Abstract: Research within a postmodern frame moves us into arenas where subjectivity is both assumed and appreciated. This framework provides an opportunity to attend to how subjectivity (of researcher[s] and of research participants) and inter-subjectivity (between/among researcher[s] and between researcher[s] and participants) can enhance the research process. In this paper, we explore a framework that involves one such model: an understanding of qualitative research as an interconnected and mutually influential series of dialogic processes. A dialogic framework allows us to view each major aspect of a research program as having, as an important hope, the creation of synergistic communication between or among participants. Because this approach relies on ideas about dialogic communication, it carries an intrinsic investment in the reflexivity of every conversant—i.e., every researcher and every participant. It emphasizes the reflexive value of conducting research in the context of a team of researchers, and it examines the role of reflexivity at each step of the research endeavor: formulating the question, gathering information, analyzing this information, collaborating with other researchers, and "returning" the fruits of the research to participants. The paper discusses the centrality of reflexivity at each of these steps, both in descriptive terms and through illustrations drawn from our own research as well as from the experiences of other researchers.

The observations and interpretations presented in this paper reflect the vision and epistemological strategies located in a postmodern view of research. Explorations conducted within this frame attend more to questions of process and relationality than to the discovery of a pre-existing truth or singular reality (CHEEK, 1999). This approach rejects the notion of meta-narratives that seek to represent a coherence of understanding and a comprehensive view of others (LYOTARD, 1984). Instead, it emphasizes the deconstruction of rigidly formed beliefs and the acceptance of a more fluid, changing perspective. Postmodern strategies recognize the transitory and ephemeral nature of reality (CHIA, 1995) while encouraging attention to the interpersonal relationships that ultimately shape and define our experience. With an emphasis on plurality, these strategies embrace the multiplicity of voices and views present in any representation or analysis (STRATHERN, 1991). They invite and privilege observation and examination of the relational nature of research and they celebrate the subjective nature of the information gathered through this process. [1]

The subjectivity found in research information originates with both the researcher(s) and participant(s), each of whom brings individual experiences and pre-existing perspectives into the research event. These subjective views have been initiated and deepened through interaction with multiple nested systems of the environment (BRONFENBRENNER, 1979), and they serve to consistently evaluate and mediate one's unique expectations and understandings of the world. These lenses of subjectivity inform and mediate each element of the research project, influencing not only the process and intended goals but also the interaction and attributions found within the event itself. [2]

BREUER (2000) identified multiple characteristics that may draw researchers toward or away from particular research topics. Included in these are the elements of intellectual and emotional comfort, individual interest in a certain phenomenon, and attraction toward certain roles or environments that complement individual style. A researcher's choice of inquiry may also be guided by professional associations that can exert influence at both the micro and macro levels (MAYS & POPE, 1995). Researchers may be enabled or confined by departmental frameworks or scientific disciplines, which in turn reflect the broader cultural norms. Topics encouraged for exploration may be ones that are pertinent to current events or reflective of the discourse of the dominant culture. While these influences are not always readily apparent, they too serve to create a type of external subjectivity that privileges or discounts certain areas of exploration. [3]

Although the image of the researcher as a value-free and objective observer (WHEATLEY, 1992) has been replaced by one that acknowledges active participation and co-authorship of research outcomes (HATCH, 1996), limited changes have occurred in the production of scientific-psychological knowledge (BREUER, 2000). According to WIESENFELD (2000), researchers have not yet determined how to operationalize the subjective nature of research in a way that provides for expanded understanding and insight into the process. AHERN (1999) notes that, although researchers report attempts to integrate reflexive strategies into their work, they frequently are unable to clarify how this process is accomplished. These reports suggest an elusive quality to subjectivity that appears to defy simple categorization or identification. [4]

The importance of reflexivity in psychological studies has gained increased recognition as researchers have begun to acknowledge themselves as co-creators of the knowledge represented in these explorations (STEIR, 1991). As researchers attempt to balance the potential benefits of researcher involvement with a commitment to accurately represent their respondents' voices (AHERN, 1999), we enlist the strategy of reflexivity as a mediator of the research process. RUSSELL and BOHAN (1999) define reflexivity as a process of honoring oneself and others in our work through an awareness of the relational and reflective nature of the task. They emphasize that researchers: "may not stand apart from their own humanity while creating new understandings and that research is not an objective rendering of reality but a form of participation in the phenomena under study" (p.404). SCHUBERT (1995, as cited in PAYNE, 2000) adds that reflexivity "helps us identify the socially and rhetorically constructed boundaries that delimit our view of the social field, to transgress those limits, and provide a basis for creative, ethical alternatives" (p.010). [5]

Reflexivity is typically represented in the literature as a process of self-examination that is informed primarily by the thoughts and actions of the researcher (BARRY, BRITTEN, BARBER, BRADLEY, & STEVENSON, 1999; PORTER, 1993). Strategies for its implementation often include the completion of self-reflective records and diaries, the examination of personal assumptions and goals, and the clarification of individual belief systems and subjectivities (AHERN, 1999). The goal of these activities is "to turn the researcher's gaze back upon oneself for the purpose of separation and differentiation" (S. HAWES, 1998, p.100). [6]

The process of reflexivity is also enlisted in a team-based approach to research. Qualitative methodologies have increasingly integrated the benefits of utilizing multiple, diverse perspectives as a means to enhance the richness of contributions to knowledge and practice (BARTUNEK & LOUIS, 1996). As research strategies have shown growing interest in the complex nature of data, they have moved toward a model that employs a critical reflection and examination from multiple positions. In this way, researchers are able to utilize their reflexivity as a means to achieve an expansion of understanding (GERGEN & GERGEN, 1991). [7]

The process of team research also illuminates the multiple levels of reflexivity that are available for implementation within the research process. The personal self-examination encouraged by individual approaches can be further expanded into a broader process of reflection that captures the interpersonal nature of the event (BARRY et al., 1999). As the complex tapestry of the research process unfolds, the reflexive strategies can expand to encompass both individual and group subjectivities that emerge from the multiple interactions among contributors. In this way, scaffolding is created that allows researchers to make transitions between and among the various relationships and dialogs that define the broader process. As a result, the team is able to generate a synergistic environment that reflects the multiplicity of voices and allows for expanded understandings (RUSSELL, 2000). [8]

In addition to enabling a richer, conceptual analysis and interpretation (BARRY et al., 1999), team research may also offer personal benefits for those involved in the process (WEST, 1994). The interaction among team members may encourage individuals to view themselves more clearly and to develop strategies to compensate for limitations (RUSSELL, 2000). Teams also can provide emotional support to counteract feelings of isolation or to hold the emotions that are often associated with highly charged or sensitive topics (ERICKSON & STULL, 1998; KLEINMAN & COPP, 1993; RUSSELL, 2000). A well-functioning team offers opportunities for expanded personal insight, a greater sense of achievement, and the sense of community that facilitates thinking well (RUSSELL, 2000). [9]

One way to gain greater understanding of reflexivity is to examine the process of research as it unfolds across the different stages of the research endeavor, including the tasks of formulating questions, defining methodological strategies, performing analyses, and disseminating results. Each of these stages is initiated and extended through the intersecting dialogic processes that create and shape the research event (LUCIUS-HOENE & DEPPERMANN, 2000; L. HAWES, 1994). According to LUCIUS-HOENE and DEPPERMANN (2000), the sense of collaboration begins in advance of the actual interview event. They suggest that as the researcher begins to imagine and later investigate the desired qualities and experiences of potential participants, (s)he begins to shape the questions and anticipate positive or problematic circumstances that could diminish or add to the success of the experience. As part of this process, the researcher enters into the fictional or theoretical steps that establish an initial communicative relationship with the participant. [10]

That communication is realized with an initial phone call or meeting that enlists the participant's assistance with the research study. The dialog has now moved from fiction to reality and henceforth evolves within the rules of everyday interactions. As the researcher and participant speculate about the potential for their discursive alliance, they establish tentative rules for interacting. Those rules will be further refined by the nature of the research event and by the researcher-participant communication. [11]

This paper will explore the dialogic interactions inherent in the various steps of the research process. In keeping with RICHARDSON'S recommendation (1995, as cited in WIESENFELD, 2000), we will both speak in descriptive terms and offer narratives that reflect the experiences of participants. Some of these narratives are drawn from our own work and some have been contributed by other researchers in conversations with us. These representations will be drawn from experiences of conducting team research that allow for an expanded view of dialog while encouraging an awareness of the critical thinking processes embedded in intellectual reflexivity. [12]

SILVERMAN (1993, as cited in RHODES, 2000) proposed that the research interview does not correspond to some external truth, but rather a way of creating one of many possible accounts that are reflexively linked to the interplay of discursive actions. He suggested that as the researcher(s) and respondent(s) learn and change through the dialog of the interview, they develop a consensual truth that enables the process. BAMBERG (1999) identified this joining as an interactive negotiation that ultimately defines a story's meaning. [13]

The concept of negotiation within the interview suggests that the process does not merely recount past events; rather, it constructs new stories out of the flow of information and interpretation of both participants. In this context, research results emerge from the interactive dialog that occurs between researcher and respondent as they position and shape the information ultimately captured in the text. In this way, the dialogic interplay enacted as part of the interview process serves to join and integrate the two independent voices into a seamless co-creation of a newly formed reality. This discursive view of research does not deny or diminish the concept of reality (SAMPSON, 1993) but offers an alternative explanation for how that reality is represented and understood. In viewing words as deeds, a discursive framework challenges the traditional understanding of research phenomena. Without searching for pre-existing truths, it considers the ways in which meaning is established through interactive dialog and recognizes the collaborative nature of the interview process. In acknowledging the power of dialog to shape and define meaning, it provides a framework to examine how knowledge is constructed and known (BAKHTIN, 1986). [14]

"Since research outcomes are produced through the creation of meanings that emerge from the interactive dialogs of interviews, the quality of that data will be influenced by the nature of the relationship between researcher and participant" (POPAY, ROGERS, & WILLIAMS, 1998 as cited in HALL & CALLERY, 2001, p.260; see also HAMMERSLEY 1987). The concept of relationality within the research process acknowledges the connectedness between researcher and participant and excludes any recognition of subject or object as constructed within the positivist paradigm. The understanding that develops out of the communicative process of the interview is a result of two human parties in conversation about meaning. Although these conversations recognize the individuality of participants, including the inequality of roles and power differentials, they have the potential to offer equality at a moral level that transcends the sense of separateness. TANNEN (1984, as cited in BRADLEY, 1995) suggested that the dialog of research may be enhanced and deepened by conversation that emerges from the joint production and coordinated interaction of the interview process. She proposed that the shared rhythms and mutual understandings embedded in the flow of conversation allow participants to experience a sense of satisfaction that goes beyond the pleasure of having one's message understood. She defines conversation as "a proof of connection to other people that provides a sense of coherence in the world" (p.373). [15]

The dialog of research must sustain an awareness of each participant's perspective. It should foster an I-and-Thou relationship (BUBER, 1970) that allows and promotes the humanity of both the researcher and the participant. In seeking a moral equality—in contrast to role inequality—the relationship invites both the researcher and the participant to grow, learn, and change through the research process. In this way, the elements of choice and possibility will appear in place of previous constraint and inevitability (L. HAWES, 1994), and a space will be created that allows each voice to be heard. [16]

In order to examine how these principles are enacted in the process of research endeavors, we will consider a typical analysis of a research project into a series of (presumably discrete) steps. Although it is both customary and convenient to apportion the research process in this way, we acknowledge that this form of analysis distracts us from what might be more aptly thought of as a seamless set of conversations with incoming and outgoing participants and with its own rhythm of ebbs and flows. The notion of stages provides a useful means for describing different aspects of the research processes. Nonetheless, these stages should be taken as heuristic devices rather than as absolute divisions between the various interacting processes involved in research. In our experience, any of the stages we are about to describe can appear and disappear throughout this seamless set of conversations that might be said to describe research processes more accurately. In fact, we will offer an example later in this paper in which the question formulation emerged in a new form during data analysis. [17]

The first (albeit admittedly artificially excised) step in the research process is the formulation of a research question. How, we might ask, can this step be understood as a dialog? To address this question, we must first recognize that the ability to ask good research questions is rooted as much in an ongoing attunement with the world as it is with any other factor. If we are listening well to the world, there is no lack of researchable questions. It is a matter of paying attention and noticing what is going on with people—i.e., listening to the dialogs of their lives: What are they doing that is interesting or troublesome or inspiring? What do they care about, wonder about, struggle with? [18]

One of us (GR), in the aftermath of an anti-gay election, overheard the pro-gay campaign spokesperson in a telephone conversation. This normally quiet and modulated heterosexual woman, who for months had been publicly speaking for a campaign on behalf of lesbian, gay, and bisexual people, suddenly raised her voice and purposefully said to the person on the other end of the line, "Listen, I know more about homophobia than most heterosexuals you'll ever know." I turned to look at this woman, who was tired from but educated through her efforts in an unsuccessful campaign. I silently answered her assertion: "I bet you do." And at that moment was born a research project focused on heterosexuals who take visible stands for gay rights. The idea for the project surely was born in the words of the spokesperson, in her dialog with the caller. It was just as surely born in my own curiosity about her words and my internal dialogs—with her and with myself. I noticed something I had never seen in this woman; I was struck by her words; I was curious about the whole picture; I wanted to know more. [19]

Good research questions spring from our "values, passions, and preoccupations" (MARECEK, FINE, & KIDDER, 1997, p.634). If we are engaged in the world and attuned to our own responses to the world, research questions arise with unexpected regularity and predictable spontaneity. Our own reflexivity is a necessary ingredient in generating questions. Even the language of "generating" questions seems problematic. It strikes us that we are no more generating them than we are hearing and responding to them—being open to and curious about what floats around us in the course of our ordinary days. [20]

Once we have decided on a research question, our next task is to take our curiosity into some sort of concrete interaction with the world, quite often with other people. However we are gathering information—be it textual, observational, interview, or any other manner of data—we are called upon to engage our entire selves in the dialogic interactions with our question, in the process of listening to the information and noticing our reactions to it. For those of us who were trained exclusively in positivist epistemologies, engaging our entire selves runs counter to the emphasis on objectivity and neutrality so central to our socialization as researchers (MERTENS, 1998). More recent training that includes an element of postpositivist analysis offers a greater possibility for recognizing the place of theory and values in research, but it does not fully prepare us to be moved in deep emotional (as well as intellectual) ways by our research endeavors. Reflexivity in domains having to do with emotions may represent a particular challenge, given training that often denied the presence or salience of emotional reactions. [21]

The listening process is at once challenging and invigorating. It is not hyperbole to say that we have, individually and together, felt our heart and breathing accelerate as we set about gathering information: What is this? What is going on here? How can I possibly make sense of that? Though very often pleasurable, it is not an altogether pleasant process; it is also disorienting and disturbing. We are thrown off-balance. It is necessary to watch ourselves carefully to see what our responses are and why these responses might be happening. Reflexivity at this point in the process is critical for alerting us to what allows us to see and to what inhibits our seeing (MICHALOWSKI, 1997). [22]

Information gathering may entail an infinite variety of challenges to researchers' abilities to stay alert to their own curiosities and responses. Our strong impression is that, when we are reflexive, other human participants join us in being reflective as well. We have been moved by moments of genuine insight offered by interview respondents. We have been surprised to find ourselves saying something we had not known we knew. We have been gratified by how interviews with multiple respondents gather force and become everyone's interview, everyone's responsibility, and everyone's prize. Looking back at some transcripts of these interviews, it is clear that there are times in the conversation when the roles of the interviewer and interviewee have become so blurred as to all but disappear. We have noticed occasions when the affect associated with the focus of the interview floods the room and all participants. [23]

One of us (GR) conducted group interviews with women who had taken a full-force self-defense class. It was common for participants to "discover" something about their experiences in and through their conversations with the other participants. It was similarly common for the participants to comment that the interview process was re-creating the joyous, connected, and empowering feelings that they had first experienced in the class. Even within the context of information gathering about very serious matters, we have often enjoyed the playfulness that SARBIN and KITSUSE (1994) regard as a regular feature of postmodern analysis. In interviews with other researchers who shared their experiences with information gathering, we have heard descriptions that are familiar. They have spoken of a sense of "profound connection" with colleague-researchers and with participants. They have described the momentum of focus groups as akin to "a snowball building on itself," and as a process in which "everybody's creativity feeds off everybody else's." [24]

In the absence of a reflexive stance, researchers might easily fail to notice the degree to which our respondents (and we) are changing before our eyes. Their (and our) engagement with the questions in an interview opens new possibilities for their (and our) understanding and insight. Their understandings of themselves may already have been altered at the moment we approach them to participate in the research process. We have often heard people second-guess our request for an interview, wondering if they really had something of interest to tell us. I was surprised by the number of people I contacted for the aforementioned study on heterosexual allies who had never considered their actions to be out of the ordinary or worthy of anyone's attention. I was heartened by the number of these heterosexually-identified activists who embraced the notion that there were others who felt and acted as they did. Our experiences with participants in this and other research ventures lead us to concur with BANYARD and MILLER'S (1998) observation that it is empowering for people to tell their stories; it is especially powerful to have their stories heard. In one study that explored the psychological effects of an anti-gay election on lesbians, gay men, and bisexuals, our coding team was stunned by the number of respondents to a long survey who took the added time to write notes of thanks. Some of these participants went on effectively to explain that they were grateful for the opportunity to describe their experiences or to do something that might help other gay people (RUSSELL, 2000). [25]

The gathering of information can be an enormously complex process that challenges our reflexivity on many levels. We have found it helpful to think of the information gathering process as the creation within the research moment of a new self: a single researcher and the text are actually a single researcher, the text, and the self-made-up-of-the-single-researcher-and-the-text. Two researchers conducting an interview with four respondents are two researchers, four respondents, and a self-made-up-of-two-researchers-and-four-respondents. If we consider the selfs (and we use this form of the plural with intent) created by the interaction of each dyad or triad within this group of six individuals, the picture becomes even more complicated. These new selfs represent the enactment by a novel constellation of people who are at this moment focused on a particular topic. Researchers' reflexivity must be attuned to themselves and their responses to participants, of course. Their reflexivity also needs to be attuned to the emergent selfs that have been created for the purpose and by the processes of gathering information. [26]

Reflexivity at this level invites us to turn our attention to all participants in the process. Echoing McNAMEE and GERGEN (1999) (who were speaking at a different level of self), we ask:

"Who is speaking and acting here, Who is listening, What voices are not being heard, What selves within are suffering, Why is this voice dominant and not some other, and How can we help these oppressed potentials into being? We can understand that the conflict, the anguish, the retribution, and so on are being played out by fractional impulses acquired from others and because, in Walt Whitman's terms, 'we contain multitudes,' we are invited to expand the retinue of guests at the table of responsibility" (p.13). [27]

If many voices compete for researchers' attention during the data gathering process, there are certainly no fewer voices striving to be heard during the process of information analysis. This is the case no matter what is the nature of the data with which researchers are engaged. In the words of MANNING and CULLUM-SWAN, "... all texts metaphorically speak with many voices" (1994, p.469). It is up to researchers to listen carefully to these voices and to listen especially carefully to the voices that are quietest and to those that may be absent. [28]

Reflexivity requires that we suspend our judgment, our propensity for foreclosed inquiry, and our enthusiasm for the early answers that usually seem to present themselves. We would do well to recall BINSWANGER'S (1967) caution against la rage de vouloir conclure (Tr: the faddish desire to come to a conclusion). The caution, borrowed from FLAUBERT, reminds us that we must overcome our passionate wish to draw conclusions. We have to sit with the information at hand long enough, and with enough openness, to understand not first what it says, but rather how it wants to talk with us. [29]

If we presume to know how research results will speak to us, we may find our analyses frustrating or fruitless. When I (GR) first began conducting qualitative research, I stumbled upon a methodology that seemed to work with the data set I was trying to understand. When I ventured into my second qualitative effort, I naively assumed that I would use the same methodology. It did not work. The information in the second study refused to be heard through the same methodology that had worked so well in the first study. Methods, it turns out, really are "the most unremarkable aspect of interpretive work" (SCHWANDT, 1994, p.119). We are inclined to draw a parallel with clinical work. There is no substitute for exquisite care to the relationship in psychotherapy no matter how reliable and valid one's "method" of therapy is. Similarly, no method of research inquiry can substitute for the need for researchers to engage in a reflexive relationship with data. [30]

The researchers' relationship with the data during analysis (the word seemed so out of place here) requires delicacy and perseverance. We have found that working with qualitative data in a group context allows us to enhance our reflexivity in a number of ways. The group context offers researchers what has been referred to as a holding environment—a stable sense of support that allows us to do things we might not be able to do if left to our individual (and individualized) devices. Research within a group context offers the "antithesis of the insular experience," as one student put it. It gives us, as another student suggested, the opportunity "to feel connected not only to the work but to the people [we are] doing it with." In a very real sense, research in a group context may allow us to keep our equilibrium even when we are faced with the "vertigo" (MARACEK, FINE, & KIDDER, 1997, p.638) that sometimes accompanies engagement with qualitative research. [31]

If there are many voices represented in our data, then we can think of many ears represented in the team that is interacting with this information. It is, in some ways, a humbling experience to work with data analysis in a group context. When a team is working well, each member is routinely reminded of what she does not hear, what he emphasizes without question, what she ignores without knowing it, and what impact this morning's chance meeting has on his understanding of the data. When a team is working well, all members attend to their own reflexivity and simultaneously serve as checks on one another's reflexivity and the reflexivity of the self-of-the-team. [32]

One of us (GR) had the experience of working on a well-functioning team, one of whose members ended a significant relationship during our data analysis. For a time, this team member frequently heard respondents' comments as reflecting a data code denoting sadness. It became clear that he was perceiving sadness far more often than were the other members of the team. Moreover, when he tried to teach us how he was hearing this sadness, only rarely were we able to discern the sadness he had seen. As a group, we considered the possibility that the recent ending of his relationship might be influencing the frequency with which he heard sadness in the data. While we all came to understand that this might be the case, none of us—including this team member—was willing to automatically disqualify his perceptions about sadness. He was indeed sensitive to sadness in a way that could impose it on the data; but he was also sensitive in a way that could ferret out hints of sadness that others of us might miss. At the same time, his experience with the dissolution of his relationship helped sensitize him—and therefore all of us and the team as a whole—to relational dimensions in the data. This team member's experience allowed us to interact with the data in ways that we might not have interacted had he been absent (or had his relationship remained intact). It is in this way that conducting data analysis in a team context can be empowering. It allows us to see other perspectives, to see more than we otherwise would have. It allows each of us to relax a bit, knowing none of us is solely responsible for understanding the data. Instead, each of us can rely on—and must nurture—each of the other researchers as well as the self-made-up-of-all-the-researchers-on-the-team. Each self requires attention; each is responsible; each depends on the self's own reflexivity and everyone else's reflexivity. [33]

We want to mention cases in which research is focused on a particular social group that is marginalized and in which the research team includes members of the target/out/marginalized group and also members of the non-target/in/dominant group. It is easy to say that having members of both groups on the research team has the advantage of bringing to bear different perspectives on the data (BARTUNEK, 1996). It is perhaps not self-evident to suggest that it would be problematic to privilege members of either one group or the other. It might be tempting to privilege the non-target group as more "objective" or the target group as more knowledgeable. However, if reflexivity is our goal, all members of the team need to play the stranger (MORAWSKI, 1991) in relation to the data. All need to approach the information with a reflexive openness that acknowledges that their perspectives hold possibilities for seeing the meanings embedded in the data, possibilities for obscuring those meanings, and other possibilities not anticipated. Reflexivity requires that group members move beyond the confines of identity politics (GERGEN, 2000; SAMPSON, 1993) and be open to what each—whether in a marginalized group or not—can see and what the self-made-up-of-all-team-members can see as they interact with the data. [34]

We have observed differing perspectives simultaneously brought to data by team members on different sides of demographic/identity politics divides (see especially RUSSELL, 2000). We have seen the data well served when there was no hierarchy based on identity categories (see also HARAWAY, 1988). This lesson was dramatically brought home when one of us who was working on a research team with two other colleagues. The project involved interviews with a number of youths who had started a Gay-Straight Alliance in Salt Lake City, Utah. Two of us, both of whom identified as lesbian, had conducted all the interviews. The third member of the team, David, was a heterosexually-identified man who had joined the team to explore with us the information gathered. As we read the interview transcripts, David raised question about the language many teens employed in their discussions of stress and suicide. At first, the two lesbian-identified members of the team dismissed his question. We both knew about the problem of gay youth suicide; we had taught and consulted about it. We were, in fact, quite steeped in the problem of gay youth suicide. [35]

It was only through David's perspective as "outsider" and his persistence in raising the question that we eventually came to share his question. Indeed, as a team we began asking questions about the data that probably could only be asked if one were not steeped in the issue. David's line of questioning pointed the way to the beginning of an understanding about what gay, bisexual, and lesbian youths may have learned from those of us who were so steeped in the problem of gay youth suicide. We came to realize that we had unwittingly given these youths a very dangerous script about their lives (RUSSELL, BOHAN, & LILLY, 2000). [36]

This experience serves to illustrate the point we made earlier: it is problematic to speak of stages of research in monolithic terms. In this instance, a moment of question formulation occurred during the data analysis stage. We proceeded to inquire about this question from that point onward. Research, like any conversation, is not bounded by some linear sequence. It is iterative, characterized by false starts and new beginnings. As such, it requires an enormous degree of reflexivity. In the absence of reflexivity, we may be tempted to accept the apparent linearity, thereby obscuring all sorts of unexpected possibilities. [37]

If research represents a series of extended conversations, then it is incumbent on researchers to continue these conversations, to do something more than compile information as fact bricks stored in libraries (BAKAN, 1977). It is incumbent on researchers to be as reflexive about what we do with the results of our research as we have been in all the prior conversations. For many researchers, the obvious task once information analysis is completed is to compose a written version of the research results for dissemination to other interested parties. Inevitably, this process of writing is "an interpretive, personnel, and political act" (DENZIN & LINCOLN, 1994, p.479); it requires that researchers be mindful of themselves, the participants, their audience, and the uses to which their research might be put. This includes the need for academic researchers to actively take into consideration how issues such as tenure, promotion, and professional visibility have affected every phase of their research. It is, of course, critical that researchers consider the effects that their research will have on the issue area which and/or the people who have been the focus of their research. In disseminating research results, we often are speaking from a very privileged position; it is important that we spend our privilege well (PHARR, 1988). [38]

We spoke with a group of students who had conducted research on narratives about divorce that had been expressed during interviews with divorced women. One of the researchers told us, "I was really sensitive that [the respondents] feel comfortable with the way that we had presented and held that information that they had entrusted to us." Her comment reflects a fitting sense of responsibility in regard to the women who participated in the study; it is no less a reflection of a fitting sense of accountability to oneself as a researcher. [39]

Dissemination of research results all too often begins and ends with presentations at professional meetings and publications in academic journals and books. Yet, powerful things sometimes occur when researchers step outside the usual venues for dissemination. The aforementioned group of students who examined divorced women's narratives decided to use a refrigerator door—their model of domesticity—as the background for a school-based poster session for student research projects. The researchers, in a stroke of creativity that seemed to capture and also enact their findings, arranged their results on the refrigerator door with magnets. The researchers reported the pleasure that they took in their creativity and in their sense of the consistency between their medium and their message. One of the researchers reported experiencing a different kind of pleasure when one of the audience members confided in her that she was about to go through a divorce and that she had learned a good deal from the poster. [40]

In the Gay-Straight Alliance study mentioned earlier, our research group knew that it was important to "return" the findings about suicidal scripting to our community. We invited all of our participants to a gathering at which we discussed our research findings in general and the suicide scripting in particular. It was gratifying to watch participants make sense of this information. They understood it; they worked with it; eventually many of them vowed to reject it through their own long, productive lives. We have since arranged to speak with gay youth groups, in efforts to alert them to the dangers of this script. We have been "giving away" these findings whenever possible. It is not without irony that our nascent understanding of the suicidal scripting of gay, lesbian, and bisexual youth came about in a study that did not begin with this topic as a focus of our interest. Reflexivity—or perhaps it is luck—may often warrant credit when the most important findings from a study were not at all anticipated when the project began. [41]

In a final note about the dissemination of research findings, we invite readers to consider the many ways that research results can be returned to communities without requiring that participants make any trips to libraries or conferences. Among the media we have used, in addition to the more mundane outlets such as community talks and newsletters, are an oratorio, a documentary for public television, and signs on city buses (see RUSSELL & BOHAN, 1990 and RUSSELL, 2000 for further details). We have been humbled to find that the ideas for innovative methods of disseminating findings do not often flow naturally from our professional training; it has been far more helpful to listen to suggestions from others who are less steeped in professional practices than are we. Reflexivity is key here, once again. We must be open to recognizing how our own position both privileges and limits us, to hearing the voices of others clearly enough that we can speak to them in the extraordinary ways that resonate with their (if not necessarily our) experiences. In short, we must find ways to speak to them, not simply to ourselves. [42]

A commitment to reflexivity suggests that we continue to look at the impact of our research at all points during the research process—including its impact on us. Certainly, we begin research projects with (among other wishes) the desire to learn something new. It has been our experience that we ourselves have typically been transformed in and through the research efforts (BRYDON-MILLER, 1997). In a way that mirrors FREIRE's (1993) observations of and lessons about education: we are educating and being educated; we are learning about ourselves as well as others. We are changed by many aspects of the research process: through engaging in real conversations, through what we learn in the course of listening well, through participation in a process that allows new creations to occur, and through our own reflexivity. As researchers, we come away with new understandings, the origins of which are not entirely clear to us. Our very participation in the research endeavor changes us. In the words of one of our student-colleagues: "I believe that knowledge can be demonstrated through discourses but that discourse changes the quality of knowing and that because the knower is the knowing, the knower is also inherently changed" (LEVINE, 2002, p.6). This knowledge can occur in a variety of areas, from the extraordinary to the ordinary ("... in areas that I live with but don't necessarily think about," as another student described it). [43]

Just as we rely on reflexivity to carry out good research, conducting good research tends to improve our reflexivity. It enhances our ability to stay engaged with our own reactions and the reactions of others. It insists that we learn more about our personal and intellectual strengths and limitations. It invites us to confront feelings and conflicts that we might otherwise avoid—aspects of experience that traditional training has, in fact, encouraged us to disavow. It insists that we trust ephemeral processes and promises and gather at the tree of no-knowledge to eat until we are full—not of knowledge (of course) but of the wonder of the ever-changing stream of conversations that occur within and around us. [44]

Just as reflexivity is important to how we think about ourselves as researchers, to our participants, and to our understandings, it is also important to how we regard the political world. The traditional focus of much of the training in research disciplines has denied the presence or the relevance of the political dimensions of our work. The call for neutral and value free research approaches and agendas denies the possibility of our being reflexive about the political roots and significance of our work. Of course, that call itself represents a profoundly political stance, as well as one that discourages a broad-based reflexivity (BRYDON-MILLER & TOLMAN, 1997). Working within a postmodern perspective, in contrast, invites us to—indeed demands that we—maintain an ongoing awareness of the political nature and implications of all our actions including those carried out under the rubric of research. We join with BRYDON-MILLER'S (1997) assertion: "Embracing the political nature of any research process allows us to act in a more direct and open manner in addressing social issues" (p.660). [45]

Perhaps one of the closest political influences on our work—the choice of topics, methods, and media for disseminating our findings—is constituted in the environments of our disciplines and our (usually educational) organizations (BAKAN, 1977; CUSHMAN, 1995). The very proximity of these influences demands a particularly sensitive reflexivity. Moreover, the epistemologies of our disciplines tend to obscure other possible ways of seeing. Working with colleagues from other disciplines helps us to see that which we have been trained to take as givens. The values and reward structure of our academic (and other) organizational homes require careful attention. We are quite literally rewarded (with tenure, with promotions, with the esteem of colleagues) for pursuing some questions and not others and for seeing with some lenses rather than with others. The more we can bring reflexivity to bear on these matters, the more aware we can be of the multiple (and often unstated) influences on our work. It is this same reflexivity, in its ability to discern the pressures to study this matter in that way that allows us the occasional possibility of rising above these very influences. [46]

Our reflexive focus need not stop with the immediate environments in which our work occurs. As researchers, we take seriously the obligation and we enjoy the privilege of inquiring as to the broader political assumptions underlying our work and the wider political ramifications of our findings. At every step in the research project, we ask ourselves the critical question: "Who benefits?" (BOHAN & RUSSELL, 1999). This question and our answers to it represent our best possibilities for holding ourselves accountable to ourselves and our participants. By including political understandings as a natural and inevitable part of our research inquiry, we close the gap between the personal and the political, between the knower and the known, and between researchers and those whom we once thought of as subjects and now understand to be our co-creators. [47]

We would like to acknowledge the contributions of the following individuals who, through their work or their conversations with us, contributed to our understandings of reflexivity. They are: Patricia AMBROSIO, Lou BARDACH, Anne CLARK, Jim DAVIS-ROSENTHAL, Sandy DIXON, Patricia DUFFY, Ruth FOLCHMAN, Kari FRASER, Pam GOSS-POWER, Maria GUGLIELMINO, Stacy HIGGINS, Barbara LANDON, Mayday LEVINE, David LILLY, Robin LOPEZ, Cay McDERMOTT-COFFIN, Jason MIHALKO, Sylvie NAAR, Rob PERL, Autumn PORUBSKY, Sean RILEY, Tammy SALTZMAN, Kristine SANDS, and Susie STEVENS. We would like to extend particular appreciation to Janis BOHAN for her extensive contributions to this paper and to the work on which it is based.

Ahern, Kathryn (1999). Ten tips for reflexive bracketing. Qualitative Health Research, 9(3), 407-411.

Bakan, David (1977). Political factors in the development of American psychology. Annals of the New York Academy of Sciences, 291, 222-232.

Bakhtin, Mikhail M. (1986). Speech genres and other late essays. Austin, TX: The University of Texas.

Bamberg, Michael (1999). Is there anything behind discourse? Narrative and the local accomplishment of identities. Proceedings of the International Society for Theoretical Psychology, Germany, 7, 220-227.

Banyard, Victoria L. & Miller, Kenneth E (1998). The powerful potential of qualitative research for community psychology. American Journal of Community Psychology, 26, 485-505.

Bartunek, Jean M. & Louis, Meryl R. (1996). Insider/outsider team research. Thousand Oaks, CA: Sage.

Barry, Christine A.; Britten, Nicky; Barber, Nick; Bradley, Colin & Stevenson, Fiona (1999). Using reflexivity to optimize teamwork in qualitative research. Qualitative Health Research, 9(1), 26-44.

Binswanger, Ludwig (1967). The existential analysis school of thought. In Rollo May (Ed.), Existence (pp.191-213). New York: Clarion.

Bohan, Janis S. & Russell, Glenda M. (1999). Conceptual frameworks. In Janis S. Bohan & Glenda M. Russell, Conversations about psychology and sexual orientation (pp.11-30). New York: New York University Press.

Bradley, Lewis (1995). Psychotherapeutic discourse analysis. American Journal of Psychotherapy, 49(3), 371-382.

Breuer, Franz (2000, June). Qualitative methods in the study of biographies, interactions and everyday life contexts: The development of a research style. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 1(2), Art. 3. Available at: http://www.qualitative-research.net/fqs-texte/2-00/2-00breuer-e.htm.

Brydon-Miller, Mary & Tolman, Deborah L. (1997). Engaging the process of transformation. Journal of Social Issues, 53, 803-810.

Buber, Martin (1970). I and thou (W. Kaufmann, Trans.). New York: Simon & Schuster. (Original work published 1937)

Cheek, Julianne (1999). Influencing practice or simply esoteric? Researching health care using postmodern approaches. Qualitative Health Research, 9(3), 383-392.

Chia, Robert (1995). From modern to postmodern organizational analysis. Organization Studies, 16(4), 580-591.

Cushman, Philip (1995). Constructing the self, constructing America: A cultural history of psychotherapy. Boston: Addison-Wesley.

Denzin, Norman K. & Lincoln, Yvonna S. (1994). The art of interpretation, evaluation, and presentation. In Norman Denzin & Yvonna Lincoln (Eds.), Handbook of qualitative research (pp.479-483). Thousand Oaks, CA: Sage.

Erickson, Ken C. & Stull, Donald D. (1998). Doing team ethnography: Warnings and advice. Qualitative Research Methods Series, 42. Thousand Oaks, CA: Sage.

Gergen, Kenneth J. (2000). From identity to relational politics. In Lois Holzman & John Morss (Eds.), Postmodern psychologies, societal practice, and political life (pp.130-150). New York: Routledge.

Gergen, Kenneth J. & Gergen, Mary M. (1991). Toward reflexive methodologies. In Frederick Steier (Ed.), Research and reflexivity (pp.76-95). London: Sage.

Hall, Wendy A. & Callery, Peter (2001). Enhancing the rigor of grounded theory: Incorporating reflexivity and relationality. Qualitative Health Research, 11(2), 257-272.

Haraway, Donna (1988). Situated knowledges: The science question in feminism and the privilege of partial perspective. Feminist Studies, 14, 575-599.

Hatch, Mary Jo (1996). The role of the researcher. Journal of Management Inquiry, 5(4), 359-375.

Hawes, Leonard (1994). Revisiting reflexivity. Western Journal of Communication, 58(1), 5-10.

Hawes, Susan (1998). Positioning a dialogic reflexivity in the practice of feminist supervision. In Betty Bayer & John Shotter, (Eds.), Reconstructing the psychological subject: Bodies, practices and technologies (pp.94-110). Thousand Oaks, CA: Sage.

Larson, Carl E. & LaFasto, Frank M. (1989). Teamwork: What must go right/what can go wrong. Newbury Park, CA: Sage.

Levine, Mayday (2002). Qualitative inquiry and evaluation research. Unpublished manuscript. Antioch New England Graduate School.

Lucius-Hoene, Gabriele & Deppermann, Arnulf (2000). Narrative identity empiricized: A dialogical and positioning approach to autobiographical research interviews. Narrative Inquiry, 10(1), 199-222.

Manning, Peter K. & Cullum-Swan, Betsy (1994). Narrative, content, and semiotic analysis. In Norman Denzin & Yvonna Lincoln (Eds.), Handbook of qualitative research (pp.463-477). Thousand Oaks, CA: Sage.

Marecek, Jeanne; Fine, Michelle & Kidder, Louise (1997). Working between worlds: Qualitative methods and social psychology. Journal of Social Issues, 53, 631-644.

Mays, Nicholas & Pope, Catherine (1995). Rigour and qualitative research. British Medical Journal, 311, 109-115.

McNamee, Sheila & Gergen, Kenneth J. (1999). Relational responsibility: Resources for sustainable dialog. Thousand Oaks, CA: Sage.

Mertens, Donna M. (1998). Research methods in education and psychology: Integrating diversity with quantitative and qualitative approaches. Thousand Oaks, CA: Sage.

Michalowski, Raymond J. (1997). Ethnography and anxiety: Fieldwork and reflexivity in the vortex of U.S.-Cuban relations. In Rosanna Hertz (Ed.), Reflexivity and voice (pp.45-69). Thousand Oaks, CA: Sage.

Morawski, Jill G. (1991, Aug.). Is ours a history of selves or discourses? Social psychology confronts the postmodern condition. Invited address to Division 8, Personality and Social Psychology, American Psychological Association.

Payne, Stephen L., (2000). Challenges for research ethics and moral knowledge construction in the applied social sciences. Journal of Business Ethics, 26(4), 307-318.

Rhodes, Carl (2000). Ghostwriting research: Positioning the researcher in the interview text. Qualitative Inquiry, 6, 511-526.

Russell, Glenda M. (2000). Voted out: The psychological consequences of anti-gay politics. New York: New York University Press.

Russell, Glenda M. & Bohan, Janis S. (1999). Hearing voices: The uses of research and the politics of change. Psychology of Women Quarterly, 23, 403-418.

Russell, Glenda M.; Bohan, Janis S. & Lilly, David (2000). Queer youth: Old stories, new stories. In Sonya Jones (Ed.), A sea of stories: The shaping power of narrative in gay and lesbian cultures (pp.69-92). New York: Harrington Park Press.

Sampson, Edward E. (1993). Identity politics: Challenges to psychology's understandings. American Psychologist, 48, 1219-1230.

Sarbin, Theodore R. & Kitsuse, John I. (1994). A prologue to constructing the social. In Theodore Sarbin & John Kitsuse (Eds.), Constructing the social (pp.1-18). Thousand Oaks, CA: Sage.

Schwandt, Thomas A. (1994). Constructivist, interpretist approaches to human inquiry. In Norman Denzin & Yvonna Lincoln (Eds.), Handbook of qualitative research (pp.118-137). Thousand Oaks, CA: Sage.

Strathern, Marilyn (1991). Partial connections. Lanham, MD: Rowman and Littlefield.

Wheatley, Michael (1992). Leadership and the new science. San Francisco: Berrett-Koehler.

Wiesenfeld, Esther (2000, June). Between prescription and action: The gap between the theory and the practice of qualitative inquiries. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 1(2), Art. 30. Available at: http://www.qualitative-research.net/fqs-texte/2-00/2-00wiesenfeld-e.htm.

Glenda M. RUSSELL is the Senior Research Associate at the Institute for Gay and Lesbian Strategic Studies. She is a psychologist and activist.

Nancy H. KELLY is a doctoral candidate in clinical psychology at Antioch New England Graduate School. She is currently completing a dissertation focused on the role of the researcher in the research process.

Russell, Glenda M. & Kelly, Nancy H. (2002). Research as Interacting Dialogic Processes: Implications for Reflexivity [47 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 3(3), Art. 18, http://nbn-resolving.de/urn:nbn:de:0114-fqs0203181.

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

ISSN 1438-5627

Creative Commons Attribution 4.0 International License