Using A Story-Building Approach To Research Comprehensive Community Initiatives
Lori Messinger

The Qualitative Report, Volume 6, Number 1 March, 2001


This article is the story of the author's experiences as a researcher-storyteller. She delineates the process used to "build a story" of the planning and development of a rural comprehensive community initiative. In a critically reflective look at this approach to community research, the author describes how she selected the research topic, defined the purposes of the research, selected the methods used in data collection and analysis, and approached the construction process of building the story through literary elements such as plot, scene, and voice. The strengths and pitfalls of this approach and suggestions for future research are discussed.


Social work researchers studying comprehensive community initiatives (CCIs) are in many ways professional storytellers, telling true-life tales of complex social programs. Regardless of whether we use qualitative, quantitative, or mixed methodologies, the purpose of the research design is to build a coherent and thorough story about the creation and implementation of a CCI, describing how it addressed formidable social problems faced by people living in a specific community or neighborhood (Kubisch, Weiss, Schorr, & Connell, 1995). When the research is complete, we write "story-reports" presenting findings about the multiple stakeholders who planned the intervention, the diverse strategies that the project used to effect community change, and the impact of this project in addressing the various aspects of the targeted social problems, while offering our analyses regarding the initiative's effectiveness (e.g., Annie E. Casey Foundation, 1995; Chaskin, 1992; Kirk & Messinger, 1996; Medoff & Sklar, 1994). In these articles and reports, we create ourselves as narrators, revealing our own perspectives, experiences, and standpoints in the kind of voice and presentation style we employ (Richardson, 1994).

Few community researchers overtly approach their research as a story-building process. Reports focus only on the "facts" of the programs' creation, implementation, and evaluation, and often fail to analyze sufficiently the context in which these programs were created. As a result, readers are able neither to identify why programs performed as they did, nor to decide whether another community could successfully adopt these approaches. Typically, if a program's context is addressed, as in Chaskin's (1992) analysis of the Ford Foundation's Neighborhood and Family Initiative, it is presented as a "given," a history of facts and events that does not begin to describe the multiple lived experiences of a community's residents.

Medoff and Sklar's (1994) book, Streets of Hope, is an exception to the rule, offering one of the best examples of story-building research on a comprehensive community initiative. Holly Sklar, a newspaper reporter, clearly wanted to "get the story" and provide insights into the planning and implementation of the Dudley Street initiative in Massachusetts. Moving beyond a strictly journalistic approach (Bogdan & Biklen, 1992), she and community activist Peter Medoff used a variety of qualitative and historical research methods to investigate the history of the community, the previous relationships between the planning participants and the other members of the community, and the culture of the Dudley Street neighborhood. This combination of a journalistic approach with traditional social science methods produced an outstanding book-length analysis of the intervention in context. Yet, these authors do not fully discuss their approach to this research project, nor do they critically reflect on their research methods.

This is my story of intentionally undertaking the role of researcher-storyteller in my dissertation research on the planning and development of Warren Family Institute (WFI), a comprehensive community intervention undertaken in rural Warren County, North Carolina. I describe how I came to this topic, defined the purposes of the research, selected the methods used in data collection and analysis, and approached the construction process of building the story of the Institute's planning. I tell this story so that other researchers might learn from my reflections on the strengths, limitations, and obstacles presented by this story-building approach to community research. I identify the ways in which my choices of data gathering and data analysis techniques helped me to create a richer story of the program's development, one that takes into account the current and historical context of the county, the diverse experiences of planning participants, and my own role in building this story. Challenges and limitations to this research project were many. I review them as they emerged and offer suggestions that might help others who undertake this research approach.

Choosing a Topic: My Introduction to WFI

I came to know the Warren Family Institute through my work on the team evaluating it and four other North Carolina community initiatives funded by Z. Smith Reynolds Foundation. The Foundation hired researchers from the University of North Carolina School of Social Work to provide technical assistance to the programs as they defined their visions and developed goals and objectives for evaluation purposes. One year after the evaluation started, I joined the evaluation team as the qualitative researcher. As a result of joining the project late, I initially knew little about the planning processes of the different programs or their internal dynamics.

I learned more about the established programs through interviews with program staff, participants, and members of local agencies who interacted with the programs. The Warren Family Institute was a comprehensive community initiative that attacked the problems of poverty in a poor, rural county using a three-pronged approach. First, case managers partnered with low-income family leaders to identify the families' strengths, needs, and goals, and then worked with the families to help them meet these goals. Second, program staff worked with staff from other social service, health, and mental health agencies to develop joint projects that would better utilize their combined resources to meet the needs of poor families. Third, program staff worked on a macro-level with other agencies, promoting the establishment of a family-centered service model, advocating on behalf of low-income families, and offering leadership development training to low-income family leaders, with the goal of establishing better relationships between the local service organizations and low-income families.

I was impressed by the accomplishments of the Institute, outlined in the final evaluation report. Innovative joint projects were established using local resources, addressing such issues as youth recreation and school-age day care, transportation, home repair, and health care. Low-income residents emerged as leaders through their participation as program volunteers, staff members, and members of the Institute's Board of Directors. Local low-income residents also noted some improvements in the delivery of services. I decided to study how this program was created, with the intention of better understanding the planning process that produced this successful program.

In my initial interactions with the Institute, I became aware of several local issues facing the program. The most immediate had to do with race and class. Few white residents participated in any aspect of the program: they did not serve on the staff or the board, nor were many poor white families willing to engage in the case management or leadership development programs. There also was hostility toward the program on the part of some white, middle-class community members, as evidenced by caustic letters to the local newspaper questioning the viability and purpose of the Institute. I wondered at the root of this separatism and hostility, and questioned its impact on the planning and implementation of this initiative. I believed that I would need to investigate the culture and history of the County in order to understand the dynamics underlying the Institute's planning. To develop an accurate account of the Institute's planning, one that could be analyzed for insights and lessons that could be helpful to other communities planning these interventions, I believed that I needed to use a multi-dimensional, story-building methodology.

Theoretical Framework: My Backstory

My philosophical perspective, informed by critical, postmodern, and feminist theories, greatly shaped my research. Critical theory calls for the investigation of mainstream interpretations of politics, social relations, and work, analyzing how those in positions of authority use their power to oppress people of different class positions (LeCompte & Preissle, 1993). Culture is a focus of research as it shapes and structures society and social interactions. I am particularly drawn to the work of Habermas (1984, 1987), who calls for a combination of rational-empirical scientific methods, historical-hermeneutic approaches, and critically oriented methods in social research (Albrecht & Lim, 1986, pp.122-123). Habermas argues that, to truly understand any social process, traditional "objective" data must be combined with data on people's perceptions and a critical analysis of all of the data. I used this approach in structuring my data collection, employing a variety of data collection methods to capture all three kinds of data.

Postmodern research builds on the critical perspective, stressing the importance of analyzing the historical deployment of power through social institutions. Postmodernists look more closely at the language processes that reinforce systems of oppression through the privileging of certain interpretations in a linguistic community (Oleson, 1994). In addition to investigating the structures and experiences of the mainstream individuals and institutions, postmodern researchers investigate the ways that oppressed people interrupt and subvert their oppression. The postmodern influences on my research design are clearest in the following areas: the selection of a diverse group of interview subjects, the focus on subjects' perceptions about the historical and current institutions of the local culture, and the investigation of discursive communities within Warren County.

Feminist critical theory calls for a research design which: (1) allows the subjects of research to participate in the project, (2) includes the voices of oppressed peoples, and (3) requires the researcher to be self-reflexive throughout the research, analysis, and writing processes. My research project employed a feminist critical research design, striving to be "multivocal, collaborative, naturalistically grounded in the worlds of lived experience, and organized by a critical, interpretive theory" (Denzin, 1994, p.509).

My approach to research as a narrative-building process was also informed by the work of Walter Fisher (1987, 1989, 1994). Fisher suggests that stories are "constitutive of people, community, and the world"(1989, p. 57), and that the world should be "interpreted and assessed using the narrative paradigm" (1987, p. xi). Narratives, Fisher (1987) asserts, should be assessed using two criteria: coherence and fidelity. I used these two criteria to shape and evaluate the story I created about the planning of Warren Family Institute.

Research Methods: Tools for Finding the Story

In this research, I used a case study design and qualitative research methods. Stake (1994) states that case study is not a methodological choice, but a choice of object to be studied (p. 236). Researchers who use qualitative, ethnographic methods often employ case study. My decision to use qualitative methods was based on their ability to help me to document phenomena of interest and identify important variables shaping the process (Marshall & Rossman, 1995; Stake, 1994). An exploratory case study using qualitative methods was an appropriate approach to learn about the planning process in Warren County and discover important contextual factors affecting this county's planning process.

Using Habermas' categories of knowledge, I designed my study to use rational-empirical, historical-hermeneutic, and critical methods. Rational-empirical methods uncovered the "facts" of the planning endeavor, such as the economic status of the county, demographic information on county residents, events making up the planning process, services offered by the Institute, and the composition of local government and services agencies' staff. I gathered this information from planning documents, program documents, prior studies of Warren County, and census data.

Historical-hermeneutic approaches allowed me to interpret subjective meanings attributed to these "realities" by the different stakeholders: residents, service recipients, local government officials, agency staff and administration, and funders. I conducted interviews with ten planning participants and 11 community residents, along with six Institute staff. Interviews with planning participants and community residents were semi-structured, focusing on perceptions of both the planning process and experiences in the community. These data were enhanced by (1) informal conversations with Foundation staff and service recipients and (2) ethnographic data on the geographical and social aspects of the community collected during several field visits to Warren County. These visits ranged in length from two weeks to one month.

I adopted a critical perspective in interviewing, examining the power relations underlying these objective and subjective data. During interviews with planning participants and Institute staff, I asked these subjects (in plain language) to offer their own analyses about the framing of social issues, the ways in which different residents identified themselves, the role and diversity of cultural norms in this rural county, and how they explained the distribution of power as reinforced by institutional and interpersonal relationships. I investigated historical and current events as to their impact on community relations and the distribution of power in Warren County.

Learning about the local culture and history was particularly important for me, as an outsider to Warren County, rural areas, and the South. Southern rural culture seemed very different from my own background growing up in a New Jersey suburb of Philadelphia. Feminist theorists Fonow and Cook (1991) and Lather (1993) call for feminist researchers to explore our own biases and pay attention to the emotional components of our research. To help me identify moments of bias or emotion, I kept a computerized journal in which I recorded my feelings, concerns, surprises, and preliminary understandings.

Reviewing my journal entries, I saw the emotions that surrounded how I was negotiating my identity in Warren County. In the first few days of my fieldwork, I wrote:

It is odd to be in Warren County. I have visions of being run out of town on a rail, of having my apartment vandalized with spray-painted "Jew dyke" on my door. My Northern paranoia, irrational fears based on limited knowledge of the rural South.
It scares me to be out here alone. I do not trust these people--too many stories, I guess. Even though I am not out to people...though I have not shaved my legs yet. I just don't want to--so I will avoid it until I feel that it impedes my interviewing.

Later, after a troubling interview, I wrote:

She [the interviewee] seemed like a fine woman, too, in some ways. I mean, she seemed really friendly and open. I could imagine her having me over to dinner. But at the same time, here I am...I'm like "Oh, no, I'm not married." They asked if I'm married. "No, I'm not married." And I'm acting... I'm hiding all these things.
I'm glad I don't live here. I couldn't live here. I could live here and be myself, the interviewer, but I couldn't be myself, the person. You know? [As] the interviewer, I am so careful what I disclose. Though, I have disclosed to a lot of people that I am Jewish. Well, that is not true. Some of them. Part of them. Not all. But more than I thought I would. Oh, my goodness, it is an experience. It is an experience.

Clearly, I was cautious about being my whole self (a Northern, Jewish lesbian), while at the same time, I fiercely clung to those physical and unspoken markers (unshaven legs, Hebrew ring) that make me who I am.

My journal and notes also revealed moments of bias rooted in my current and past experiences. As a resident of Durham, NC, a large urban area with a wide range of housing options, I was surprised by the living conditions of some professionals in Warren County. Several of the professionals I interviewed lived in trailers or dilapidated houses. "It never occurred to me," I wrote, "that the housing problem was affecting them as well." In these notes, and in commentary on these notes, I critically assessed my entries, looking at the ways that my own background, experiences, and emotions were shading and shaping the collection and analysis of the data.


I collected data in a series of waves spanning two years; even with all of that time, there are still significant limitations to the data. Some of these limitations were situational and can be avoided by others undertaking such research in the future. Since there was no official file of planning documents in any of the planning agencies, I could only analyze those documents individual planning participants had kept. Researchers who plan to analyze planning documents might try to establish their own collection of memos, proposal drafts, and surveys throughout the planning process.

Other limitations are endemic to the methods used. There is no way to represent all aspects of the community in interview samples. Such a claim to representativeness would be impossible, because there are countless differences among community members. A researcher would have to interview all members of the community to get a true "sample." The categories I used to select interviewees (gender, race, age, native/non-native, economic status, area of residence) were not exhaustive categories. Another researcher might identify other categories of importance, searching for a distribution across levels of education, types of education, church affiliations, or levels of community or political activity. More importantly, I collected data from those who agreed to participate (Yow, 1994). As a result, only a limited number of perspectives truly inform the final analysis. It is important to remember these limitations during the data analysis/story-building stage.

Data Analysis: Building the Story

The data analysis was an iterative process in which I constantly created and revised codes. Data collected during the (earlier) evaluation of Warren Family Institute shaped the preparation for my study of the program's planning. Conducting and transcribing each round of interviews shaped my areas of interest, created possible leads, introduced new subjects, and gave rise to new questions that informed another round of data collection. These insights were further informed by demographic and historical data gathered throughout the research process.

I used ATLAS/ti qualitative data analysis software to code all of the data. Data collected during the second round was transcribed and analyzed concurrently with the formal coding and analysis of the earlier data. Themes, patterns, and categories emerged as I transcribed the interviews, finally taking shape in the stories of the planning process and the county itself.

Two historic events seemed to have an identifiable connection to the planning of Warren Family Institute. The first significant event I identified was the racial integration of public facilities in the late 1960s. The second event, the planning and ultimate demise of Soul City, followed closely on the heels of the first event in the 1970s. Soul City was to be a new city, supported by federal "New Communities" funds and planned by a coalition of African American civil rights activists, state planners, and state University personnel.

The relationships between these two historical events and the Institute emerged clearly after analysis. Several of the professionals involved in the Institute's planning had been leaders in the integration struggle and the development of Soul City. As a result, strategies employed by the Institute's planning participants were similar to those they had used in the past. Race and class issues also emerged as important factors in integration, Soul City, and the planning of the Institute. Discussions about the Institute reflected ways of talking about race, in particular, that had roots in the local historical struggle for civil rights.

I used the data to identify different perceptions of these events by planning participants and community members. By investigating people's actions and viewpoints in the 1960s and 1970s, I was better able to understand their roles in and perceptions of the Institute's planning process in the 1990s.

I was consistently aware of the fact that I was interpreting these events through my own standpoint (Collins, 1990; Harding, 1991). My own current context, lived experiences, and limited knowledge of the Civil Rights Movement in North Carolina shaped my understanding of the data. To try to combat these biases, I relied on Warren County natives and Civil Rights scholars to provide critical feedback regarding my interpretations. I discussed my growing understanding of specific phenomena during interviews and casual interactions with these residents and scholars and invited them to offer alternative interpretations. This happened throughout the project in both unstructured and structured ways.

One clear example of this process occurred in the area of current and historical intraracial divisions among African Americans. While I was aware of current political and social divisions among African Americans in this country, I had a particularly idealistic view of blacks in the South during the Civil Rights Movement. Popular media led me to believe that all Southern blacks supported integration and all Southern whites opposed it. When I started to recognize the lack of consensus around integration among African American interviewees who had been Warren County residents in the 1960s, I began to ask them how they would characterize these different positions. This led me to ask them questions about other divisions among African Americans in the county. People identified political and social divisions, as manifested in people's participation in different social groups, churches, and social change activities. Some of these divisions mirrored divisions between and among the members of the two groups involved in the Institute's planning: the Planning Group, comprised of local professionals and government officials, and the Leadership Development Team, a group of low-income residents. Acknowledging and understanding these divisions was integral to my understanding of the local context, undoing the common myth of the Civil Rights Movement as a homogeneous black Southern community who all worked together for integration.

As my analyses progressed, I created a "construction group," composed of Warren County residents, who critically reflected on my interpretations of the data. I talked with these residents about the stories I was developing about the Institute's planning and life in Warren County, and asked for their thoughts on the stories' accuracy and completeness. They confirmed my growing understanding of a more complex "black community." It was my hope that these processes would make the project less corrupted by my own biases and more clearly reflective of the perspectives of local residents.


The software program was helpful in sorting the data into common themes, issues, and concerns; yet it was incapable of producing a coherent story. The responsibility for creating the story lies with the researcher-storyteller. One way I started putting together the story was by reviewing my notes, written after the interviews. Each time a person articulated a response that was unexpected or different from prior responses, I noted my surprise or interest.

When an African American woman stated that she had not supported the integration of the public schools in the late 1960s, claiming that she "did not have a child she was willing to sacrifice for integration," I was surprised and noted it in my journal. Soon after, a white woman condemned the integration of the schools in the same language that the black woman had used. I used my journal to sort out my thoughts about the similarities and differences in the meaning of these words as used by the white and black women.

Both comments were offered from the perspectives of mothers and used the gendered language of mothering. Yet, even though these two women used the same words-"not having a child to sacrifice for integration"-the meanings they attached to the words were rooted in different discourses on race. While the black woman was concerned that her child would be physically and psychologically assaulted, the white woman was afraid that her child would receive an inadequate education. After analyzing my own commentary, I then went back to the data to compare all of the different responses about this historical topic, which had been assigned a topic code. I discovered that the divisions on the issue of racial integration were embedded into distinct discourses shaped by race, class, gender, education, political orientation, and experience.

My analysis was in many ways limited by the design of the research project, my own biases and blindspots, and the limitations of the construction group. Construction group members were chosen for their willingness to read the manuscripts and provide feedback; therefore, they were obviously a select group. However, I completed the analysis and composition of these stories with a feeling that they were stronger than they would have been without the construction group. I felt that my own understanding of the local culture, history, and politics was much more nuanced at the end of the project as a result of the critical insights of the construction group members. Researchers who use this "construction group" method should remember to strive for diversity among group members. They should also be aware that all analyses will be incomplete and imperfect, and will be structured by the theories, experiences, and identities of the researchers and the construction group members. For these reasons, the storyteller must try to identify these aspects of her (and the group's) perspective(s).

Writing It Up: Composing the Story

Stories of all kinds, from great literature to dime store novels, rely on the same elements: plot, scene, and voice. In writing the narrative of the planning of Warren Family Institute in my dissertation (Messinger, 1999), I concentrated on developing these three aspects of the story. In this section, I will recount my approach, offering the story of my story-writing process.

Penning the plot

The basic plot of the planning was fairly straightforward, told in chronological order. Using planning documents, along with individual planning participants' personal documents (i.e., notes, journals, memos), I mapped out the timeline of events. The Z. Smith Reynolds Foundation issued a call for initial planning grant proposals in January 1990. Warren County officials, who had participated with ZSR staff on several regional projects, wrote and submitted an initial proposal within the month. The Foundation's staff reviewed the proposals and awarded the county an initial planning grant of $20,000. This money supported the organization of a Planning Group (PG), composed of county government staff, local and regional agency staff, and local community leaders.

The planning grant also supported the establishment of a group of low-income residents, called the Leadership Development Team (LDT), who assisted the Planning Group by implementing a needs assessment survey. The LDT members knocked on doors and asked residents about the strengths and needs of Warren County. The Team reported the results to the professionals in the Planning Group, who then proceeded to write and submit the final proposal. The Foundation awarded a three-to-five year grant of $1,000,000 to the county in January 1992.

This description of the planning process served as the basic plot of the story. As I gathered more information, I added smaller steps and events and, slowly, the outline grew. The structure of planning story was formed.


As stated earlier, some very important documents, vital to accurately dating each step of the planning process, were lost in the passage of time. Several planning participants disagreed about the timing of certain elements, and there was often no paperwork to establish who was correct.

Crucial elements also were missing in some earlier versions of the planning story. In initial interviews, planning participants did not identify points of strain, disagreement, or hostility during the planning process. This was particularly true as relates to the tensions between the Planning Group and the Leadership Development Team. Initial interviews with PG members denied or downplayed strains between the two groups. Later interviews with members of the Team revealed a more contentious relationship. This was a challenge in writing the plot. I addressed this problem by revising the story as more information developed. I confirmed the details with earlier interviewees when I could, and relied on the memories of the majority when I could not get complete agreement.

I also noted points of disagreement around the role of different planning participants, interactions among them, and the effects of the process on the low-income women participants. Some planning participants characterized themselves as integral to the process, while others dismissed those participants' contributions as marginal. Several participants portrayed the planning process as effective and uncomplicated, while others viewed it as conflicted and sometimes ineffectual. Some members of the Leadership Development Team felt that they had gained from the experience, while others harbored negative feelings and saw little benefit to them. Rather than seeing these as discrepancies to be resolved, I concentrated on fleshing out these differences when creating the voice(s) in the finished piece.

Setting the scene

In a narrative, "scene" consists of several elements: physical setting, time, local cultures, and historical background of the characters and the place. Placing the plot in space and time was, for me, the easiest aspect of establishing the scene. It was much harder to define and describe the local culture and history of Warren County in a way that is clear and digestible for the reader.

One of the ways I attempted to clarify the local culture was through the definition of common norms. Through interviews with residents and planning participants, I was able to identify seven ethical stances and four community beliefs shared by all. The ethical stances addressed issues such as the need to care for one's family, the importance of education for one's children, and the importance of maintaining strong community relationships. Shared beliefs included a reluctance to see change, an appreciation for solitude and quiet, and a desire to be close to family members. Most of these norms corresponded to norms described in the literature on rural agricultural communities (Carlton-LaNey, 1992; Ginsberg, 1976). I reflected on these norms as I fleshed out the planning process, noting how they shaped participants' interactions and the reactions of community members to the emerging Institute. I learned about the history of the county by reading historical documents, including old editions of local and regional newspapers, previous studies of Warren County residents, published family and academic histories of the county, and historical documents in local and state archives. Reviewing all of these data was challenging; the irrelevant detail overwhelmed me. I worked to sort out the important historical events. I focused on specific events identified by subjects as most important to shaping either their lives or the county as a whole. This led me to investigate historical information about the 1960s and 1970s, specifically in the areas of integration and Soul City. In the final analysis, these events provided the backstory for the planning participants while offering possible motives for their actions during the Institute's planning. The more thorough history also helped to locate the planning process as one event in the history of a 100-year-old county in North Carolina.


My expectations shaped the creation of the scene as much as the words of residents, historians, and previous researchers. Often uncomfortable during my field visits in this rural setting, I brought my own emotions and beliefs to the interviews. It seemed reassuring that the data on local norms mirrors the literature on similar rural areas. However, it also made me wonder if the identified norms were simply those I sought out as typical in rural settings. I might have missed more uncommon norms that were not articulated so overtly. Again, the work of the construction group strengthens these findings, as they were in agreement about the centrality of the identified norms to local culture.

The same was true of the historic events that emerged as relevant. Soul City and racial integration interested me early on. Soul City was an enigma, words on a sign that loomed like a question mark over the small group of houses and buildings sitting somewhat adrift on the vast expanse of undeveloped land. As I noted in my journal:

Soul City reminds me of [my hometown in the Northeast] -- the way folks treat it, the way it is castigated because of the black majority, the crime rate, the sense of hope that brought residents, the fear of others who live around it. It makes me feel like a natural ally.

My curiosity about, and affinity for, the town easily could have influenced my decision to focus on Soul City.

My experiences in my hometown, a racially diverse New Jersey suburb, also served as my baseline for understanding and interpreting race relations in Warren County. I was discomfited by the stares that I received from restaurant patrons as I ate lunch with the black women who worked at the Institute. I was curious about how formal racial integration-so much more recent here than in New Jersey-was accomplished, and whether it was still shaping the informal social segregation that still exists in Warren County. While I was able to connect these events to the current planning, it was possible that there were more important factors that I overlooked in my interest in these historical topics.

There is a need, then, for researchers to identify their own interests and biases when building the scene of the story. My own "character," my "backstory," undoubtedly shapes this part of the construction of the story and belongs in the final story. Journal entries and personal notes were very helpful in identifying how my own interests factored into the construction of scene. This can also be addressed in the positioning of the researcher as storyteller.

Giving Voice(s)

Voice, also called narrative style, includes the positioning of the author, the positioning of the subjects of the research, the tone of the piece, and the piece's form. Authors of mainstream positivist research articles and reports often write in the third person, hiding their own personalities and character behind the "mask" of the objective researcher (Fonow & Cook, 1991). This approach is unacceptable for a researcher-storybuilder whose emotions and decisions shape the research project so thoroughly. Instead, the storyteller must reveal her role in designing the research, implementing it, and writing the final story. This approach is common to most current ethnographic and qualitative research as a way to improve the validity of the research (Lather, 1993). Feedback from Warren County community members, other Southerners, and friends from the Northeast helped me to identify moments when my own cultural and theoretical background clashed with those in Warren County. I recorded this feedback in my personal journals.

One of the most memorable moments for me was a painful interview with a very racist, white professional couple. After calling friends in New Jersey and North Carolina, I wrote in my journal:

You know, they started off and I guess I sort of thought they were liberal. But they just implied that somehow, miraculously, [the community] would have become a more racially equal and just place-without any provocation, without sit-ins, and social movement, and things like that. Without the court ordering, without the Civil Rights Act & the Brown V. Board of Education decision. Instead, they were so hung up on [telling me] that black people [at that time] couldn't write, that black people [at that time] were stupid, and you know... It was too much for me. And they were even trying to be careful! That is what they said when they were careful.

My friends helped me identify that I wasn't surprised that this couple harbored racist beliefs; instead, I was surprised that their racism could come out through such a thinly-disguised veneer of civility. I was shocked and offended that they felt that they could say these things to me, someone they didn't even know. Disdain for blacks and a dismissal of blacks as people were underlying themes throughout the interview that pushed all of my buttons.

Feminist theorists (Fine, 1994; Fonow & Cook, 1991; Lather, 1993) have argued that just as researcher-authors should identify themselves and their own raced/classed/sexed place, they also need to open the space for the research participants to present themselves in the story's text. There is little consensus among these theorists about how to present (and represent) participants' stories in the final text, but I have incorporated some of their suggestions into writing the final draft of the story of the Institute's planning (Messinger, 1999).

Lather (1993) and Fine (1994) advocate the inclusion of multiple voices throughout the text, providing space for the participants to disagree. Participants speak from their specific subject position when large quotes from their interviews appear in the final story. I have tried to include as many diverse opinions and perspectives as possible in the finished story, without bogging down the text in too many quotations (Chenail, 1995). The story (Messinger, 1999) identifies the ways in which participants' perspectives on planning were rooted in professional and personal identities, prior planning experiences, and individuals' social status. I don't provide any adjudication as to which opinion or perception is "correct." Instead, I challenge the reader to join me in finding the truths and, more importantly, interpreting the meanings in all of the responses individually and as they inform one another (Noblit & Engel, 1991).

Fonow and Cook (1991) and Lather (1993) also suggest that the power of giving voice be shared with the research participants, inviting them to join in authorship of the final text. This was not feasible in my research for two reasons. First, I was writing a doctoral dissertation and I felt that I should be the sole author on this work. Second, I chose the topic to meet academic standards rather than a more popularly focused article. I did not imagine that any of the participants would want to help write an academic paper on the role of context in planning comprehensive community initiatives. Instead, I employed the construction group as a means to include participants in the story's telling, letting the group members review the story as it was being written and suggest changes. These changes were incorporated into the final version in two ways: if I agreed with the suggestion, I made the change; if I disagreed, I added the alternative interpretation to my own.

In qualitative research, the tone and form of the story vary with the topic of the research and the forum for which it is being written (Bogdan & Biklen, 1992). The story of planning Warren Family Institute concerned a rural Southern setting, where storytelling and plain-speaking are established practice. This folksy approach contradicted the story's desired forum-an academic journal article that was part of a larger doctoral dissertation-which usually include some academic jargon, references, and a standard format (Madsen, 1990). The final version was a negotiation of both styles, opting for a story format, a more narrative journalistic style, and limited academic references.


Just as feminists and other qualitative researchers have been arguing for researcher reflexivity and the inclusion of multiple voices in the presentation of research, they have been cynical about the possibility of achieving these goals. Fine (1994) is very clear on this point, as she presents the challenge of "working the hyphen, reconciling the slippery constructions of Self and Other and the contexts of oppression in which both are invented."(p.78) As stated above, the limits of the construction group, the decision not to include all of the research participants in the writing, and the choices of an academic topic and forum for the story raise questions about the ability of the final story to meet the standards of feminist research. The ethic of the "holistic injunction" (Noblit & Engel, 1991) in qualitative research served as an ideal, pushing me to learn as much as I could about the county, incorporate all of the data into the analysis process, and tell as complete a story as possible. (p.128) My struggle with this ideal is acknowledged as a tension in the final piece.

Editing: Refining the Story

How do you decide when a story is done? Whether a story is good and/or accurate? Walter Fisher (1987) suggests that stories can be evaluated using two criteria: coherence and fidelity. These were the two criteria that I employed when I was editing my final account of the planning story.

Narrative coherence, or probability, is what constitutes a coherent story (Fisher, 1987, p. 64), one that hangs together and makes sense (p. 68). Narrative fidelity is whether the story rings true with stories people experience in their own life (pp. 64-65); that is, "the soundness of its reasoning, the value of its values" (p. 68).

I used members of my dissertation committee, along with friends and colleagues, to test my story for narrative coherence: did the story I proposed about the ways in which the culture and historical context shaped the planning make sense to them? What were the shortcomings of the story? As outsiders to Warren County who were unfamiliar with the institute, did the story help them understand the planning process? These respondents offered written and oral feedback.

Similarly, I relied on interactions with the Warren County construction group, as well as members of the evaluation research team and the foundation funding the intervention, to assess narrative fidelity. I asked them a number of questions: Did my story sound right to them? Even if it was not their own personal interpretation, was it "sound" and reflective of their own experiences? Could it be true for them? Throughout the research and writing process, these individuals reviewed my manuscripts, listened to my arguments, and offered feedback.

The final story was one that satisfied the members of my committee and the construction group, and therefore met Fischer's criteria for narrative coherence and fidelity. This approach to evaluation of the writing process helps the researcher to insure that the final story is of good quality.

Conclusion: The Moral of this Story

The story-building approach to community intervention research requires a variety of research methods, a greater focus on historical and current context, and an attention to form in writing the finished story of the intervention. These elements of the story-building approach produce a story that ultimately reveals more about the intervention than standard research approaches. This was clearly the case in my story-building research on Warren Family Institute.

My exploration of the historical and cultural forces that shape planning and implementation processes revealed the importance of looking at backstory and contextual issues. My investigation of the relationship between the historical struggles, such as the integration of public facilities and the establishment of Soul City, and current interracial and intraracial politics in Warren County revealed possible explanations for the particular process used in planning this CCI and the reactions of the community to the intervention. This storytelling approach is a vast improvement over mainstream approaches to research, which only lay out the "facts" of the planning and describe current tensions shaping the planning process. In my use of a story-building approach, I tried to explain why the planning progressed as it did.

The more narrative approach to writing the finished piece allowed me to present a more complex understanding of the supports and barriers facing the intervention. The multivocal story of Warren Family Institute is more representative of the diversity of perspectives in the community than the mainstream program evaluation report based on survey and outcome data. Moreover, telling the findings in a story should make them more accessible to social work practitioners, the target audience for social work research.

There has long been a concern among social work researchers that practitioners in the field have not used our research. Nowhere is that more true that in rural social work. Practitioners often criticize scholarly articles for using jargon and writing in ways that are uninteresting or unintelligible. Offering findings in a story, using this more popular style, might encourage readership and the adoption of suggestions for practice among these rural social workers.

Being conscious of the research project as a story-building process can help community intervention researchers remember to think about plot, scene, and voices, capturing all the contextual, historical, and perspectival data necessary for the story. We want the story we tell to be a good one, which accurately and fairly represents those individuals included and excluded from the intervention process, their relationships, and the context and the history in which their interactions take place. These stories we produce will increase our learning about community initiatives and help us to continue to create strong, effective initiatives in the future.


      Albrecht, J., & Lim, G. C. (1986). A search for alternative planning theory: Use of critical theory. Journal of Architecture and Planning Research, 3(2), 117-131.

      Annie E. Casey Foundation. (1995). The path of most resistance: Reflections on lessons learned from New Futures. Baltimore: Author. [On-line publication]. Available:

      Bogdan, R., & Biklen, S. K. (1992). Qualitative research for education. Boston: Allyn & Bacon.

      Carlton-LaNey, I. (1992). Elderly Black farm women: A population at risk. Social Work 37(6), 517-523.

      Chaskin, R. J. (1992). The Ford Foundation's Neighborhood and Family Initiative:Toward a model of comprehensive, neighborhood-based development. Chicago: University of Chicago, The Chapin Hall Center for Children.

      Chenail, R. J. (1995). Presenting qualitative data. The Qualitative Report 2(3). [On-line] Available:

      Collins, P. H. (1990). Black feminist thought: Knowledge, consciousness, and the politics of empowerment. New York: Routledge.

      Denzin, N. K. (1994). The art and politics of interpretation. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp.500-515). Thousand Oaks, CA: Sage.

      Fine, M. (1994). Working the hyphens: Reinventing self and other in qualitative research. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp.70-82). Thousand Oaks, CA: Sage.

      Fisher, W. R. (1987). Human communication as narration: Toward a philosophy of reason, value, and action. Columbia, SC: University of South Carolina Press.

      Fisher, W. R. (1989). Clarifying the narrative paradigm. Communications Monographs 56(1), 55.

      Fisher, W.R (1994). Narrative rationality and the logic of scientific discourse. Argumentation, 8(1), 21-32.

      Fonow, M. M., & Cook, J. A. (Eds.). (1991). Beyond methodology: Feminist scholarship as lived research. Bloomington: Indiana University Press.

      Ginsberg, L. H. (1976). Social work in rural communities: A book of readings. New York: Council on Social Work Education.

      Habermas, J. (1984). Theory of communicative action: Lifeworld and system. Vol. I: Reason and the rationalization of society (T. McCarthy, Trans.). Boston: Beacon Press.

      Habermas, J. (1987). Theory of communicative action:Lifeworld and system Vol. II:A critique of functionalist reason (T. McCarthy, Trans.). Boston: Beacon Press.

      Harding, S. (1991). Whose science? Whose knowledge? Thinking from women's lives. Ithaca, NY: Cornell University Press.

      Kirk, R. S., & Messinger, L. (1996). Opportunities for Families Fund: Final evaluation report. Unpublished manuscript, University of North Carolina at Chapel Hill.

      Kubisch, A. C., Weiss, C. H., Schorr, L. B., & Connell, J. P. (1995). Introduction. In A. C. Kubisch, J. P. Connell, L. B. Schorr, & C. H. Weiss (Eds.), New approaches to evaluating community initiatives: Concepts, methods, and contexts (pp. 1-23). Washington, DC: The Aspen Institute.

      Lather, P. (1993). Fertile obsession: Validity after post-structuralism. Sociological Quarterly, 35, 673-694.

      LeCompte, M. D., & Preissle, J., with Tesch, R. (1993). Ethnography and qualitative design in educational research (2nd ed.). New York: Academic Press.

      Madsen, D. (1990). Successful dissertations and theses. San Francisco: Jossey-Bass.

      Marshall, C., & Rossman, G.B. (1995). Designing qualitative research (2nd ed.). Thousand Oaks, CA: Sage.

      Medoff, P., & Sklar, H. (1994). Streets of hope: The fall and rise of an urban neighborhood. Boston: South End Press.

      Messinger, L. (1999). History at the table: Planning a comprehensive community initiative in the rural South. Unpublished doctoral dissertation, University of North Carolina, Chapel Hill.

      Noblit, G. W., & Engel, J. D. (1991). The holistic injunction: An ideal and moral imperative for qualitative research. Qualitative Health Research, 1(1), 123-130.

      Oleson, V. (1994). Feminisms and models of qualitative research. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp.158-174). Thousand Oaks, CA: Sage.

      Richardson. L. (1994). Writing: A method of inquiry. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp. 516-529). Thousand Oaks, CA: Sage.

      Stake, R. E. (1994). Case studies. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp. 236-247). Thousand Oaks, CA: Sage.

      Yow, V. R. (1994). Recording oral history: A practical guide for social scientists. Thousand Oaks, CA: Sage.

Author Note

+ Dr. Lori Messinger is an assistant professor of social work at North Carolina State University. She received her MSW and her Ph.D. in social work from the University of North Carolina at Chapel Hill, specializing in community planning and program evaluation. She also has an MA in political science from Rutgers University, with a concentration in Women and Politics. Her primary areas of research include comprehensive community planning processes, cultural competence in social work practice, feminist theories and research methodologies, and gay and lesbian studies. She is currently using qualitative methods to investigate the experiences of gay and lesbian social work students in field placement.

This article is based on research supported by a dissertation grant from the University of North Carolina at Chapel Hill School of Social Work.

Correspondence concerning this article should be addressed to Lori Messinger, MA, MSW, PhD, Assistant Professor, Social Work Program, North Carolina State University, CB# 7639, 1911 Building, Room 216, Raleigh, NC 27695; (919) 513-1990 (phone); and (919) 515-4403 (fax). Her email address is

Article Citation

      Messinger, L. (2001, March). Using a story-building approach to research comprehensive community initiatives. [74 paragraphs]. The Qualitative Report [On-line serial], 6(1). Available:

Lori Messinger
2001 copyright

Return to the top of the paper.

Return to the Table of Contents.