Assessment in Music: A Practitioner Introduction to Assessing Students

What Is Assessment?

Idealistically, assessment has been presented as a method for gathering information relevant to both teachers and students about teaching and learning, centered on student knowledge and skills (Parkes, Rohwer, & Davidson, 2015). Goolsby (1999) suggested that four types of assessment are, in his experience, commonly used in instrumental settings: (a) placement (auditions, chair tests, etc.); (b) summative (concert, festivals, etc.); (c) diagnostic (identifying performance skills and knowledge difficulties); and (d) formative (the regular monitoring of students for learning outcomes). Eisner (1998) also underscored the diagnostic, positioning, content evaluative, and reflective roles of assessment in the arts. Conceptual definitions and goals, such as those mentioned above, provide the foundation for the creation and implementation of assessments. It may, therefore, be necessary to have a basic understanding of what assessments accomplish.

Students have been found to benefit from specific and reliable feedback gained from both formative and summative assessments (Colwell, 1998; Salvador, 2011; Sicherl Kafol, Kordeš, & Holcar Brunauer, 2017; Zerull, 1990). Formative assessments have been defined as assessments conducted during learning, and have been one of the primary aspects of designing and differentiating instruction (Salvador, 2011; Saunders & Holahan, 1997; Sicherl Kafol et al., 2017). Music teachers have been said to engage in formative assessment whenever they listen critically to student performance/practice, make judgments, and provide feedback (Saunders & Holahan, 1997). While practitioners may often use formative assessment to ensure that students do not simply skirt the instructor’s acceptance of mistakes (Goolsby, 1999), formative assessment has often been relegated to error detection, identification, and communication (Colwell, 1998). True formative assessment, according to Colwell (2008), only occurs when “some learning action follows from the testing, otherwise it is merely frequent summative assessment” (p. 13).

Summative assessment, or the end measurement of either learning, growth, or achievement (Sicherl Kafol et al., 2017), can also be thought of as answering the following question: How did the student(s) do? (Mastrorilli, Harnett, & Zhu, 2014). In music, individual performances and competitions often serve as summative assessments and have been found to be the source of many student grades and teacher evaluations (Ciorba & Smith, 2009; Latimer, Bergee, & Cohen, 2010). Russell and Austin (2010) found that the vast majority (95%) of the 352 secondary music teachers surveyed worked in schools that used traditional summative grading systems. Comparable systems have been noted as extremely common; however, Harrison, Lebler, Carey, Hitchcock, and O’Bryan (2013) suggested that simple grades lack meaningful feedback. Due to their perceived and real importance, summative experiences often influence instructional decisionmaking, from sequencing to curriculum.

Why Does Assessment Matter?

Researchers have found that students may view assessment, particularly when measured by grades, as important and a powerful motivating factor (Colwell, 1998; McClung, 1996; Reimer, 2009). Boud, Cohen, and Sampson (1999) claimed that “assessment is the single most powerful influence on learning in formal courses and, if not designed well, can easily undermine the positive features of an important strategy in the repertoire of teaching and learning approaches” (p. 414). Lehman (2008) also emphasized the interconnectivity of instruction and assessment. In this view, assessment has been considered an essential component of quality instruction and necessary for improvement in teaching and learning to take place (Eisner, 1998; Lehman, 2008). Formative assessments, in their role of providing data for reflection and adjustment of instruction, may improve specific lessons, general content delivery, sequencing, and/or curricular focus. Additionally, the cumulative nature of summative assessments may be used to evaluate achievement, influence the development of future goals, serve as a motivating factor, and provide meaningful data about teaching and learning in the music classroom. It may be important, however, to note that these positive factors may be predicated on the proper implementation of assessment.

Increases in accountability, spurred by laws such as No Child Left Behind and grant programs such as Race to the Top, have exacerbated the preexisting natural emphasis on assessment (Ciorba & Smith, 2009; Wesolowski, 2014). This increased public attention, in turn, has increasingly affected policy decisions about music assessment (West, 2012; Zerull, 1990). Music teachers have been asked to teach more than one area of music, in many cases, to free up resources to support subjects struggling with high-stakes testing (Lehman, 2008; West, 2012). The implementation of standardized music tests, while attempted several times in the past 40 years, has not been widespread due to the inherent difficulties in uniform measurement of music skills (Parkes et al., 2015; Zerull, 1990). Researchers have also pointed to the breadth of music as a field, the expressive nature of the discipline, and the intrinsic value of all art forms as factors contributing to music’s unsuitability for assessment (Lehman, 2008; Zerull, 1990).

Assessment has been found to be of such importance that evaluating music and music performance is included in the current national standards (Shuler, Norgaard, & Blakeslee, 2014). Music has been recognized at both national and state levels as being a core curricular subject, which has brought more focus on assessment in the music classroom (Fisher, 2008). As such, Asmus (1999) noted that some objections to assessment in music may stem from a lack of specificity as to students’ expected knowledge and skills. Fisher (2008) provided several reasons for the adoption of more regular assessments in music classrooms, including easily understood and consistent accountability, legitimization in the perceptions of those outside music, and protection of music instruction time. Wright, Humphrey, Larrick, Gifford, and Wardlaw (2005), however, countered that there are also viable arguments that music is not suitable for formal standardized testing, including the inconsistent schedule of music instruction and the miniscule impact on subject credibility testing may provide. Likewise, the researchers listed additional contributing factors, including the breadth of music as a field, the expressive nature of the discipline, and the intrinsic value of all art forms. Asmus (1999) suggested that music, possibly more than other subjects, operates in more unique circumstances and has longer teacher-to-student interactions. Therefore, formative and summative assessments may affect long-term learning outcomes.

The importance of music assessment extends beyond the need to evaluate for pedagogical purposes. A number of studies have shown that assessments can be used to communicate value and advocate for school music to those outside the discipline (Colwell, 1998; Reimer, 2009; Zerull, 1990). Accordingly, Reimer (2009) remarked that the use of nonmusic criteria for formative assessment, or even worse for summative grades, can lend credence to claims that music is a less serious subject than other academic disciplines. McClung (1996) found that only 18% of administrators surveyed perceived choir as having equal educational status as other core subjects. McQuarrie and Sherwin (2013) found that many of the most common assessments used in music classrooms were nonachievement based, which deviated sharply from the assessment literature in the Music Educators Journal. In general, music ensemble instructors have embraced the necessity of assessment in their classes, but often have no training in developing or administering assessment (Colwell, 1998).

How Do We Assess in Music?

Within the differing forms and goals of assessment, scholars have noted a multitude of approaches (McQuarrie & Sherwin, 2013; Rohwer, 1997; Russell & Austin, 2010). Teachers and researchers have used content-specific rating scales, checklists, rubrics, report cards, aptitude testing, observations, and portfolios to assess student learning and growth (Parkes et al., 2015; Rohwer, 1997; Salvador, 2011). Many of the common assessments found in schools, however, have been noted as either nonmusic or containing no actual measurements (Barkley, 2006; McClung, 1996; McQuarrie & Sherwin, 2013; Russell & Austin, 2010; Simanton, 2000). McQuarrie and Sherwin (2013) found that the top five assessments used in elementary general music were based on (a) participation, (b) effort, (c) individual performances using informal observation, (d) group performances, and (e) behavior. These results were echoed by Barkley (2006) who reported that observation was the most commonly used assessment strategy in elementary music classrooms, followed by concert performances and written tests. Russell and Austin (2010) found similar results among secondary music teachers, with the top five assessment criteria in order of use being (a) performance/skill, (b) knowledge, (c) attendance, (d) attitude, and (e) practice. Additionally, Simanton (2000) reported that secondary band directors used (a) participation/attitude, (b) band music performance, (c) attendance, and (d) technique/sight-reading as the primary means of assessing students, and that 56% of assigned grades in band classes were derived from nonperformance criteria. McClung (1996) noted that students perceived individual participation and attitude as the most frequent predictor of grades in choir.

The proliferation of nonmusic assessments, such as attendance or attitude, in classrooms has contrasted with much of the published literature, which has focused on assessments like individual and group performance, standardized testing, assessment software, and others (McQuarrie & Sherwin, 2013). While simple to execute and often supported by administrators (McClung, 1996; Russell & Austin, 2010), nonmusic assessments have not been noted to support music learning and growth to the same extent as content-based assessments built on demonstrations of music knowledge and skills (Reimer, 2009). Russell and Austin (2010) determined that content-based assessment, such as evaluation of music performance or music composition, was the most effective approach to improve teaching and learning.

What Challenges Occur in Music Assessment

A number of inherent challenges exist when trying to develop music assessments. From a broad perspective, the lack of agreement on music curricula or the end goals of instruction have created divisions in assessment approaches, which in turn may have erected barriers to assessment (Lehman, 2008; Reimer, 2009). The subjective nature of any music value judgments may have further hindered the development of any consistency in assessment; however, the rise of the standards movement may assuage some of these concerns (Colwell, 2008; Lehman, 2008). The use of standards may also align with McClung (1996), who commented that grades should be linked to specific learning objectives.

Logistical challenges may also influence assessment decisions and impede improvement in assessment practice (Ferm Almqvist, Vinge, Väkevä, & Zandén, 2017; Harrison et al., 2013; Russell & Austin, 2010; Salvador, 2011). In a qualitative study of three purposely selected elementary music teachers, Salvador (2011) noted that participants viewed the number of students taught, time constraints, and the lack of administrator support as being obstacles to improved assessment. Similarly, Harrison et al. (2013), Lehman (2008), and Ferm Almqvist et al. (2017) all echoed the concern over class size issues and their impact on effective assessment. Russell and Austin (2010) also found that administrative guidance or assistance with assessment was rare, yet when administrators provided help, nonmusic assessments were less likely to be present in the classroom. Additionally, Reimer (2009) suggested that the complexity of grading individuals in ensembles may create confusion or disagreement between music educators and administrators, and that administrators may be concerned that such confusion could lead to grade inflation.

A lack of transparency in grading procedures used in ensembles can lead to perceptions of favoritism (Harrison et al., 2013). Providing written grading policies to students and parents may alleviate transparency concerns. On the contrary, Ferm Almqvist et al. (2017) found that assessment often drove instruction and learning goals directly, stating that “assessment procedures and techniques [became] so persistent that they completely dominate[d] the teaching and learning experiences” (p. 6). Ferm Almqvist et al. (2017) further suggested that transparency contributed to the dominant role assessment played in participant classrooms. Overall, student understanding of teacher expectations has remained crucial for learning, and assessments have been suggested as meaningful indicators of expectations, goals, and objectives (Reimer, 2009). Including rationales and reasoning in assessments has also been offered by practitioners as a way for music educators to document rigor (Wesolowski, 2014).

Even when transparency exists, assessing creativity in music has often proven especially challenging for music educators (Hickey, 2001). Scholars have described creativity as novel, appropriate, and valuable (Barbot & Lubart, 2012; Stefanic & Randles, 2015) and have noted its historical importance in music education (Rohwer, 1997). Despite the educational value of creativity, the inherent subjectivity may prove problematic for educator assessments (Hickey, 2001; Rohwer, 1997).

Researcher-designed standardized assessments and product-based assessments have been used to evaluate and measure creativity in research literature. The most prominent researcher-designed standardized measures of creativity in music were developed by Guilford and Torrance during the middle of the 20th century (Rohwer, 1997; Stefanic & Randles, 2015). The Torrance Tests of Creative Thinking (TTCT) have been renormed as may serve as an example of standardized creative measures (Stefanic & Randles, 2015). Individuals or groups respond to activities (five activities on the TTCT-Verbal and three activities on TTCT-Figural) in writing and in drawing, which are then scored (Kim, Crammond, & Bandalos, 2006). These standardized tests have been used as a foundation for creativity research, although concerns about reliability have been raised (Stefanic & Randles, 2015).

In contrast, product-based assessment has been favored by researchers in recent years, with many studies using Amabile’s (1983) Consensual Assessment Technique (CAT) (Barbot & Lubart, 2012; Eisenberg & Thompson, 2003; Hickey, 2001). Counter to standardized tests, CAT measurements involve domain-appropriate judges using a Likert-type scale to rate artistic products on multiple criteria. This approach has been found by researchers to be reliable when used with both improvisation (Eisenberg & Thompson, 2003) and composition (Hickey, 2001; Stefanic & Randles, 2015). Stefanic and Randles (2015) also found that CAT was less reliable when evaluating group compositions or when there was only one judge. The need for multiple expert judges may pose a logistical problem for the implementation of CAT in some settings (Lu & Luh, 2012).

General tenants from both researcher-designed measurements and general assessment strategies may still be valuable to practitioners in developing appropriate assessments for music creativity. The lack of criterion or standards for evaluating creativity, for instance, may free music educators to focus on growth in novelty or appropriateness of the creative product. To this end, practitioners may use assessment tools such as checklists, rating scales, peer/teacher responses, or portfolios to evaluate student creativity (Rohwer, 1997).

Finally, the common use of nonmusic assessments may point to external factors influencing assessment choices. Russel and Austin (2010) found that practitioner practice and district/campus policies were often at odds, and music teachers were rarely provided assistance in reconciling any discrepancies. Barkley (2006) found that elementary music educators believed nonmusic factors, such as itinerancy, district/campus policies, planning time, training, and resources, all influenced assessment practices. Simanton (2000) suggested that the extensive work load of band directors may contribute to the lack of specific, individual assessments.

Conclusions

In a climate increasingly focused on accountability, how do we implement effective assessment without sacrificing effective music making? This process may begin with the instructor’s deep understanding of music. Colwell (1998) argued that developing a meaningful assessment, particularly in a discipline with a high level of subjectivity, requires a strong foundation of music knowledge. Difficulty in assessment may increase if one attempts to keep the flexibility needed for creativity and artistry in the music classroom alive. The authentic performance assessments that so often constitute the majority of content-specific grades in music ensembles may, if developed and implemented poorly, contribute to the restriction of creative thought by reinforcing the power dynamics of teacher-led decisionmaking. As such, knowing music from a formalist and expressionist perspective might allow practitioners to create learning environments tied to a holistic ideal of music making, instead of simply continuing the deficiency rote performance assessments often seen in school.

Assessing specific knowledge may also be important; however, knowing exactly what and how to assess appropriately may remain problematic. Therefore, a significant understanding of assessment fundamentals and strategies may help practitioners gather truly meaningful data about their students (Colwell, 1998). Yet, in spite of this self-evident fact, many teachers may be unaware of validity concerns in either teacher-created or externally provided assessments. Additionally, deficiencies in practitioner assessment knowledge may lead to misuse of assessment strategies, and in turn promote incorrect conclusions and negatively influence instruction. To avoid misuse, researchers have suggested that assessments be linked with learning objectives (McClung, 1996; Parkes et al., 2015) and be implemented in an instructionally meaningful sequence (Wesolowski, 2014). While nonmusic assessments may be common, they may devalue the overall learning goals and public perceptions of the rigor of school music.

Accordingly, practitioners should understand the knowledge and skills related to assessment in order to evaluate and instruct students effectively. Therefore, intentionally varying assessment approaches and material assessed may provide a more holistic picture of student progress and the efficacy of teacher instruction. Additionally, disparate learning objectives may be best assessed by disparate techniques. Consequently, practitioner planning may be an essential part of meaningful assessment.

Planning may also allow for practitioners to gain a more holistic view of student learning. For instance, a student composition may be assessed through both a teacher checklist to provide feedback about specific areas of growth and a written peer evaluation to address more general aesthetic attributes (Rohwer, 1997). Planning may also mitigate issues common to assessment. Peer feedback, for example, may require careful introduction and practice of procedures to develop a safe environment for the students and facilitate the generation of relevant and meaningful feedback. Sicherl Kafol et al. (2017) found that proper establishment of peer feedback procedures led to improved student perceptions of assessment.

Practically, assessment can direct student and teacher focus. Ferm Almqvist et al. (2017) found that summative assessments drove student learning, as students were coached to perform well on assessments. In light of the power of assessment to shape content, Duke (2005) advocated for planning instruction with the assessment clearly in mind. For practitioners, this may mean deciding on aspects of the material to assess and approaches prior to developing lesson plans. As an illustration, when working with a secondary ensemble, instructors may select the music, decide on the standards and objectives for musicality/expression/accuracy/and so on, develop several approaches for formative assessment (such as an overall rubric for the selected music, checklists for each song, or designated moments of peer feedback), and culminate with the summative assessment of concert performance. Each specific learning objective or standard may need multiple assessments. Such frequency may serve to increase the opportunities to transition content knowledge about music into procedural knowledge of making music while lessening the consequences for failure (Duke, 2005). Consequently, practitioners may want to have frequent, intentional, formative assessments in their classrooms to facilitate teaching and learning.

As discussed previously, several different approaches to assessment may be best suited for differing objectives. Performance remains one of the most common forms of assessment in music; however, practitioners may wish to extend their conceptualization of performance beyond the concert stage. Performances for peers may provide opportunities for reflection, the development of evaluation skills, and practice in the procedural demonstration of music understanding. Furthermore, performances in varying settings may allow for students to exhibit music learning for the teacher, administration, community, and themselves. Practitioners may then use the information gathered to inform future curricula and instruction. For instance, students may be assigned solos and ensembles as part of the class curriculum. During the learning stages, performances for the instructor and other students may offer frequent, low-impact opportunities for feedback and assessment. Building from these performance assessment experiences, students may then participate in community performances of their music. At this stage, students have further opportunities to demonstrate their understanding of the music, and if recordings are made they may again reflect, either individually or with others, on their performance. Finally, performing the solo or ensemble as part of a competition may give a summative assessment of students’ understanding of the piece and individual instrument. By having multiple assessments in various settings, teachers may mitigate some of the noted subjectivity of performance assessments (Bergee & McWhirter, 2005; Latimer et al., 2010; Reimer, 2009; Ryan & Costa-Giomi, 2004).

Similarly, rubrics may help practitioners clearly communicate expectations and increase consistency in certain assessment areas. Practitioners may use rubrics as a way to communicate the desired learning outcomes of a particular unit or topic, and give structure to planned assessment. To illustrate this, practitioners may create a rubric for a piece of music prior to distribution to students that describes the intended learning objectives that will be assessed. This rubric may prime students for engaging with the music once rehearsals begin. Of particular application, the high rate of interjudge reliability found with the use of rubrics (Latimer et al., 2010; Norris & Borst, 2007) may facilitate assessments where more than one practitioner might be involved, such as placement auditions. Correspondingly, rating scales and checklists may also be used to communicate expectations.

Portfolios, while potentially time-consuming for practitioners, may provide a wider picture of student learning by incorporating various diverse artifacts related to music. Student music theory work, performance, composition, and writings may all contribute to a holistic picture of understanding and growth. Practitioners who implement portfolios may desire to predetermine a specific approach to artifact inclusion in the portfolio to avoid student confusion and effective assessment of the learning objectives. Parkes et al. (2015) noted that blind portfolio reviewers occasionally struggled to understand teacher artifact selection. To reduce extraneous time spent evaluating portfolios, practitioners may choose to have a priori categories or descriptors for artifact inclusion and establish overarching goals and procedures for the portfolio. Structured planning may allow practitioners to synthesize artifacts with greater ease. Conversely, too detailed criteria may limit creativity by once again focusing students and teachers toward the assessment instead of the learning objective.

Assessment and accountability are inherent in music education. As a profession, we strive for improved teaching and learning, leading toward instilling in our students a lifelong love of music. Consequently, the effectiveness of our assessments matters, as motivation (Colwell, 1998; McClung, 1996; Reimer, 2009), external value judgments (Colwell, 1998; Reimer, 2009; Zerull, 1990), and most important, student learning (Boud et al., 1999; Eisner, 1998; Lehman, 2008) all can be linked to assessment.