«INDIVIDUALIZING ELEMENTARY GENERAL MUSIC INSTRUCTION: CASE STUDIES OF ASSESSMENT AND DIFFERENTIATION By Karen Salvador A DISSERTATION Submitted to ...»
…very diverse, both socioeconomically and racially... My building qualifies as a Title One school but is also situated in a very nice subdivision where the homes are valued at probably $300-500,000 and up. We have many nationalities/races represented in our student population, including many different languages spoken in the homes of our families. (Email communication; February 4, 2010).
All three participants taught in school districts that consistently ranked at the top of their state by many metrics. Each district achieved high ratings for its academic programs on the state report card, with strong test scores and much emphasis on college preparation, including offering numerous Advanced Placement (AP) courses. These districts had high graduation rates and high percentages of graduates who continued on for post-secondary education. By state law, students from other districts could choose to attend these schools if there was room; there was a waiting list for those slots each year in all three of these districts.
Data Collection Methods of data collection for this study comprised (1) field notes of observations, (2) videotape observation forms, (3) verbal protocol analysis of selected video excerpts, (4) teacher journals, and (5) interviews. I received human subjects approval from the Michigan State University Institutional Review Board. Although I videotaped each class that I observed, the tapes were of the teachers, and the tapes were not viewed by anyone other than the teacher in the video and me. My only known impact in the classroom was as an observer of typical general music instruction.
5 This designation indicates that about 40% or more of the families served in a school building qualify as “low-income” as described by the US Census (Elementary and Secondary Education Act, 2002).
6 To put this in perspective, in the fourth quarter of 2009, the median home value in Ms.
Stevens’s economically diverse, mostly suburban, county was about $130,000 and the urban county about 3 miles south of her school had median home values of about $92,000.
Observation. Naturalistic observation of elementary music classes was the primary data collection instrument in this study. My experience as a general music teacher and the knowledge of assessment and instruction that I have developed over the course of graduate study provided the lens through which I viewed each class. This is a typical practice in descriptive research (Creswell, 1998). I attempted to be as unobtrusive as possible in order to have the least impact, but I recognized that my presence in the classroom had the potential to change the classroom climate (Angrosino & Mays de Perez, 2000). Occasionally, students would check for my reaction to some event, or they would talk, sing, dance or play to me or to the camera. In general, students appeared accustomed to various adults coming in and out of the room and seemed to adjust quickly to my presence. In addition to videorecording each class, I took field notes on my computer as I observed; this is how I write most efficiently, and it facilitated data storage.
When I designed the study, it was my goal to spend 6 weeks observing every meeting of two classes taught by each participant teacher, for a total of 12 observations of each class.
Optimally, one class would be upper elementary and one lower. However, this goal was flexible to accommodate the needs of participant teachers as well as emergent issues. The observation period for Ms. Wheeler was from Jan 15-March 1, which resulted in a total of 11 30-minute observations of both a kindergarten and a fourth grade class. The classes that Ms. Wheeler selected for me to observe happened to fall on a Monday and a Friday, which resulted in several days with no school during the observation period: Martin Luther King, Jr. Day, and President’s Day. The students also had a Monday cancelled due to inclement weather, so the observation period was extended to 7 weeks and still did not reach the goal of 12 observations. Ms. Wheeler and I opted not to do a final make-up because the students were scheduled to miss the next two music days due to a mid-winter break.
I observed Ms. Stevens from February 4 to March 25. Ms. Stevens was ill for one observation day, her students had mid-winter break (resulting in one missed observation day), and school was cancelled on one observation day due to inclement weather. We persisted, and, over the course of 7 weeks, we were able to meet our goal of 12 40-minute observations of a first grade and a third grade class.
Observation for Ms. Davis was different, because we were nearing the end of the school year and because there was a unique opportunity in her setting to observe fourth grade students with cognitive impairments receive music instruction in both mainstreamed and self-contained settings. Therefore, I observed three classes—one third grade, one fourth grade, and one selfcontained class of upper elementary students with cognitive impairments each time they met from April 19 to May 26. Observations of individual classes were cancelled on several occasions due to field trips or assemblies, Ms. Davis was ill on one observation day, and I attended dress rehearsal for both the third and fourth grade end-of-the-year programs when these fell on observation days. This resulted in ten observations of each class as it met normally, plus observations of entire grade levels at dress rehearsals.
Videotaping. Each class I observed was also videotaped with a camera sitting on the desk near where I was taking notes. I would occasionally reposition the camera so that it was capturing the teacher as she moved around the room. The videotapes served two purposes. First, one week after each observation, I watched the videorecording and filled out a video response sheet (adapted from a sheet designed by Dr. Mitchell Robinson, based on examples in Miles and Huberman, 1984, pp. 53-55, see Appendix A). The video response sheet and the teacher’s journal for that class (see below) provided triangulation for my field notes. Videotapes were not transcribed for data analysis. Because the video data included singing, moments of classroom noise, unidentifiable voices, group work, and other extraneous or unintelligible information, I found it unlikely that transcription of every video would result in data that were more meaningful than field notes, teacher journals, video response sheets, and verbal protocol analysis.
Verbal protocol analysis (think alouds). The videotapes also provided brief excerpts to watch with the participant teacher for verbal protocol analysis (VPA). VPA, in which the participant is invited to pause a video and to describe what they were thinking as they were teaching or to reflect on what they are seeing in the video, is a method borrowed from psychological research traditions (Flinders and Richardson, 2002). This technique, also referred to as a “think-aloud,” can provide valuable information on the practices of teachers “in the moment.” Video excerpts for VPA were selected by the researcher and comprised segments of teaching when the participant seemed to be delivering instruction based on the needs of individual students. Each session of VPA was audio-recorded and transcribed for inclusion in data analysis. Ms. Wheeler and Ms. Stevens both participated in two sessions, one about four weeks into observations, and the other after observations were completed. These sessions lasted 35 to 50 minutes. Due to her shorter observation period, Ms. Davis had a single session of VPA that lasted nearly two hours.
Journals. After each meeting of the targeted classes, the teacher completed a journal
entry and emailed it to me. Each journal entry was based on the following questions:
(1) What opportunities for individual or small group response did you give, and what interested you in the students’ responses?
(2) How did you keep track of what individual students know and can do?
(3) How and when did you deviate from your plans in order to individualize instruction?
(4) How will what you learned today about what your students can do affect your instructional planning?
Teachers chose to answer the questions that were most applicable to the class they were describing and could also add comments unrelated to the questions if they wished. In addition, I occasionally asked them to comment on specific behavior I had witnessed or to comment on something we had discussed in the moment. Journal entries had several functions as data in this study. First, they were a source of triangulation—the teachers could present their thoughts about their teaching to enrich what I observed. Second, the journals offered the teachers a chance to reflect on their practice. Finally, the journals informed what video clips were chosen for verbal protocol analysis and suggested questions to be asked during final interviews.
Interviews. In addition to soliciting the teachers’ thoughts through verbal protocol analysis and teacher journals, I also interviewed the teachers prior to beginning classroom observations and after data collection was complete. Initial interviews followed a semistructured interview protocol, guided by a list of questions (see Appendix B) and supplemented by additional questions to clarify responses or to investigate interesting statements. The initial interview informed my observations and my interpretations of the teachers’ journaling. General interview topics included: (1) the school setting (demographics, other topics of interest) (2) the teacher’s views on assessment and individualization of instruction, (3) what kinds of assessment had already taken place in the classes I was about to observe, and (4) the music learning goals the teacher was working on while I was observing. In this interview, participants also were given the opportunity to ask me any questions that they may have had about this study.
The exit interview took place several weeks after the completion of the observations for all three participants. By this time, each teacher had performed a member check on the transcript of their initial interviews and both think-alouds. In addition, I had already performed preliminary analysis for themes within and across cases, based on all the information collected so far (preliminary interviews, my field notes, teacher journals, verbal protocol analysis, and video response sheets). The exit interview questions were derived from this preliminary analysis and were intended to allow the teacher to share her opinions of the credibility of my findings (Appendix C). This was an important part of the research design, because the results should “ring true” to the participants and, if they did not, it was important to know why. I also asked how the act of being studied (including journaling and verbal protocol analysis) affected the teacher’s pedagogy and/or thoughts about assessment. The exit interview allowed me to refine my initial themes in conference with the teacher(s) to whom they applied.
Trustworthiness/Credibility This study attempted to reveal experiences of public school elementary music teachers as they used information gleaned from assessments to help individual students progress musically.
The study was only successful to the degree that it described these interactions in a manner that seemed meaningful and authentic to the reader. In order to ensure the trustworthiness or credibility of my data, I used several techniques. First, I used multiple sources of data, including observation field notes, teacher journals, video, and interviews. These various forms of information and the viewpoints they represented allowed for triangulation of data. That is, the sources were checked against one another to bolster credibility (Miles and Huberman, 1984). In addition, transcriptions of interviews were returned to the participant for “member checks,” in which a participant could ensure that her thoughts were accurately portrayed by editing or adding to the transcript (Janesick, 2000, p. 393). Each participant also was asked to comment on the credibility of my initial data analysis as a further member check. Finally, preliminary findings and entire case studies were submitted for peer review by faculty members and fellow doctoral students at Michigan State University. The combination of triangulation, member checks, and peer review should enhance the trustworthiness of findings.
Limitations This qualitative case study took place in three specific settings taught by three individual teachers. While the settings differed from one another, they each were associated with mediumto-large, suburban school districts in the Midwest. Due to the qualitative nature of this project, I did not attempt to find any kind of “sample” that might be construed to be widely representative of any group. Instead, I purposefully chose the participant teachers and settings based on recommendations by leaders in music education in an attempt to study promising practices.
Because other elementary music settings and teachers differ from those described in this study, it would be inappropriate to expect that the results of this study could necessarily be generalized to other settings. However, perhaps teachers could adapt or modify ideas illuminated by this research for use in their own classrooms. Information from qualitative studies may be transferrable to similar situations (Creswell, 1998), and the results that resonate with particular teachers may be appropriated.
I did not describe or evaluate the curricula being assessed, except as this information directly informed this investigation into assessment and differentiation of instruction. While some of this information may be apparent to the reader as I describe instructional practice and how the results of an assessment were used, a discussion of the relative merits of various curricular goals was beyond the scope of this paper. Participant teachers in this study came from different educational backgrounds and used a variety of methodologies. Similar to curriculum and assessment methods, the methodologies being used in these classrooms may become apparent to the reader, but I did not set out to discuss the relative merits of methodological approaches.