading the paper and exclaiming out loud. Keith was making dinner and would come into the living room each time I did so to see what it was that had engendered the reaction. Several times I said out loud, "Thank God, I thought it was just me!" It was very interesting-and heartening!-- to see that several other participants had similar thoughts and experiences during their doctoral program. While you're going through it, you feel so alone ... it's tough to imagine or realize that others have had many of the same experiences (or perhaps even worse) and also felt that there was no light at the end of the tunnel.
Maybe you should develop a new course for doctoral students ... before they graduate, they have to go through a "debriefing" to ensure that they will be able to survive and thrive after the process is over?!
Well, as I've said before, I've really enjoyed and learned from this experience, and I am happy to see that you have been able to put my thoughts and those of the other participants to such good use. Only wish I could be there to clap for you at the session! Have a great time!
Sarah
Bobbi,
I just got through reviewing the second/final draft of your paper. WOW!!!!!!
When I was reading through Barbara's story when you sent me the more preliminary version I started to cry. So I had to stop reading and finally managed to get through the whole thing this afternoon.
My reaction today is that I wish I could march into my dept and impress upon the faculty there that this is a real issue. The faculty at my university have been confused and incredulous of our complaints about not being supported financially and/or intellectually/emotionally (morally?).
Well, I'm probably not going to do that. But it has been striking to me to understand that what I have experienced and some of my friends is a phenomenon that is present (at the least) across the whole of the humanities. I don't feel nearly so alone.
I am realizing that I have little fear of the final oral defense because there hasn't been one in so long; only three that I can think of in my 6 years affiliated w/ this institution. But no horror stories from the three. So that's promising.
I will be sharing what you've written about with several of my friends, many of whom already know about my participation in your group because I have thought it to be so interesting and so necessary. I will mostly discuss it with them but I know a few may ask to see the paper. I hope you don't mind if I forward it. I realize that there is a risk involved in distributing not yet published work. But I won't pass it to anyone I don't trust and wouldn't pass my own work to.
Thanks for doing this study, Bobbi. I am impressed with what you've sent me. Please be proud of this work. I am proud of you and of my participation and ability to help you.
[insert sound of tremendous applause here]
many smiles,
Tracy
The pilot study proceeded very much the way I had anticipated it would. However, two minor changes in procedure were thought to be useful for the larger study.
Implications of the Pilot Study
From among the 35 original women respondents, I invited 7 women who had continued to write with the greatest detail and clarity about their doctoral experiences to formalize their long term commitment to participate in the study by signing letters of informed consent. Criteria for inviting the women to participate in the study included their willingness: (1) to continue written exchanges with me about their doctoral experiences, (2) to see the project through to the end; this involved the willingness to maintain contact with me until I had written the final chapters of my dissertation and to provide me with ongoing feedback about my work. While I recognize that this group of seven women necessarily is not representative of all women doctoral students there is, nonetheless, much we can learn from the experiences they recount herein.
- I had thought that integrating my 'questions' into my responses to the women's stories would be the most effective way to engage them in a dialogue about their experiences. This strategy worked very well except with a few women in the very beginning stages of the pilot study. Some of these women were probably anticipating a more structured or formal approach such as a questionnaire and when I embedded the initial questions in my replies to them, some women did not recognize them as being part of the study. Three or four of them responded to these notes and then asked when the study was going to begin. I quickly adjusted my approach with subsequent respondents so that the initial questions were numbered and clearly distinct from other conversation in my notes. This seemed to work well and thereafter, when I embedded my questions in my responses to their notes and the participants were able to follow the exchanges quite easily. My initial purpose in embedding the questions in my notes (as opposed to explicitly structuring them as questions to be answered) was to personalize my responses to the women. However, a more structured format, at least in the beginning stages of our exchanges, seemed to be preferable.
- I thought that some demographic information about the women would be helpful in contributing to individual portraits of the participants. There seemed to be no easy way to integrate questions about age, for example, into the conversation. Therefore, I developed a short biographic questionnaire for the women to complete that addressed issues such as baccalaureate and master's degree completion dates, registered time in degree programs, time out of program, part-time employment, number and ages of children, etc.
 |