Engaging the Whole Musical Self Through Expert Music Analysis Processes

Materials of Music III (.pdf) is the third course in a four-semester core music theory and aural skills sequence that is required for all music majors. This course builds on prior skills and content and asks students to take that work deeper as they gain more fluency. For some students, this is the last music theory class they will take; for others, this course is a pre-requisite for upper-level classes in analysis and composition. At my institution, music theory classes are taught as two 100-minute classes a week with no tutorials, and with class sizes ranging from 9-15 students.

A stereotypical music theory class has a set list of typical activities that tend to stay fixed while the content changes or deepens. These include:

• writing single examples of each concept using music notation
• writing passages of music using examples of each concept, following specific guidelines for how to connect one concept to the next
• recognizing each concept visually in musical scores and aurally in musical pieces
• singing examples of each concept
• playing examples of each concept at the piano

Often, students practice these skills on homework or other out-of-class activities; however, I practice a flipped model of teaching. I divide my class time into smaller chunks to engage with the various tasks described above. A typical class period might be structured as follows:

• review of returned assignments and/or quizzes with discussion of common difficulties
• short lecture to expand on a current topic or introduce a new one
• individual or small-group practice on topic through music writing (sometimes at pianos)
• individual quiz to test mastery on a previous concept
• analysis of a piece or music passage as a full class or in small groups, through listening and score study

My classes use a variety of assessment tools with various grade weights.

What most music theory classes share is a tension between two types of activities and goals: those that address mastery of content and improvement of musical literacy (that is, the ability to read, write, hear, and play musical concepts quickly and accurately) and those that address the application of content and literacy to solve a problem or defend an argument. Both are important, but, returning to my specific target course, I see this class as the place in the overall music theory sequence where my expectations of student learning changes. Mastery of content and basic musical literacy is now assumed to be primarily the responsibility of the student. My focus in this course is to deepen student learning by encouraging critical thinking and application to musical literature from a range of genres.

Music analysis thus became my primary focus in this course redesign, since it synthesizes content and encourages critical thinking and creativity. It also involves skills of prose writing and oral presentation, if analysts choose to share their work. I also had a hunch that good music analysis could involve many of the skills that music theory classes aim to improve: score study, performance, listening, and, to a lesser extent, composition.

My implementation involved three stages:

• preparatory work to determine the processes involved in an expert music analysis
• revisions to in-class activities in my target course
• revisions to out-of-class activities in my target course

Pre-Work: What is an Expert Music Analysis?

My statement that music analysis synthesizes numerous musical skills is a new discovery. I learned this by using the Decoding the Disciplines framework, developed by David Pace and Joan Middendorf. Decoding the Disciplines is “a process for increasing student learning by narrowing the gap between expert and novice thinking.”

The full Decoding process is a seven-step framework to help teachers identify and address difficulties in student learning across a range of disciplines. Step 1, “define a bottleneck,” encourages instructors to find a place in their course where students typically struggle. For me, this was music analysis.

The second step in the Decoding process involves inquiry into the behaviors conducted by experts in order to overcome that bottleneck. The investigator thinks deeply about the skills and strategies that s/he uses as an expert, often by participating in a “Decoding Interview” (a sample interview is available here). I had a thirty-minute conversation with a colleague from another discipline about what I, as an expert music analyst, do when I analyze music. As a result of this conversation and subsequent reflection, I came to the following conclusions.

An expert music analyst, when presented with a piece of music, will typically do three activities:

1. listen to the piece
2. perform the piece in some way: traditionally, by sitting at the piano and playing from a score
3. compare what s/he hears in listening/playing with what s/he expects to hear (comparing the music with learned conventions)

       Through repeated passes through activities 1-3 above, the analyst will form theories about how this piece of music conforms to convention, or how it changes or challenges convention in some way. Then, the analyst returns to activities 1-3, but this time with an aim to find evidence to support her/his developing theories.

       Next, the analyst will reflect on whether s/he thinks the theory is valid, and consider its strengths and weaknesses. This reflection will happen through the preparation of formal findings for presentation/publication (that is, through the activity of writing), and through further passes through activities 1-3.

       Finally, the analyst will share her/his theory with other experts, and respond to feedback from those experts.

I discovered that analysis involved all the musical skills that were goals of my course: listening, performance, score study, and composition. Music analysis also synthesizes content and promotes creativity, critical thinking, and the articulation of ideas in prose. Expert music theorists bring their “whole musical selves” into the act of analysis: their listening ears, performing bodies, and critical minds. But since these processes are normally hidden in the final, written, analytical product (even at the professional level), I wondered if my students were using their whole musical selves in analysis, and if I was teaching them to do so. These two questions would form the basis of my CHRP project.

Revisions to In-Class Activities

I focused on four areas where I could redesign my course to better teach students an expert process of analysis in my course. Two redesigns were for in-class activities and two were out-of-class activities.

My first in-class focus was the way I modeled the expert process of music analysis for students. Through analysis of my lecture notes (.pdf) from the start of the fall 2015 iteration of the course (see a sample here), I learned that I mostly emphasized the act of looking at the score, although I sometimes mentioned or demonstrated listening and performing (and often ad libbed these elements in class). I rarely, if ever, mentioned revision of prior analytical work. Over the 2015 iteration of the course I worked informally to improve my use of listening, playing, and analytical revision.

In the 2016 iteration of the course, I continued this informal emphasis, but also deliberately planned several points across the semester to model expert analytical processes:

• a lecture (.pdf) early in the semester that demonstrated the “whole musical self” expert process of analysis
• a class activity (.pdf)in week 4 where students began an analysis unguided by me, but where I deliberately modeled expert processes when I assisted them
• in conjunction with the first Writing for Analysis assignment (described below), gave a demonstration (.pdf) of how an expert would work with a piece of music to find a thesis
• a class activity (.pdf) in week 11 where students prepared an analysis for homework, and then in class used expert processes to evaluate that analysis and search for deeper meaning

My second in-class focus was to adapt and introduce more activities that would allow students to practice expert processes in a low- or no-stakes environment. A sample from fall 2014 (.pdf) from fall 2014 shows that I already do this in my teaching, but what is missing is drawing students’ attention to the processes as useful for the future (and which in many cases they’re already using). In 2015, I gave students a preparatory assignment (.pdf) for the following class that was given a completion grade. I also designed in-class activities that would encourage students to use multiple musical modalities, and to revise their work before submitting (example activity (.pdf) from the end of the semester).

In the 2016 iteration of this class, I made student awareness of process even more explicit. In addition to the modeling just discussed, and informal comments throughout the semester, assignments at the beginning and end of the semester drew students’ attention to their own processes for music analysis. In the first class period I asked students to discuss how they analyzed music and recorded the results (.pdf), before sharing with them the purpose of my research. In the final class, I asked students to write advice (.pdf) to next year’s students about the best methods for music analysis. I continued to take informal opportunities to point out moments when students were (or were not) using revision, listening, and performance as part of their analytical processes.

Revisions to Out-of-Class Activities

My first out-of-class focus was to change the standard template I use for short music analysis homework. Prior to this redesign, I might have written an analysis assignment like this:

Prepare a formal analysis of Beethoven’s Sonatina in F Major, Op. Posthumous, second movement (p.99 in your anthology). Draw phrase structure diagrams and label phrase types as specifically as possible.

Given my current focus, I see now that this assignment emphasizes the product of analysis (diagrams of phrase structure and labels of phrase types) but not the process. Additionally, it does not remind students to listen to and play the piece.

In an analysis assignment (.pdf) from Fall 2015, I deliberately didn’t prompt students to listen or play, because at this point in my project I was trying to assess student process rather than change it. However, there are more cues for students at moments when they might have difficulty with the analysis (question #1). The assignment also encourages them to reflect on what they already know (question #2), and back up opinions with evidence (question #3). Question #4 was meant to be a data-collection question for me, but it could also have given students a chance to reflect on what techniques they used in analysis.

A further redesign (.pdf) in Fall 2016 gives students specific directions about when and how to listen to the piece. Again, the third page was mostly for data collection but had the added benefit of prompting students to reflect on their own analytical process.

My second out-of-class focus was to modify the way I prepared students for the “Writing for Analysis” (WA) assignments. These assignments (.pdf) represent the highest-order thinking required of my students, since they must synthesize prior concepts to analyze music, and then write a short analytical/argumentative paper in disciplinary style that has an original thesis statement and supporting evidence.

In 2015, I had a grant to work with a student Writing Center Fellow from my university’s Writing Center to improve student writing in my course. I assigned WA1 (.pdf) before being matched with a Fellow, but was able to redesign both the process and assignment itself for WA2 (.pdf) and WA3 with her help.

The most striking changes are from WA1 to WA2. While each assignment asks students to combine music analysis and writing, in WA1 I maintained an artificial separation between the two, while in WA2 writing and analysis are linked, reflecting at least one expert’s process of analytical writing (my own). At the same time, WA2 gives students more guidance in their analysis, and suggests strategies linked to the expert process of analysis that I wish to teach (listening, playing, transcription, music writing, comparison of new piece with present knowledge).

Finally, the way I used class time between WA1 and WA2 changed. In WA1, we spent some class time together discussing the blues form and completing the chart in groups. I left them to complete their drafts on their own, gave them feedback on drafts by email, and then collected final versions.

For WA2, I added several new elements:

• delivered a ‘think aloud’ lecture where I demonstrated how to approach an analysis and develop a thesis statement for a specific song
• asked the Writing Center Fellow to deliver a presentation on how to give effective peer feedback
• asked students to write up a thesis statement and supporting evidence; students gave online peer feedback
• gave students feedback on a draft, via email
• met with Writing Center Consultant to give my overall impressions of drafts
• had students meet with Writing Center Consultant to revise draft
• students submitted final version on paper

In the fall 2016 iteration of this course, I will no longer have a Writing Center Consultant. I kept the redesign of the assignment itself, while changing the topics. In terms of how I taught the skill of completing such a high-level analysis, I used the pedagogy above for WA1 (.pdf) rather than WA2, and modified the order as follows:

• delivered a ‘think aloud’ lecture where I demonstrated how to approach an analysis and develop a thesis statement for a specific song
• asked students to write up a thesis statement and supporting evidence; students gave in-class feedback and revised their work
• gave students feedback on a draft, via email
• met with students individually to discuss feedback
• encouraged students to meet with general Writing Center Consultants to revise drafts
• students submitted final version on paper

For WA2 and WA3, I continued to discuss the assignment in class and model certain aspects for students. Only the draft and final submissions were required; the rest was optional.

I collected and analyzed student work in three ways:

1. experimented with methods for data collection of student processes of musical analysis
2. explored the potential link between use of expert processes of analysis and higher quality musical analysis
3. analyzed student performance on Writing for Analysis assignments

Data on Student Process of Analysis

Figuring out how to collect data on student processes of analysis was difficult because those processes often happened out of class, and because students were not self-aware about the expert processes they were using. (It should be noted that most experts are not self-aware, either.)

In Fall 2015 I asked students to self-report their analytical strategies using ‘process reports’ where they wrote down everything they did to complete an analysis assignment, and then submitted the report anonymously. I struggled with difficulties including student forgetfulness and lack of self-awareness (as detailed in my Year 2 portfolio (.pdf), but I did glean some useful information. In process reports (.pdf) collected in fall 2015, I saw some evidence that students were using their whole musical selves in analysis. I also noticed that students were focusing on what I considered to be lower-order analytical strategies, indicating a lack of mastery of concepts from prior semesters. This question of musical literacy is a problem I continue to see in student work, and its implications are addressed in the reflection section of this portfolio.

More helpful in Fall 2015 was direct observation of students working alone or in groups in class. Not only could I observe processes that students were as-yet unaware of, I could also highlight expert processes used while students were working, and use the experience as an opportunity for student learning. I eventually integrated this into homework assignments where student groups were given musical excerpts for homework and then gave presentations that included both their analysis and a discussion of the process they used to evaluate alternatives. In the class presentations, I took notes (.pdf) on their reported processes, but also asked guided questions and highlighted expert processes for students that they themselves had suggested, empowering and validating students as experts themselves.

In my Fall 2016 iteration I abandoned process reports and developed strategies that would help benchmark student use of expert processes at the start and end of the semester. The activities are summarized below, with links to the data:

• first class - asked students ‘how do you do music analysis’ (.pdf)
• second class - observed student processes (.pdf) on an in-class analysis assignment then designed an assignment that included reflection (.pdf) on process.
• last class - asked students ‘give advice to next year’s MUS 211 students’

Data on the Link Between Expert Process and Expert Results

In the first two iterations of my target course, I struggled to find data that could support my hypothesis that an expert process of analysis would lead to expert results (or at least, better analyses). In part, this is because there are so many variables. For instance, in the fall 2015 iteration of the course, when I assigned students to present on both their analysis and their process I found that the group using the most expert process had the best results. But this could have been because the group had three members rather than two, or because my note-taking was more accurate for this group than the others, since they presented first. Students also frequently forgot to submit work that would help with my data: in Spring 2016 I asked students to submit an analysis once before listening, and once after, but only 1 of 9 students submitted.

A strategy that worked better in Spring 2016 was to give students prose questions that guided them toward the use of expert processes. For instance, this prompt (.pdf) asks students to listen to a piece of music first, and then to listen and look at a score. I was happy to see that 5 of 8 students referred to these two experiences in their writing, suggesting that I successfully encouraged expert processes. Analyzing the answers for content, I note that answers are generally more specific than on other questions on the assignment where there was no cue to listen. This suggests that listening to a piece enriches the analytical statements that students make about it.

In Year 3 of the project, I felt more hesitant to investigate this link further because of the complications described above. In one analysis assignment, I looked for common features (.pdf) of those assignments I’d given ‘check-plus’ grades (the highest possible grade in my grading system). However, as I began to analyze the data I realized that a full answer to question is outside the scope of this project, and thus left it for future work.

Data on Writing for Analysis Assignments

Student performance on Writing for Analysis assignments was much easier to measure, and is relevant to my project since WA assignments are where students behave the most like “expert analysts.” In Fall 2015 I revised the assignment process and design between WA1 and WAs 2 and 3, and in Fall 2016 I used the revised assignment process throughout the semester. I analyzed the results of my interventions in two ways: by looking at raw grades, and by examining the content using an adaptation of my grading rubric.

In my Year Two Portfolio (.pdf) I discussed how a comparison of 2015 student performance through grades on WA1 (pre-revision) and WA2 (post-revision) showed a stunning change from two of seven “A” grades on the first assignment to six of seven “A” grades on the second. While external factors may have influenced the jump, it’s hard to dismiss the raw data here. In 2016 I had a larger sample size, and found that the raw scores (.pdf) were less easily understood. There was a bigger spread across grades on WA1, although the percentage of students who scored above a C was the same as in 2015. There was, again, a striking rise in grades on WA1 to WA2, but this time the only change to my teaching strategy was the particular prompt for the assignment. I attribute the change in 2016 to a combination of the following: students learning how to better meet my grading expectations; students valuing the scaffolded process and taking it more seriously; students investing more time on the assignment because of their first grade; or, better student mastery of the analytical concept needed for the assignment. Unlike in 2015, 2016 saw more lower grades on the final WA3. I speculate that this has more to do with end-of-semester burnout than any pedagogical strategies on my part.

In 2015 I collected one sample (.pdf) of actual student work on WA1 and WA2 and analyzed it using an adapted version of my original grading rubric (.pdf). In 2016 I collected student samples from WA1 and WA3, and grouped students into particular categories based on their assigned grades. I then used my adapted rubric to analyze the work from those students whose grades rose from WA1 to WA3. These two groups collectively accounted for about half of the students in the class, while the rest of the class either maintained a high performance throughout (several students), or a low performance throughout (a couple of students).

Reading through all 6 WA1s as a group (.pdf), I notice several features informally. Every student is meeting the basic assignment requirements: using music theory terms to complete a musical analysis, and making some kind of argument. The missing piece in all cases is detail and specificity: the thesis statements are quite general, and the musical evidence to support the thesis is quite broad, general, and not described in detail.

The change from WA1 to WA3 was more striking in some students than others, but I note that by the final assignment, most students are better at using meaningful musical details to support more specific arguments about the pieces they are studying. This might be the primary factor contributing to the overall rise in grades.

My initial Decoding the Disciplines exercise made it clear just how difficult it is to uncover expert musical processes, even for an expert. For students, articulating how to analyze music was even more challenging. Both experts and experts-in-the-making often forget that music is experienced with ears and fingers and body as well as the mind and the eye, leaving their “whole musical selves” out of the musical analysis. By highlighting the process of analysis as well as the results, both my teaching (and, by extension, the scholarship of teaching and learning); and my discipline of music theory can change.

In terms of my teaching, I realized that I am fairly lecture-averse, preferring to guide student learning through activities and assignments rather than lecturing. However, I now see a value in lecture as a way to model expert behaviors: whether in thinking through a problem, crafting a thesis statement, or, in the case of this project, demonstrating the expert processes that are involved in a high-level musical analysis. If I use lecture to show students how listening, performing, composing, comparing theory with case studies, and revising are all part of the process of musical analysis, then students see those as valued activities that they themselves should practice.

I also realized that, even if I never came up with a fail-safe way to uncover student processes of analyses, simply mentioning it in class more often caused students to become more self-aware of their processes, and led to more class discussion around how to improve those processes. By my third iteration of the course, I was regularly referring to the process of analysis informally in lectures, and drew student attention to the “whole musical selves” approach throughout the semester.

In terms of my discipline, I connect my difficulties in linking expert processes to expert analytical results to criteria in my field that are vaguely defined. The answers to the question “what makes a good analysis?” would likely be varied and potentially contradictory or unspecific. I’m inspired to continue investigating this question within my field, and also to consider the role of self-reflection in answering that question, and teaching it to students.

I’ve also come to understand how deeply my teaching is influenced by the traditional pedagogies of my field, where analysis and composition (that is, reading and writing music) are seen as the primary activities in undergraduate music theory teaching. Prose writing is often assumed to be more in the domain of music history teaching. In my music theory courses, I’ve emphasized prose writing across the curriculum, but when it came time to evaluate the ‘process of analysis’ and the connection of expert processes to expert results, I was initially hesitant to use the Writing for Analysis assignments as a means of assessment. Now, at the end of the project, I see how that bias came out of my discipline. Rather than an extra bonus, I now see the Writing for Analysis assignments as the most representative assignment in terms of expert process, since they essentially ask students to behave like music analysts. The assignments required students to listen, perform, compare, and revise as they work to answer the assigned prompt. And the best Writing for Analysis assignments were the ones where students spent the most time doing this (as well as working on refining their abilities as writers). Far from being an added bonus, then, these writing assignments, like the “whole musical self” approach to analysis, are where the most vital and interesting thinking about music happens.

Contact CTE with comments on this portfolio: cte@ku.edu.