ESTABLISHING A CONCEPTUAL MODEL OF COMMUNITY-BASED RESEARCH THROUGH CONTRASTING CASE STUDIES

This study seeks to provide insight into the process of conducting community-based research. In order to do so, the study utilizes a qualitative case study approach to examine the methodology of community-based research. Two contrasting cases of CBR are described and analyzed in order to understand the issues that arise when conducting CBR, the factors that facilitate or hinder the process, and the benefits of conducting CBR. Finally, these contrasting cases are considered to determine what this study can contribute to the field of CBR. This chapter details case study methodology as well as multiple case design. It also describes the methodology of community-based research, the participants of the study, data collection and analysis, and issues around credibility, including my own subjectivities that may have influenced the research.

In order to explore the collaborative process of conducting community-based research, this study utilizes a qualitative case study approach. Case studies can be particularly useful for studying a process, program or individual in an in-depth, holistic way that allows for deep understanding (Merriam, 1998). As Merriam points out,

There are some differences in how researchers define case study. Some researchers think of case study as the object to be studied (Stake, 2000), while others define case study as a process of investigation (Creswell, 2002). Creswell defines case study as "an in-depth exploration of a bounded system (e.g., an activity, event, process, or individuals) based on extensive data collection" (p. 485). Creswell recommends case study as a methodology if the problem to be studied "relates to developing an in-depth understanding of a 'case' or bounded system" (p. 496) and if the purpose is to understand "an event, activity, process, or one or more individuals" (p. 496). Patton (1990) suggests that case studies are valuable in creating deep understanding of particular people, problems or situations in comprehensive ways.

This study is particularly suitable for a case study design because it is a bounded system, it is contextual, and it is a study of process (Merriam, 1998). Like Creswell (2002), Stake (2000) defines case study as the study of a "bounded system" (p. 436). According to Creswell (2002), "'Bounded' means that the case is separated out for research in terms of time, place, or some physical boundaries" (p. 485). In other words, it is possible to create limits around the object to be studied (Merriam, 1998). A case study can focus on a variety of different things. A case could be an individual, a group, a school, a community (Merriam, 1998), or a case could also include "a program, events, or activities" (Creswell, 2002, p. 485). The bounded systems in my contrasting case studies are my collaboration with the Coalition for Schools[2] in a western city and my collaboration with community members in a small, rural, mountain community to carry out community-based research. The boundaries of these two cases are determined by the people and groups that I collaborate with in the CBR process.

I chose a case study design because it involves "detailed, in-depth data collection involving multiple sources of information rich in context" (Creswell, 1998, p. 61). Context is a key factor. According to Merriam (1998), in focusing on a particular phenomenon in a case study, it is impossible to separate the phenomenon from its context. However, in this study, it is important that the context is understood as part of the process. As Yin (2003) says, "you would use the case study method because you deliberately wanted to cover contextual conditions-believing that they might be highly pertinent to your phenomenon of study" (p. 13). Thus, using a case study approach allows for the possibility of gaining significant knowledge about the process of conducting community-based research in particular contexts. According to Sanders (1981), "Case studies help us to understand processes of events, projects, and programs and to discover context characteristics that will shed light on an issue or object" (p. 44).

The two case studies each took place over an extended period of time. The first CBR project lasted nine months, and the second CBR project lasted eight months. I worked with my collaborative partners to define research problems and questions, develop research designs, collect data, and analyze data. However, this study does not focus on the data that I collected as part of that CBR work. Instead, this study focuses on the process of the collaborative experience. Since the study focuses primarily on the procedures of conducting community-based research, the study is considered a process study. According to Patton (1990), when carrying out a process study, the "focus is on how something happens rather than on the outcomes or results obtained" (p.94). And, as Merriam (1998) points out, "Case study is a particularly suitable design if you are interested in process" (p.33). Therefore, case study was chosen since it allows for detailed monitoring of the collaborative process (Merriam, 1998).

Stake (2000) delineates three types of case studies: intrinsic, instrumental, and collective. Intrinsic case studies focus on a case that is unusual and is of particular interest to the researcher (Creswell, 1998; Stake, 2000). The intent is not to build theory (Stake, 2000). An instrumental case study is pursued in order to provide insight about a particular issue that may be generalizable (Creswell, 2002). The primary purpose of an instrumental case study is to help advance understanding (Stake, 2000). The collective case study encompasses more than one case "in order to investigate a phenomenon, population, or general condition" (Stake, 2000, p. 437). Since the purpose is to help advance understanding, a collective case study is a grouping of instrumental case studies (Stake, 2000). Using a collective case study approach can allow for the possibility of stronger interpretation and "perhaps better theorizing" (Stake, 2000, p. 437).

Though Stake (2000) uses the terminology "collective case study," this approach is known by other names such as, multiple case studies, cross-case studies, comparative case studies, and contrasting cases (Merriam, 1998; Yin, 2003). With multiple case studies, data are analyzed for insights both within each case and across cases (Merriam, 1998). Yin (2003) points out that multiple cases may be chosen to try to replicate insights that you find within individuals cases or to represent contrasting situations. Regardless of whether the purpose is replication or contrast, multiple case studies are "considered more compelling, and the overall study is therefore regarded as more robust" (Yin, 2003, p. 46).

When this study was first proposed, the original intent was to pursue a single case study of my experience of collaboration in carrying out a community-based research project. After completing my work with the Coalition for Schools, I felt dissatisfied with the experience in that I did not view it to be a success. Instead of focusing on that one experience, I decided to pursue another research option in the small town in which I live in order to have a contrasting experience to write about. It turned out that the project I completed in my small town was more successful, therefore allowing me to present contrasting cases. Since this study seeks to add insight to the field of methodology in CBR, it is important to understand the factors that impact the process of collaboration and the factors that support successful collaborations (Strand et al., 2003a).

Since the purpose of this study is to explore the process of carrying out CBR, it is important to understand the methodology of community-based research. As mentioned in chapter two, community-based research is not as concerned with methods as it is with methodology (Hills & Mullett, 2000; Strand et al., 2003a). Either quantitative or qualitative methods may be used; the choice depends on what would obtain the most useful data for the community (Greenwood & Levin, 2000). The methodology of CBR is guided by the three principles outlined by Strand et al. (2003a): 1) collaboration, 2) validation of the knowledge of community members and the multiple ways of collecting and distributing information, and 3) "social action and social change for the purpose of achieving social justice" (p. 8). Though community-based research is not limited to specific methods, it does follow the typical stages of research that most traditional academic research would follow: defining the research question, developing a research design, collecting data, analyzing data, and writing up the results. The difference is that the researcher collaborates closely with the community throughout the research process (Strand et al., 2003a). The community is involved in determining the problem and research questions, creating the research design, collecting data, analyzing data, and creating a presentation of findings (Strand et al., 2003a). The researcher also continues to play a role in the final stage by assisting with the enactment of solutions to create change (Greenwood & Levin, 1998).

Regarding knowledge, community-based research seeks to redefine how we conceptualize knowledge in relation to academic research (Strand et al., 2003a). Researchers who conduct CBR projects recognize the important knowledge that community members possess on the subject of their environment and the issues they are dealing with (Cordes, 1998a, No Concrete section, para. 2; Hills & Mullett, 2000, p. 1), what Strand (2000) calls "local knowledge" (p. 88). This knowledge is key throughout the research process. This acceptance of community knowledge does require the researcher to rethink his or her role. As Stringer (1996) says, "The role of the researcher is not that of an expert who does research, but that of a resource person" (p. 22). The expertise that the researcher brings to the equation is still valued; however, the local knowledge that the community brings is recognized as integral to the research process (Strand et al., 2003a).

I have provided a brief overview of the methodology of CBR. However, the purpose of this chapter is to describe the case study methods that I used to carry out this process study. The descriptions of data collection and data analysis that are included in this chapter pertain to the data that were collected and analyzed for the contrasting case studies. A description of the data collection and analysis that was conducted for the CBR projects in each case study will be included in the case descriptions in chapters four and five.

Though I came into contact with a variety of people in each case study, my primary research collaborators are the main participants of my study. In the first case study that I carried out, my collaboration with the Coalition for Schools, there were initially two primary collaborators, one of the co-chairs of the Coalition, Marge Bowline, and the director of the Coalition, Lisa Brown. As my collaboration progressed, I worked primarily with Lisa Brown.

The Coalition for Schools is an organization that has been created to support greater academic achievement in an urban school district in a western city. The Coalition has focused its efforts toward a feeder pattern of schools in a quadrant of the city that has a high percentage of students who are eligible for free or reduced lunches, a high percentage of minority students, and a high percentage of English language learners. This feeder pattern includes five elementary schools, two middle schools, and three small high schools that were originally part of one large high school and that are housed in one building. The Coalition is an alliance of non-profit organizations, foundations, parent organizations, universities and colleges, and the school district working together to support achievement in these low performing schools. The Business and Schools United (BSU) organization is the lead partner for the Coalition, and the Coalition is housed at BSU. Marge Bowline is the director of BSU and one of the co-chairs of the Coalition for Schools. She helped to create the Coalition and to procure funding for the organization. The Coalition was a year old when I began my work with them. Lisa Brown was hired to direct the Coalition and replaced the first director. She had been in her position for about six months when I began my work with the Coalition.

The two primary collaborators in my work in a small, western, mountain town are John Brewer and Maria Swenson. The town is a small rural community that has a rapidly growing immigrant population from Mexico, about half of which are Indians from a remote area of the country. Both John Brewer and Maria Swenson work in positions that have direct contact with this population. John Brewer is the director of the literacy program which offers free English courses for English as a Second Language (ESL) students. He is also a member of the city council. Marge Swenson, who is herself a former immigrant from South America, is the coordinator of the diversity office which provides services to immigrants in town. The case descriptions in chapters four and five provide greater detail of the participants and setting.

As I progressed through each case study, I pursued two streams of data collection; the data collected to pursue the CBR projects and data that were collected as part of this case study to study CBR. This section describes only the data that were collected for the case studies. A description of the CBR data that were collected for each collaboration is included in the case descriptions in chapters four and five.

Since the purpose of case study research is to provide an in-depth exploration of the person, program, or process under study, it requires intensive data collection (Merriam, 1998; Yin, 2003) using "multiple forms of data" (Creswell, 2002, p. 486). Data collection for case studies usually focuses on three sources of data: observations, interviews, and documents (Merriam, 1998). Though all qualitative research is to some extent based on the idea of emergent design, this study was truly emergent. Though the research questions that this study proposed to address did not shift throughout the study, the methods of data collection changed to accommodate emerging issues or ideas. According to Patton (1990),

Though I collected all three forms of data (observations, interviews, and documents) for each study, there are some variations that are detailed in the following sections. Appendix A provides a list showing the dates of meetings and interviews for each case study.

My primary source of data collection for both case studies was observation. Since I was essentially observing myself as I collaborated with my community partner, all of the observations that I completed for my case study data collection were participant observations. Creswell (2002) defines participant observation as "an observational role adopted by researchers when they take part in activities in the setting they observe" (p. 200). In this role, the researcher "actually engages in activities at the site begin studied" (p. 200). Glesne (1999) describes a continuum of participation that "ranges from mostly observation to mostly participation" (p. 44). Based on this continuum, I was what Glesne (1999) describes as a "full participant" in every interaction relating to my collaborative work with my community partners since I was concurrently a member of the collaborative partnership as well as the researcher investigating the process.

In all of the meetings that I conducted with my community partners in relation to our CBR work, I collected data around those interactions. I utilized Merriam's (1998) checklist of elements to structure my observations: physical setting, participants, activities and interactions, conversation, subtle factors, and my own behavior (pp. 97-98). When working on my first CBR project with the Coalition, I initially only maintained field notes. I was concerned that if I taped our meetings that it would be intrusive and would impact the openness of our conversations (Merriam, 1998). However, as my study progressed I realized that it was difficult to take effective notes while participating in the conversation. I then asked my community partners if I could tape subsequent meetings. After that, most of the meetings I had with Lisa Brown or Marge Bowline were taped and then transcribed. As part of the transcription process, I added notes that clarified or contextualized the dialogue. When I began my work with my community partners in my small town, I asked during the first meeting if I could tape all of our meetings; both John Brewer and Maria Swenson readily agreed. I found that after the use of the tape recorder became routine, they did not seem to be inhibited by being recorded. Using the tape recorder allowed me to collect much more extensive data from my observations of our meetings.

As part of the data collection for both case studies, I collected both formal and informal interview data (Patton, 1990). Informal conversational interview questions were interwoven into meetings that we had in relation to ongoing research (Merriam, 1998) and were recorded as part of observation transcriptions. These informal questions typically addressed how the community partner felt the research process was progressing, whether the research was meeting their needs, or addressed immediate questions that arose through the process of continued interaction.

I also collected formal interview data for both case studies; however, I conducted fewer formal interviews with my community partners from the Coalition for Schools. As my work with the Coalition progressed, I sought to determine particular data collection procedures that would address my research questions. Since I was working within a collaborative relationship, part of the consideration when choosing methods was the impact that various methods would have on the relationship with my community partner. In this first case study, as I show in more detail in chapter four, it was challenging to develop a collaborative relationship with my community partners. The lack of trust and communication within this relationship made it difficult to carry out formal interviews discussing our collaboration. I felt that these kinds of interviews would create greater distance between us. Instead I relied primarily on other forms of data collection, observations and documents. However, I did interview both Lisa Brown and Marge Bowline once formally toward the end of our partnership. This interview included questions about the work of the Coalition as well as questions relating to community-based research (Appendix B). I also conducted a follow-up email interview with Lisa Brown after beginning the process of data analysis (Appendix B).

In my collaboration with John Brewer and Maria Swenson in my small town, I was able to develop a much more honest and open relationship from the beginning and felt very comfortable conducting formal interviews about the process. I interviewed John and Maria individually three times throughout our collaboration (Appendix B). I used a semi-structured approach (Rubin & Rubin, 1995) when designing the interview protocols. I prepared questions as a starting point, but allowed the conversation to flow in whatever direction was helpful to providing insight. The first interview focused on getting a sense of their background and experiences with research, their expectations for our research, and strategies for effective communication. The second interview focused on their satisfaction with how things were proceeding, whether they felt we were communicating effectively, and whether they were having the input they wanted to have in the process. The final interview focused primarily on the research questions of the case study: what were the issues that arose, what helped or hindered our collaboration, and what benefits did they receive from the research. I transcribed each interview and added additional notes for interpretation.

As part of the data collection process, I also collected or created a variety of documents including: email communications, a reflective journal, a phone call log, and other items that were provided by my community partners such as newsletters and meeting minutes. As part of my collaboration with the Coalition for Schools, we relied extensively on email for communication since I found it difficult to schedule face-to-face meetings with Marge Bowline and Lisa Brown. These email conversations are an important source of data in compiling a picture of our collaborative experience. I also collected email data during my second case study. However, these email communications focused primarily on setting up logistics. Most important conversations were conducted face-to-face.

Throughout both case studies, I sought to engage in a reflective stance toward my role in the research process. In order to aid my reflection, I maintained a journal in which I transcribed my thinking in relation to my experiences and the perceived experiences of my community partners. Merriam (1998) expresses some concern about using personal documents such as journals as data. Merriam (1998) says,

However, Merriam (1998) does point out that one of the goals of qualitative research is to "reflect the participant's perspective" (p. 116). Since this is a process study, the perceptions of all participants are a key consideration (Patton, 1990). As I am a participant in this study, my perceptions of my experience of the process are important.

The other documents I collected consisted of a phone call log and documents obtained when meeting with my community partners. The phone call log consisted of a brief description of phone calls that were made during the research process. If the conversation was extensive, I tried to recreate the conversation as closely as possible. The phone call log was used primarily during my collaboration with John Brewer and Maria Swenson. I also obtained various documents from my community partners. These mostly included newsletters, meeting minutes, and data collected from previous research. Most of the documents related to the CBR work we were conducting; yet some of the documents also provided information for my case study research.

After completing both case studies, I had accumulated large volumes of data (more than 500 pages of data for each case study). I organized the data from both cases into what Yin (2003) calls a case study data base. I organized my case study data base in a chronological order so that I could move through the data from the beginning to the end of the process. This allowed me to perceive the progression of the process and my changing views throughout. However, I felt that I needed an additional frame from which to organize the data.

Data analysis was an ongoing process throughout the implementation of each case study. Periodically I composed analytic memos to begin to formulate ideas around particular findings. As each study progressed, I looked for events with common elements within the data that had "issue-relevant meaning" (Creswell, 1998, p. 154) or significance for the study. As I recognized these common elements, I focused on determining whether they continued to be supported throughout the data collection process. Creswell (1998) calls this process categorical aggregation. As categories within the data began to emerge, I began to look for patterns or themes that connected these categories. Based on the literature and the categories and themes that emerged while conducting the cases, I created an analytic framework from which to organize and think about the data.

The analytic framework is composed of four categories: community, collaboration, knowledge creation, and change. In creating this framework, I was influenced by Stoecker's (2003) delineation of radical and mainstream CBR. I view each of the four constructs of my framework as existing on a continuum. At one end, there is radical CBR, in the middle, mainstream CBR, and at the other end the professional expert model or consulting (see Figure 1). Based on how I conceptualize this framework, the closer on the continuum the researcher moves toward radical CBR, the greater the potential for change that will benefit the community with which the researcher is collaborating.

Figure 1. The Four Constructs of CBR

Consulting

Mainstream CBR

Radical CBR

Non-Representative Orgs.

Midlevel Orgs.

Grassroots Orgs.

Community

Researcher Holds Power

Partial Collaboration

Shared Decision Making

Collaboration

Researcher Controls Knowledge

Partial Participation

Community Creates Knowledge

Knowledge Creation

No Discernable Change

Programmatic Changes

Structural Change

Change

When considering the category of community, the goal is to work as closely as possible with the community. Since the ultimate goal of CBR is "social change for social justice" (Stoecker, 2002a, p. 9), the closer the researcher is to the members of the community who are dealing with the problem (Stoecker, 2003), the greater the potential to empower. The community continuum includes grassroots organizations on one end and organizations which do not represent the community or use practices that "disempower the community" (Strand et al., 2003a. p. 73) on the other (see Figure 1). In between are organizations that are a level removed from grassroots organizations but still seek to represent the community democratically, what Strand et al. (2003a) call "midlevel organizations" (p. 74). Conducting CBR projects with midlevel organizations is what Strand et al. (2003a) label "doing CBR in the middle" (p. 73).

Within this analytic framework, I conceptualize collaboration as shared decision making. The goal is that the community should have equal power with the researcher and that decision making should be a shared process throughout (Strand et al., 2003a). When considering this concept within the continuum, shared decision making is at one end of the continuum and at the other end the decisions are made primarily by the researcher (see Figure 1). A companion to collaboration is the concept of participation in knowledge creation. The primary goal in relation to this aspect of the framework is that the community assists in the creation of all knowledge that is generated during the CBR process, thus leading to community empowerment. This point of the framework is based on the principle that the knowledge of community members is valid (Strand et al., 2003a) and integral to creating strong results. At one end of the continuum, the community is involved in all aspects of knowledge creation, at the other end, the researcher controls the creation of knowledge (see Figure 1).

The final point of the analytic framework is change (see Figure 1). If you consider CBR within the radical framework described by Stoecker (2003), the goal for change is "massive structural changes in the distribution of power and resources through far-reaching changes in governmental policy, economic practices, or cultural norms" (p. 36). This goal can be difficult to achieve. More often, CBR work leads to programmatic changes within an organization or other more limited changes (Strand et al., 2003a). However, each change within a community can have a cumulative effect that can lead to broader change. Community-based research that does not involve the community in close collaboration and knowledge creation is less likely to create change that benefits the community.

Since this study utilizes contrasting cases, data analysis occurs at two levels: within-case and across cases (Merriam, 1998). Merriam (1998) describes this process:

For each case, I analyzed observations, interviews, and documents to develop a description of the case. This description depicts the setting and participants as well as a general chronology of events and provides the reader with an understanding of the particulars of the case (Creswell, 1998). This allows the reader to develop an understanding of the case within the larger context (Creswell, 2002). Then using the analytic framework I developed, I did some within-case analysis and organized the categories that emerged during each study around the four constructs of my analytic framework. This within-case analysis focused on answering the primary research question: What is the process of collaborating with a community partner on a community-based research project? Thus each case analysis consists of "both description and thematic development" (Creswell, 2002, p. 486).

After completing the within-case analysis, I focused on the cross-case analysis to address three of the sub-questions of the study: What kinds of issues arise when collaborating on a community-based research project? What facilitates or hinders the process of collaboration? and, What does the researcher gain through this collaborative process, and what are the benefits for the community? In the cross-case analysis, I used data from both case studies to address these questions. I explored the categories that had emerged throughout each case study and then compared to see if these categories were supported in both cases. I used the categories and themes that emerged during the within-case analysis and the cross-case analysis to determine "naturalistic generalizations" (Creswell, 1998, p. 154) concerning the field of community-based research. Creswell (1998) defines naturalistic generalizations as "generalizations that people can learn from the case either for themselves or for applying it to a population of cases" (p. 154). These naturalistic generalizations address the final question of the study: What can we learn from these experiences to inform the field of CBR?

In order to lend credibility to the findings of my study, I incorporated a variety of validity procedures. The first validity procedure I employed was prolonged engagement in the field (Creswell & Miller, 2000) or what Merriam (1998) calls "long-term observation" (p. 204). I worked on my case study with the Coalition for a period of nine months, and I worked with John and Maria for a period of eight months. During each of these case studies, I had consistent contact with my community partners. Collaborating with my community partners for this length of time allowed me to develop tentative categories in my findings and then follow up on these preliminary findings through observations or interviews (Creswell & Miller, 2000). Therefore, the length of each case study and the consistent contact I had with my community partners lends credibility to my perceptions of this experience.

In addition to prolonged engagement in the field, another important validity procedure I employed, which is integral to case study design, was triangulation (Creswell, 1998). Merriam (1998) defines triangulation as "using multiple investigators, multiple sources of data, or multiple methods to confirm the emerging findings" (p. 204). I employed methodological triangulation (Creswell & Miller, 2000) since I collected three forms of data: observations, interviews, and documents. I also employed multiple sources of data since interviews were conducted with several participants (Creswell & Miller, 2000). I used the process of triangulation to seek convergence in the data and to confirm or disconfirm emerging categories and themes (Creswell & Miller, 2000). As part of this process, I employed another validity strategy, disconfirming evidence (Creswell & Miller, 2000). Categories or themes that emerged in the within-case analysis were compared across cases. If a category did not hold true across cases, it was generally deemed to be unreliable. However, I did utilize what Creswell (1998) calls direct interpretation. In direct interpretation, "the case study researcher looks at a single instance and draws meaning from it without looking for multiple instances" (p. 154). I did recognize that there were single incidents specific to only one case that were significant to the study as well.

Since this case study focused on the study of process, my perceptions were an integral component of the research. However, since I did write interpretations of what I considered to be the perceptions of others, I used member checking to ensure accurate portrayal (Creswell & Miller, 2000). I conducted member checking toward the end of the study so that it would not potentially disrupt the collaborative process. I shared an outline of findings with Lisa Brown with the Coalition and also John Brewer and Maria Swenson in my small town and allowed them the opportunity to provide feedback. Lisa Brown responded to the findings through email and said, "Thanks for sharing [these findings]. I feel it is accurate, and that it was a learning experience for all of us." Maria Swenson also responded to the findings that I shared with she and John. She said, "I looked at [the findings] and it sounds good. I agree with all said." John also said that he thought that the findings looked good.

Finally, I used the validity procedure of thick description when writing about the study in order to give the reader a sense of being there and to capture the essence of the experience (Creswell & Miller, 2000). This is an important feature in case study design that is presented to the reader through the case description. The case description for each contrasting case is included in chapters four and five.

Another method of creditability I used continuously throughout the research process was researcher reflexivity (Creswell & Miller, 2000). I incorporated researcher reflexivity by constantly questioning my assumptions about what I thought was happening. I sought to maintain a heightened sense of awareness of the biases that I brought to the study and maintained this awareness when adding contextual data to field notes, observations transcriptions, and interview transcriptions, and also when writing journal entries.

Since my perceptions of the research process played a major part in the findings of the study, it was important that I attend to the idea of subjectivity. Peshkin (1988), defines subjectivity as "the quality of the investigator that affects the results of observational investigation" (p. 17). Peshkin (1988) points out that an individual's subjectivity is not something that can be removed, and it is therefore something researchers need to be aware of throughout the research process. Peshkin (1988) identified the various facets of his subjectivities through a series of I's, for example, the "justice-seeking I" (p. 18) and "the community-maintenance I" (p. 18). Though Peshkin does not view subjectivity as necessarily negative, he does feel it is something that researchers need to realize and acknowledge. It was important to examine my own subjectivities throughout the research process so that I was aware of how these subjectivities could influence my interpretations and portrayal of events. As Strand (2000) points out, "the researcher's values, experiences, and personal points of view are as much a part of the research process as those of the people studied, and they should be discussed and acknowledged" (p. 91).

Since the two CBR projects I worked on were in different settings and related to different types of work, I dealt with different subjectivities within each case study. In my work with the Coalition for Schools many of the subjectivities that I brought to that collaboration arose from my past experience as a classroom teacher. I hold the perception that people who do not have experience in a K-12 classroom do not generally understand the issues that classroom teachers have to address. I can be defensive and overly sensitive to criticism that I feel puts the blame on teachers. There were many times during my partnership with the Coalition that I realized this subjectivity was influencing my reactions to statements made by Lisa Brown or Marge Bowline. I also think that this perception at times clouded my view of the knowledge that Lisa brought to the equation. Though I felt that she was very knowledgeable in certain areas, I questioned her understanding of what was actually happening in the schools that are part of the Coalition. I tried to be aware of my bias in this area, though I do not believe I was always successful in controlling how this bias influenced my work with Lisa.

Another bias that I brought to my work with the Coalition was the idea that a successful partnership should not have conflict. I tend to avoid conflict in my personal life. I have difficulty at times recognizing the benefits that conflict can bring. Because of this, I did not communicate as effectively with Lisa as I could have. If had been more willing to risk conflict, we may have been able to develop a more productive working relationship. When I began my work with John Brewer and Maria Swenson, I determined that I would not avoid conflict in this collaboration. When a situation did arise where John and I disagreed, I engaged him, and we talked through the matter. The outcome was that we both were able to see the value of the other's viewpoint.

Though I was able to address the issue of conflict avoidance in my work in John Brewer and Maria Swenson, there were other subjectivities and biases of which I had to be aware. I am liable to have the perception that small towns tend to discriminate against minorities. Since all of the projects that I completed with John and Maria involved the immigrant population in town, I felt at times that I was waiting for someone to say something that would demonstrate their prejudice. At times, I would jump to the conclusion that a particular statement was pejorative. When looking back again at the statement in the context of the full conversation, I realized at times that I may have misinterpreted particular statements. I had to make a concerted effort not to single out statements just because they supported my bias. Nevertheless, this subjectivity did influence whom I chose to partner with during this case study. I had originally planned to include Maria's supervisor, Jennifer Payton, in our collaboration. However, after meeting with Jennifer in October 2003, I decided not to collaborate with her since she made several comments during the meeting that I perceived to be pejorative. If I had decided to work with Jennifer, I may have found that these comments did not represent discrimination but rather a lack of understanding of the impact of language choices.

Two other subjectivities that I brought into my work on both projects related to my experience with previous CBR projects. As I was involved in another community-based research project before working on my dissertation, I already had an initial perception of how the process works. One concern that arose during my previous experience was the issue of communicating with my community partner. I had difficulty developing a research question because the conversations that I shared with my community partner seemed circuitous. We talked around questions during several meetings before I was finally able to gain a sense of what she was hoping to achieve from the research. Though these past experiences with community-based research helped me to anticipate some of the issues that arose, I tried to make sure that the anticipation of issues did not create issues.

When entering into CBR projects, it is important to me that I am doing work that I view as meaningful. Work that is meaningful to me would be research that allows me to consistently interact with members of the community on a personal level. However, I tried to maintain the awareness that the research that I wished to pursue was not necessarily the research that the people I was collaborating with wished to pursue. I continued to remind myself that these discrepancies should not interfere with the development of a research design that was beneficial to my community partner and had the potential to bring about effective change. Since change is the goal of community-based research, I needed to be sure that the change I was assisting to create was the change that the community partner was seeking to make rather than the change that I would have liked to pursue.

Finally, when a researcher carries out a qualitative study, it is also important to attend to the subjectivities that the researcher brings based on gender, age, ethnicity, and socioeconomic status. I feel at times that I lack self-awareness of how these orientations impact the way that I view the world. Though I tried to be conscious of these factors while doing my research, I am not sure that I was successful in completely exploring how these subjectivities may have influenced my research. I do feel, however, that my status was an issue in the work that I conducted with the Coalition for Schools. My status in relation to my age (under 40) and my position as a graduate student influenced how my community partners at the Coalition viewed my role, and my socioeconomic background impacted the level of confidence that I felt when working with members of the Coalition. I come from a working class background while my community partners at the Coalition come from backgrounds of higher status both in relation to levels of education and socioeconomic status. At times, I did feel out of place moving through the world of the Coalition in that I often felt that I was from a lower class than many of the people with which I came into contact. I felt most comfortable when interacting with teachers or parents.

In order to minimize the impact of my subjectivities, I closely monitored my feelings as I carried out my research. I looked for situations where I felt uncomfortable or that I wanted to avoid as well as situations where I felt comfortable and that I wanted to continue. When these feelings arose, I realized that I was usually being influenced by subjectivity (Glesne, 1999; Peshkin, 1988). I analyzed my feelings and considered how they related to my subjectivities, then took note of these occurrences in my journal (Peshkin, 1988). Throughout the research process, I was mindful of previously identified subjectivities. I also tried to be aware of newly emerging subjectivities that I may not have considered (Peshkin, 1988) that would potentially influence my research.

This study seeks to compare two cases of conducting community-based research. However, there are differences between the two experiences that may have impacted the findings of the study. In my work with the Coalition, I was a paid employee. Though I was hired with the understanding that I would be a collaborative researcher, I believe my position as an employee impacted how Marge Bowline and Lisa Brown viewed my role, and it also impacted my reactions to various situations. The fact that I was an employee in the first case study when collaborating with the Coalition but in the second case study I was independent, may have created some of the differences that were apparent in the two cases.

Another limitation of this study is that it primarily focuses on the researcher's experience of this process. Though I did interview my community partners, the number of interviews in the first case study was more limited. If I had conducted additional interviews throughout the first case study, I might have additional information to support or contradict some of my observations. However, the purpose of this study is to provide insight into this process for practitioners in the field of community-based research, thus it is beneficial to explore the researcher's perspective of these two experiences.

The final limitation of this study relates to the timeline of the completion of the study. Since I only recently finalized data collection in relation to my work with John Brewer and Maria Swenson, I am not really able to make an assessment at this point as to whether any of the work we completed will affect change. My work with the Coalition was completed almost a year ago so it easier to assess the impact of that work. However, even with the first case study, there is a possibility that some of the work that I completed could eventually lead to change. If I were to conduct a long-term case study in relation to either of these collaborations, it would be more feasible to assess the impact of our work.

This chapter provided an overview to the case study methods that were used to conduct this study. I detailed a rationale for choosing this method, then described data collection, analysis, and procedures in relation to validity. Since this is a process study of the methodology of CBR, I also described the foundations of this methodology. The next three chapters will present the findings of this study. Chapters four and five provide a synopsis of the within-case analysis of each of the contrasting cases. I begin each chapter with a chronological overview of the major events of the case and then present within-case analysis organized around the four concepts of my analytic framework. In chapter six, I present the findings from the cross-case analysis that address the sub-questions of the study and identify the "naturalistic generalizations" (Creswell, 1998, p. 154) that emerged from the study with recommendations for further research.

Tinkler: Establishing a Conceptual Model	COMM-ORG Papers 2004	*http://comm-org.wisc.edu/papers.htm*
Contents \| Chapter 1 \| Chapter 2 \| Chapter 3 \| Chapter 4 \| Chapter 5 \| Chapter 6 \| Notes & References \| Appendices

Chapter 3: Research Methods