© Please note: Rhoda Summers owns the copyright for the dissertation below. You may download and print one copy for educational purposes only. These pages may not be duplicated, distributed, redistributed or republished in any manner without express permission from the author.
The Skin Horse had lived longer in the nursery than any of the others. . . . He had seen a long succession of mechanical toys arrive . . . and pass away. . . . He knew they were only toys, and would never turn into anything else. For nursery magic is very strange and wonderful, and only those playthings that are old and wise and experienced like the Skin Horse understand all about it.
Margery Williams, The Velveteen Rabbit
I used a qualitative case study design for this study since it would allow me to draw from the wisdom and insight of those who have experienced and understand the myriad facets of teaching middle schoolers. This design allowed me to listen to the stories of those who have "lived [long] in the [middle school classroom]" (Williams, 1981, p. 12) and have blazed a trail for others to follow in the quest for excellence in middle school education.
Merriam (1988) defines a qualitative case study as "an intensive, holistic description and analysis of a single entity, phenomenon, or social unit. [They] are particularistic, descriptive, and heuristic and rely heavily on inductive reasoning" (p. 16). A case study is particularistic because it focuses on a specific phenomenon such as a program, event, process, person, institution, or group. When Merriam states that a case study is descriptive,
she is referring to the end product of the study which is a "rich, thick
description of the phenomenon under study" (p. 11). Heuristic refers to a case
study’s power to "illuminate the reader’s understanding of the phenomenon under study" (p. 13), and Merriam (1988), quoting from Stake, says "previously unknown relationships and variables can emerge from case studies leading to a rethinking of the phenomenon being studied" (p. 13). Case studies utilize inductive reasoning since new understandings, concepts, and relationships arise from studying the data (Merriam, 1988).
According to Yin (1994), a case study "investigates a contemporary phenomenon within its real-life context, especially when the boundaries between phenomenon and context are not clearly evident" (p. 13). Using this method is appropriate when contextual conditions impact the phenomenon under study. Yin also notes that a case study relies on multiple sources of data collection, triangulation of data, and benefits from prior development of a theoretical framework which guides data collection and analysis.
Though Tellis (1997) does not specifically define case studies, he introduces another idea by saying that a case study "is done in a way that incorporates the views of the ‘actors’ in the case under study" (p. 2). The voices of the participants in the research study are included in reporting the research rather than only the voice of the researcher.
Drawing from Merriam (1988), Yin (1994), and Tellis (1997), my study fell well within a qualitative case study design. It focused on four individual middle school teachers as I sought to describe their beliefs, practices, and lives. I used inductive reasoning as I analyzed the data, searching for relationships and themes. Yin (1994) stresses that a case study occurs "within its real-life context" (p. 13) and this was an important emphasis in my study. I not only listened to the voices of the teachers participating in my study as Tellis suggests, but I also observed in their classrooms to gain an understanding of the context within which they teach. As Yin suggests, I developed a theoretical framework through my review of the literature, but I chose not to use it while collecting data so I could, as Clandinin and Connelly (1994) suggest, be "open to a rich and sometimes seemingly endless range of possible events and stories and . . . be prepared to follow leads in many directions" (p. 417). Consequently, my research study and design were emergent.
Within the qualitative case study design, I used narrative inquiry. Connelly and Clandinin (1990) state that "humans are storytelling organisms who, individually and socially, lead storied lives" (p. 2). Connelly and Clandinin (1995a) specifically connect teachers and the use of narrative: "Teachers know their lives in terms of stories. They live stories, tell stories of those lives, retell stories with changed possibilities, and relive the changed stories" (p. 12). Mattingly (1991) reiterates this idea when she says that "storytelling is something that practitioners already do" (p. 255).
By listening to teachers’ stories I was able to share in their experiences and learn about the deeply held beliefs which inform their lives and teaching practices. At times it may be difficult to clearly articulate one’s theories, but teachers tell stories of their experiences within which are hidden their beliefs (Mattingly, 1991). Schmidt (1997) believes that the "stories of [teachers’] experiences constitute [their] authority and inform what [they] do in [their] classrooms" (p. 169). Narrative is the best way to understand teachers’ knowledge (Connelly & Clandinin, 1995a), and it gives us "insight into what motivates a teacher’s actions" (Sparks-Langer & Colton, 1991, p. 42).
Narrative inquiry was an appropriate tool to use in my study. In order to describe the beliefs of effective middle school teachers and what has influenced them to become effective teachers, I first needed to understand those deeply held, often implicit beliefs and experiences which precipitate teachers’ actions and decisions. This understanding came through listening to and analyzing the teachers’ stories.
A major task was to find effective middle school teachers willing to give of their time and themselves to participate in my study. I wanted the participants to be in the Michiana area, yet I wanted to work with teachers that not only I believed were effective, but ones whom others also recognized as effective middle school teachers. Further selection criteria included middle school teachers with more than 5 years of experience, and I wanted one male and one female teaching in two different public schools and one male and one female teaching in two different Christian schools.
These boundaries (Miles & Huberman, 1994) led me to follow Maxwell’s (1996) suggestion of using purposeful sampling when persons are "selected deliberately in order to provide important information that [cannot] be gotten as well from other choices" (p. 70). I combined my purposeful sampling with reputational selection, or participants "chosen on the recommendation of an ‘expert’ or ‘key informant’" (Miles & Huberman, 1994, p. 28). I drew from others’ expertise in choosing participants because they, like the Skin Horse, have had experiences I have not been privy to.
I used reputational selection twice in the process of identifying effective middle school teachers. My first step was to contact professors at Michiana colleges and universities who have direct contact with middle school principals and teachers, requesting them to identify schools where they have met effective middle school teachers or schools which they believe have effective middle school programs. I phrased my inquiry so that these experienced educators would use their own criteria in nominating schools or specific teachers. From these conversations I formed a list of approximately 12 public and Christian schools. One school I added to the list as a result of my own reading. George and Alexander (1993) included a roster of middle schools which have "become known because of their level of excellence to researchers and practitioners on a local, state, national, or international basis" (p. 535), and one of the schools on the roster is in the Michiana area.
The second step in my search was to reduce this list of schools to four public and four Christian schools based on a number of factors: (1) Proximity to Andrews University, (2) reputation for excellence in Michiana, and (3) previous involvement with Andrews University. I then contacted the eight principals via letter and asked them to select teachers whom they consider effective. Principals used their own criteria in nominating effective teachers.
After the principals selected effective teachers in their schools, each teacher was given a questionnaire concerning his or her beliefs and practices. This questionnaire, designed by Smith (1992), was "found to be statistically valid and reliable in discriminating among teachers whose self-reported knowledge, practices, attitudes, and beliefs [are] or [are] not aligned with the findings about effective middle school teachers in a review of the literature" (p. 108). From the returned questionnaires, I contacted four teachers who had expressed interest in participating in my study. These four effective middle school teachers committed themselves to journey with me through my study.
Data were collected and analyzed in three phases. Phase one began when I contacted professors and principals for their recommendations concerning participants for my study. During this time I also conducted semi-structured interviews with each principal to discover his or her professional training, educational experiences, basic information about the school, and beliefs concerning characteristics of effective middle school teachers.
After conducting semi-structured interviews with the eight principals and asking them to discuss criteria they used to identify effective middle school teachers in their schools, I compiled a list of characteristics from their criteria. I then placed these in six different categories, looking for similarities among the characteristics. I concluded my task by naming the categories: Personality traits, classroom atmosphere, understanding middle schoolers, enjoyment of students, curriculum and instruction, and communication with stakeholders.
Of the 38 questionnaires I distributed, 30 of them were completed and returned. After analyzing the returned questionnaires, I targeted the 17 questionnaires completed by teachers who indicated an interest in participating further in my study. I divided these into four categories: Male teachers in public schools, male teachers in Christian schools, female teachers in public schools, and female teachers in Christian schools. I then contacted the four teachers whose questionnaires indicated that they were the most effective middle school teachers in their specific category. All four were enthusiastic about participating in the next phase of my study.
Yin (1994) describes six sources of data used in qualitative case study research: Documentation, archival records, interviews, direct observation, participant observation, and physical artifacts. As part of observation, Merriam (1988) suggests completing accurate field notes which include both actual occurrences as well as the observer’s comments or interpretations. Of the data collection sources mentioned, I relied on classroom observations, interviews, and field notes. I also collected some documents such as student handouts, assessment instruments, and letters.
Since data collection consisted largely of observations, interviews, and field notes, all of which were directly influenced by me, I became the primary instrument of data collection and analysis. Being the primary instrument allowed me to view the context within which my research phenomenon occurred. This gave me freedom to clarify and summarize while collecting data and to pursue new ideas and lines of thought. A degree of data analysis occurred simultaneously with collection and allowed for member checks to enhance the trustworthiness of my interpretations (Merriam, 1988).
I conducted the more formal phase of analysis by coding the interview transcripts and field notes. My initial coding categories consisted of significant life experiences, classroom practices and atmosphere, beliefs, and motivation. Later as I spent more time writing about the teachers’ lives and classroom practices, I revised my coding categories and called them life experiences, classroom practices, and beliefs. For me, this was largely an intuitive process, but I also, as Merriam (1988) stresses, considered issues such as frequency, uniqueness, and previously unrecognized areas. After I had identified the overarching categories for my data, I sorted through each category and looked for ways to break the data down into manageable pieces which fit together. Each teacher’s story is unique, which required that each chapter I devoted to him or her be unique. My goal was to have each chapter reflect the teacher’s lived experiences, classroom practices, and beliefs.
After completing chapters 3-6, I presented a copy to each teacher and asked him or her to review it. I also gave each teacher an open-ended questionnaire (Yin, 1994) to provide direction for member checks (Merriam, 1988). A final interview was conducted with each one to discuss the accuracy of my portrayal, needed corrections, and suggested additions and items to delete. Based on these final interviews, I revised each teacher’s chapter.
I began my study with broad categories of questions concerning the stories of each participant. I wanted to hear about their lives as children, academic preparation, and circumstances surrounding various teaching positions during their careers. My questions became more focused as the study emerged. Though I knew when I began my study that I wanted participants to draw a life map of significant events, I waited until the final months to make this request. Even then I was apprehensive about voicing my desire because of the walls I often build around the stories which constitute my life, but each teacher willingly and openly shared his or her life map.
Prior to each interview, I prepared a list of questions though each conversation took a life of its own. Many of the questions came from classroom observations or comments made during earlier conversations; these questions often led to the telling of a story as explanation for a given practice or belief. In many ways I felt like a sleuth, searching for experiences which have shaped each participant. I worked hard to be a considerate sleuth who treated these teachers with respect and sensitivity because we sometimes discussed less than pleasant experiences in their lives. Though Jack and I laughed after his comment that "This is like psychotherapy" (Vol. II, p. 128), the parallel seemed accurate as I probed for explanations and clarification.
Instead of addressing issues of validity and reliability, qualitative researchers use terms such as trustworthiness, credibility, dependability, and consistency. Merriam (1988) suggests that these are enhanced through member checks, triangulation of data and methods, an audit trail, and stating researcher biases.
During data collection and analysis, I used member checks by asking for clarification during conversations with the four participants. Each of the teachers reviewed his or her own chapter, suggesting corrections, additions, and commenting on interpretations. By using interviews, observations, and collecting documents, allowing for triangulation of data and methods, I was able to corroborate what I saw emerging as each teacher’s beliefs and practices.
By describing the steps taken during data collection and analysis, I am creating an audit trail. Each researcher’s biases, past experiences, and implicit thoughts impact his or her research interpretations (Eisner, 1998); however, another researcher would be able to use the trail I am leaving behind as an "operating manual" to recreate my study (Goetz & LeCompte, 1984, p. 216). Part of my audit trail includes an organized research notebook of data collected from each of my four participants, and the "narrative . . . report [in chapters 3-6 are] a history of the research project" (Polkinghorne, 1997, p. 15).
The relationship that I as the researcher negotiated with each teacher participant varied (Maxwell, 1996). My face-to-face conversations began in November and continued through June of the following year. The number of classroom observations ranged in number from five to seven, but some visits were longer than others due to each teacher’s schedule, classroom activities, and my personal constraints. I conducted from four to six interviews with the teachers. The length of the interviews varied, the time spent conversing following the interviews varied, and with two participants I also had significant telephone conversations which I included in my field notes. By the end of data collection, I felt like each participant had become a friend; in each case I viewed them as my personal "teachers" as I sought to understand their lives and settings (Spradley, 1979, p. 25).
During my study I found myself periodically going down "rabbit trails" concerning possible conclusions about the characteristics and beliefs of effective middle school teachers. At times I discussed these with one or more of my participants and other educators, but as I returned to the data I realized that these interpretations were premature and not consistent with the experiences and beliefs of all four participants.
Merriam (1988) writes that "one selects a case study approach because one wishes to understand the particular in depth" (p. 173). The purpose of my study was to describe the experiences, practices, and beliefs of individual, effective middle school teachers. It was important to me that I understand enough of their lives and characteristics that I be able to re-present (Glesne, 1997) them for other educators to read and restory their own lives (Clandinin & Connelly, 1991). Patton says that qualitative research should "provide perspective" (as cited in Merriam, 1988, p. 175) rather than identify truths reflective of a vast number of individuals. In my study I was able to, according to the participants, provide an accurate perspective of their experiences, practices, and beliefs. I have provided readers with a description of four effective middle school teachers that "they would have missed without [my] observations" (Eisner, 1998, p. 114).
In the chapters following this one, I provide a "rich, thick description" (Merriam, 1988, p. 177) so that those who read the stories of these teachers will be able to evaluate the degree of transferability to their own settings. I do not expect all other effective middle school teachers to "share identical . . . features but rather that these are features one might look for in other [effective middle school teachers]" (Eisner, 1998, p. 103). My goal has been to articulate one perspective of effective middle school teachers and disseminate it for others to consider and find applicable truths.
I have already mentioned that I, as the researcher, was the primary instrument of data collection throughout the study and would like to make some of my biases explicit. I have spent time as a middle school teacher and have given leadership to implementing the middle school philosophy at the school where I was teaching. I believe middle school instruction and curriculum should be student-centered and teachers need to build relationships with middle schoolers. From my personal exposure to middle schools and the literature, I believe that schools which have adopted and implemented the middle school philosophy are more effective in reaching students than schools closer to the junior high-school end of the continuum.
Merriam (1988) sees tolerance for ambiguity, good communication skills, and sensitivity to context, data, and personal bias as characteristics needed by the researcher as the primary instrument. My tolerance for ambiguity has grown considerably through my graduate studies and I enjoy the challenge of a less-structured process which allows me the freedom to "search for pieces to the puzzle" (Merriam, 1988, p. 37). I consider myself a good communicator and have had this verified by those around me. I find it relatively easy to establish rapport with others and to listen carefully.
Yin (1994) also discusses commonly required skills for case study researchers. Some of his ideas mirror those discussed by Merriam; skills he adds are the ability to ask good questions, being a good listener, having a firm grasp of the issues being studied, and the ability to remain unbiased by preconceived notions (Yin, 1994, p. 56). Several of the participants indicated that they found me to be a good listener during our conversations, and as I reviewed transcripts of our interviews, I noticed that the amount of words recorded for the questions I asked are minute compared to the length of the answers given. This to me is indicative that I was both a good listener and able to ask appropriate questions.
Understanding the issues surrounding middle school education has occurred through both my personal experiences, review of the literature, and conversations with other educators. I liken this to Eisner’s (1998) concept of connoisseurship which he describes as "the means through which we come to know the complexities, nuances, and subtleties of aspects of the world in which we have a special interest" (p. 68). My past experiences as a middle school teacher and observer in middle schools allowed me to recognize the practices of effective middle school teachers. I see this as an asset rather than a hindrance. I have seen both effective and ineffective instruction in middle schools which allowed me to compare the responses of students, non-verbal messages, and classroom atmospheres of the four classrooms I observed with those of less-effective middle school classrooms I have visited.
Yin’s (1994) comments about a researcher’s need to be willing to set aside preconceived ideas deserve special attention. I did submit a partially completed theoretical framework with the proposal of my study, and one of my concerns was that it not blind my eyes to seeing and hearing things which were outside or contrary to the framework. Consequently, I laid aside the framework and refused to review it until I was ready for the cross-case analysis. At one point during the data collection phase of my study, another educator questioned me about the theoretical framework and I was able to articulate only its broad concepts. Though the incident was slightly embarrassing, I also experienced a sense of relief that I had been able to remove its contents from the forefront of my thoughts.
Ethical considerations in my study include confidentiality of data, anonymity of participants and sites, and informed consent. The middle school teachers who participated in the interviews and opened up their classrooms for observations, as well as their principals, were given an abstract of the intended research study and a consent form explaining the research protocol. During my initial interviews with the participants, I alluded to Eisner’s (1998) comments about the fallibility of informed consent; that "researchers usually do not know what will emerge . . . and therefore are not in a good position to inform those to be observed about what to expect" (p. 215).
The method used in this study was a qualitative case study design. I relied both on the wisdom and insight of expert informants and a questionnaire to find effective middle school teachers to participate in my study. Data collection included classroom observations, interviews, and document collection during the course of 1 school year. Participants provided feedback concerning the interpretations of the study. The results of data analysis and the participants’ journey toward becoming Real constitute the remainder of this document.