«Conceptualizing Pedagogical Content Knowledge from the Perspective of Experienced Secondary Science Teachers Committee: Julie A. Luft, Supervisor ...»
3. Roger, the only male participant in the present study, teaches Integrated Physics and Chemistry (IPC) and Geology, Meteorology, and Oceanography (GMO) in high school. He has ten years of teaching experience.
4. Emily has 16 years of experience teaching sixth and seventh grade science as a certified public school science teacher. She has a bachelor’s degree in Biology and Chemistry and a master’s degree in Integrated Science. She worked in middle school when I began to interview her and moved to high school one year after that. She is currently working as a high school science
The invited participants (teachers and program personnel) received a study description that included the purpose of the study and explained the possible risks and benefits of participation. Teachers who were interested in participating in the study were asked to submit their contact information, such as telephone numbers and email addresses.
A consent form was sent electronically to these participants for further review.
Participants were also given tentative interview schedules. The interview schedules were flexible, according to the availability of the participant. A specific timeline is shown in Figure 5. (Also see Appendix A for detailed research timeline.) The research aspect of this study is limited to examining the components of PCK and the specific elements within each component, based on the data from the four secondary science teachers who mentored beginning teachers. Data were generated for each participant in the following ways: three semi-structured interviews, two classroom observations, a collection of lesson plans, and monthly reflective summaries of participating teachers. Interviews were conducted on the days, times, and locations mutually agreed upon by the P.I. and study participants. Each participant was interviewed three times between November 1, 2003, and March 23, 2005. The first two interviews lasted no more than one to one and half hours, while the third interview took more than two hours for each participant. The interview protocols are attached in Appendix D, E, and F.
throughout the Spring semester, classroom observations were conducted during the Fall semester of 2004. Relevant documents (e.g., syllabi, lesson plans, handouts, monthly reflective summaries) were requested to supplement interview data. Monthly reflective summaries of participating teachers were utilized to better understand their perceptions of PCK.
After getting a list of possible participants in this study from the “Teachers as Mentors” project director, I contacted them individually by email. In the email, I included an online-form that allowed the participants to fill out their biographical information (see Appendix for the on-line form). I then met individually with each participant who indicated an interest in participating in the study. I explained the purpose of the study and the required time commitment. All seven teachers agreed to participate in the study and signed the Informed Consent Form (see Appendix B for the Informed Consent Form). At the request of the Internal Review Board, the Teachers as Mentors project director needed to give informed consent, as the research was taking place with the mentor teachers in the program (see Appendix C for the Informed Consent Form).
As the study proceeded, three of the participants who agreed to the study were dropped from the study one after the first interview, two after the second interview.
The reasons for their withdrawal were time constraints and the discomfort they felt at classroom observations. Therefore, the results of this study report based on the remaining four participants’ data.
Figure 5. Specific timeline of the study
There were three data sources: (1) interviews with mentor teachers and the program personnel, (2) class observations, and (3) supplementary documents (e.g., handouts, lesson plans, and reflective summary). Data from each source in this study were complementary and helped me obtain a holistic picture of teachers’ conceptualization of pedagogical content knowledge. The data sources complement each other in the following manner.
The main data source was interviews. The interview set consists of three individual interviews for each participant. Merriam (1998) classifies interviews by the
structured/standardized to semi-structured to unstructured /informal (p.73). The interviews that I conducted fell on the continuum between highly structured/ standardized to semi-structured. For my interview protocols, I did prepare a list of questions that I developed based on discussions between my supervisor and myself. The interview protocol served as a tool to enable me to visualize their internal conceptualization. For each participant, the interview data was transcribed. The transcribed data were analyzed and coded prior to the next interview.
In addition, interviews with program personnel helped me understand the purpose of the mentoring program and mentoring activities, as well as their expectations of the program.
As I began to identify categories after the first interview, I was able to generate a list of additional questions to refine emerging categories. Through the interviews with
participants, I expected to obtain the following information:
1. The nature and components of pedagogical content knowledge (PCK) within
In the first interview, biographical information was collected (see Appendix D for the first interview protocol).The second interview had two purposes: one was to further explore the knowledge components required to teach science, and the other was to clarify the observed teaching practice during classroom observations (see Appendix E for the second interview protocol). The third interview was conducted to provide participants with an opportunity to reflect on the results of the data analysis (see Appendix F for the third interview protocol). The participants were also asked to rate each component according to its importance in teaching science and to draw a concept map to represent their interrelationships.
Classroom observations were another data source for this study. By observing the participants teaching in their own classrooms, I was able to gain a better understanding of their teaching practices and the context in which they taught. And also, classroom observation also served as a source of interview questions. In the second interview, participants were asked to describe various activities observed during the classroom observations. Participants were asked not to make any changes in their teaching because of my being there. Certainly my presence in their classroom had some effect on their teaching, and more extensive classroom observations may have helped to address that concern.
Each participant was observed a minimum of two class periods for each class they taught. I arrived five to ten minutes prior to the start of the class period. Since I choose to take the role of classroom observer rather than participant, I usually sat in the back of the classroom. This decision enabled me to keep more detailed field notes. By actually observing teachers’ classroom practice and taking field notes, I was able to capture the details of how teachers act in their classrooms with respect to PCK.
I also collected lesson plans, project flyers, and monthly reflective summaries from each participant. From a collection of these supplementary materials, I was able to understand how PCK is represented in their lesson plans. In addition, examining the mentor teachers' monthly reflective summaries helped me to understand how the teachers’ conceptions of PCK emerge from their own reflections and how they utilize those conceptions of PCK in their mentoring practices.
The major unit of analysis was interview transcripts. I also considered field notes that I took during classroom observations, lesson plans for the observed classes and reflective summaries collected as a requirement for the mentoring as significant units, in addition to their interviews. Data analysis was based on the following process. Codes were formulated from the data in process and modified as the data collection proceeded.
Given that qualitative research is an open-ended and on-going process, once analysis of the collected data began, the procedures continued to move recursively through the process of constant-comparative analysis (Lincoln & Guba, 1985; Patton, 2002). In other words, it was difficult to separate analysis from interpretation because the two procedures were interwoven. Moreover, the data analysis of each interview provided a foundation for developing subsequent interview questions.
Three types of coding, as mentioned earlier, were sequentially conducted to analyze the data: open coding, axial coding, and selective coding (Strause and Corbin, 1998). In the following paragraphs, each coding procedure will be described in detail.
The graphic shown in Fig 6 displays the sequential procedures of analysis.
For open coding, the researcher relied on a qualitative analysis tool called QSR NVivo 2.0 (QSR International, 2002), a software application that allows a researcher to import transcripts as text, create codes (termed “nodes” in the program), and highlight and code pieces of text ranging from a few words to a complete transcript. This software was helpful for coding data visually. Figure 7 illustrates an example of the process of open coding using NVivo 2.0 on a small section of transcribed dialogue.
For naming elements and components, I initially put in vivo codes using NVivo2.0 program. After that, I examined a list of codes for each interview transcript, and then I named codes using more
terms. During this work, I was able to come up with broader, more comprehensive, and more abstract labels for the codes and groups of codes.
Figure 6. Sequential procedures for data collection and analysis Figure 7.
An example of the process of open coding using NVivo 2.0 Another source of code names is the literature. Since data analysis was initiated with the conceptualization of PCK for science teaching proposed by Magnusson, Krajcik, & Borko (1999, see Figure 8), their conceptual model of PCK provided a guideline for understanding PCK in the process of exploring teachers’ perspectives of PCK in this study. However, as tentative components representing recurring patterns of each teacher’s conception emerged, I continued to reshape and modify the categories over the course of the data analysis.
In an attempt to avoid using the same categories as those in Magnusson et. al.’s study (1999), the names for the categories came mostly from the list of concepts discovered in my data. Among the lists of concepts that emerged from the data analysis, ones that stood out as broader and more abstract than the others were used to denote categories.
Through the data analysis process, interestingly, seven common components emerged in all four cases. While the components and elements within each component were initially based on my interpretation of interview data and field notes, I kept reconstructing according to each participant’s input over the course of the study. Thus, it is, in part, a co-constructed interpretation between the participating teachers and myself. In the third interview, every teacher agreed upon the components of PCK, with some minor modifications of the elements within the components. The categories and elements that emerged from data analysis for this study will be described at length in Chapter 4.
Figure 8. Components of pedagogical content knowledge for science teaching.
( Magnusson, Krajcik, & Borko, 1999, p.99) Additionally, the interview transcripts and field notes were used to construct pictures of the teachers’ understanding of PCK through the axial coding process. This picture was incorporated into the initial diagram that each participant and I coconstructed during the third interview. We then modified it several times through email communication. During the modification process, I encouraged them to further develop the construct by adding linking words and explaining the relationships among the components.
The diagram was developed by a combination method of “card sort tasks” and “concept mapping” (Baxter and Leaderman, 1999). I showed them the components and elements that emerged from their data and asked them to weigh them according to their importance and to explain them. After that, I asked each participant to make connections among the components to show how they are interrelated within the context of teaching science. After getting a manually-created concept map, I converted it into an electronic version and sent it to each participant. The teachers was asked to check it and to modify it if necessary.
My study incorporated several techniques in order to meet the standards of validity for naturalistic inquiry. First, I achieved triangulation by employing multiple sources for collecting data, including (a) interviews with participants and project personnel, (b) observations of mentoring program activities, and (c) reflective summaries and supplementary materials. Not only would triangulation provide me the means for observing data that might have been overlooked by relying on only one source of data collection, but it would also allow me to see the same data from various perspectives and, in the process, to clarify the meaning of the data in its larger context. Second, I conducted a member check (Lincoln & Guba, 1985) to establish credibility, to clarify meaning, and to check the accuracy of my understanding of the data. Third, I had also discussed my on-going investigation with colleagues. Such discussions served the purpose of “peer debriefing” (Lincoln & Guba, 1985 p. 243).
This process was also helpful in developing interview questions and in developing and testing the emerging categories and subcategories.