RTRL.70: Demographic Characteristics and Music Achievement of Eighth-Grade Students (Elpus, 2022)


Elpus, K. (2022). Middle school music uptake and achievement: Evidence from the 2016 Arts National Assessment of Educational Progress. Journal of Research in Music Education, 70(3), 248-270.

What did the researcher want to know?

What are the characteristics of 8th grade students enrolled in school music classes?

What did the researcher do?

Elpus accessed data from the National Assessment of Educational Progress (NAEP), “a series of standardized assessments intended to chart American students’ educational progress” (p. 249). Specifically, Elpus analyzed data from the 2016 Music NAEP, which encompassed a nationally representative sample of 4,340 students. Data included demographic information (e.g., race, sex), indicators of socioeconomic status (e.g., parents’ education, books in the home), and questions on music in and out of school.

Elpus calculated the proportion of 8th grade students enrolled in school music classes; how music enrollment varied across school-level characteristics; and demographic characteristics of students enrolled in music classes. He also used multiple regression to compare NAEP music scores among students in various kinds of music classes and to examine how various characteristics as associated with NAEP Music Achievement scores.

What did the researcher find?

Overall, 64% of 8th graders nationwide were enrolled in a school music class in 2016. Of these, just over half were enrolled in band, orchestra, or choir, and just under half were enrolled in a general music class. 

Male students were overrepresented in general music classes and underrepresented in choir classes. White students were overrepresented in band and choir classes and underrepresented in general music classes, while Black students were underrepresented in band and orchestra and overrepresented in general music classes. Students enrolled in ensembles tended to have higher socioeconomic status than those not enrolled in music classes.

Specifically looking at general music classes, students were more likely to be male, eligible for free/reduced lunch, and/or receiving special accommodations under an IEP or Section 504 plan and were less likely to be White than were nonmusic or ensemble students, suggesting “students with these characteristics in middle school may be either systematically prevented or actively discouraged from enrolling in music ensembles and channeled into general music courses to fill needed music or arts requirements (p. 256). In terms of school-level characteristics, music enrollment was highest at private and public middle schools and lowest in charter middle schools and lower in the West than other regions of the U.S.

When comparing Music NAEP scores in relation to music class participation, “the kind of music course in which a student was enrolled was significantly associated with Music NAEP scores, with ensemble students significantly outscoring both their non music and general music-only peers” (p. 260). In addition, “there was no significant difference between nonmusical and general music-only students” (p. 260). Elpus also found significant differences by gender (female students scored higher), race (Asian and White students scored higher), socioeconomic status (students eligible for free/reduced lunch and having fewer books in the home scored lower), special education and English language learner status (both scored lower), and attendance record (students who missed fewer days of school scored higher).

What does this mean for my classroom?

Several previous studies have shown distinct differences between the populations of students who choose to participate in high school music and those who do not (Elpus & Abril, 2019, Kinney, 2019), but Elpus’s results show that this begins earlier than high school. According to Elpus, “Indeed, it is quite likely that these disparities first emerge much nearer to the point in schooling when ensemble music making first becomes elective” (p. 264). Music teachers should be aware of what populations may be less likely to enroll in school music classes once it becomes optional, examine possible reasons why certain populations may elect not to participate, and work to make their classes a more inclusive and welcoming space for those populations. 

We also need to examine whether over- and/or under-representation of certain populations in ensemble versus general music settings may be an issue of access. Elpus asks, “Are general music courses more readily available than ensembles at middle schools serving greater proportions of students of color and students living in poverty? Or, perhaps if neither access nor adult channeling underlies this finding, is there something about the content and form of general music classes that make these courses more attractive to Black and Hispanic (Latinx) students” (p. 265)?

RTRL.56: Student Preferences for Music Learning And Music Courses (Pendergast & Robinson, 2020)


Pendergast, S., & Robinson, N. R. (2020). Secondary students’ preferences for various learning conditions and music courses: A comparison of school music, out-of-school music, and nonmusical participants. Journal of Research in Music Education, 68(3), 264-285.

What did the researchers want to know?

In terms of music class, what are secondary students’ preferences for teacher role, group size, repertoire, and class options? Do these preferences differ by socioeconomic status, ethnicity, or whether students participate in school music, out-of-school music, or no music? Why do some students choose not to participate in music?

What did the researchers do?

Pendergast and Robinson surveyed 827 middle and high school students from one urban and one suburban school district who were chosen through stratified random sampling to ensure diversity. The survey included items regarding demographics (gender, ethnicity, free/reduced lunch status, participation in school music, participation in out-of-school music), their learning condition preferences  (group sizes, amount of independent learning, degree of input on repertoire selection), and their interest in various music courses (including piano/guitar, composition with technology, popular music group, large ensemble, history/theory, and world music group).

What did the researchers find?

Overall, the least preferred teacher role was exclusively teacher-led instruction, with only 23.0% of students preferring this option. Teachers sometimes leading was preferred by 43.5%, and learning independently with teacher intervention only when necessary was preferred by 33.5%. Among in-school music participants, only 20.1% preferred exclusively teacher-led instruction. In terms of group size, only 13.3% of students preferred exclusively large group, with 35.9% preferring small groups and 50.8% preferring a mix of large and small groups. In terms of repertoire choice, most students (61.8%) preferred when the teacher and students choose music together, with 28.7% preferring that students select all the music and only 9.6% preferring when teachers select all the music.

All students—even those who participate in school music—expressed the most interest in piano/guitar class. While large ensemble received the second-most interest from in-school music participants, it was rated second-to-last by out-of-school music participants and nonparticipants. The second and third most interesting to those students were music composition class with technology and popular music group.

Pendergast and Robinson found a significant difference in repertoire choice preferences by ethnicity, with Latinx and Black students significantly more likely to prefer that students choose the repertoire. When comparing preferences by music participation group, Pendergast and Robinson found that out-of-school participants and nonparticipants were significantly more likely to prefer small group learning and students choosing the repertoire learned.

When responding to questions on why they were not enrolled in school music, 80.1% of nonparticipants said they were not interested, while 19.6% said they don’t have time and less than 5% said they couldn’t afford an instrument. Among out-of-school music participants, 35.4% said they had no interest, 37% said they did not have time, 9.1% said they couldn’t afford an instrument, and 20.1% indicated “Other.”

What does this mean for my classroom?

While teacher-led, large-group instruction may be the dominant mode in traditional school music classes, this kind of learning may actually be least preferred by students. Music teachers might consider providing opportunities for small group and/or student-led learning experiences. Since Pendergast and Robinson found that very few students prefer having the teacher select all their repertoire, teachers also might consider allowing students to have a voice in choosing the music they will perform. This may be especially critical for teachers who work with Latinx and Black students, as these students were significantly more likely to prefer choosing their own repertoire. Small group learning and student-chosen repertoire may be especially powerful in attracting more students, as they were significantly more preferred by out-of-school participants and nonparticipants.

In addition to adjusting how current music courses operate, that out-of-school participants and nonparticipants rated large ensemble as one of their least desired music class options suggests we should consider adding other types of courses to the curriculum. Doing so might draw in students who are not interested in traditional band, orchestra, and choir offerings but would be interested in other types of much learning. Since over 80% of nonparticipants said they did not enroll in school music because they simply were not interested, it is possible that these students would enroll in another type of music offering they found more compelling. Piano/guitar, music composition class with technology, and popular music group may be promising options, as these were most preferred by out-of-school participants and nonparticipants.

RTRL.48: Predictors of Music Class Enrollment in Urban Middle/High Schools (Kinney, 2019)


Kinney, D. W. (2019). Selected nonmusic predictors of urban students’ decisions to enroll and persist in middle school and high school music ensemble electives. Journal of Research in Music Education, 67(1), 23-44.

What did the researcher want to know?

Who enrolls in urban middle school and high school music classes?

What did the researcher do?

Kinney obtained student demographic information from one large midwestern metropolis area. This district was comprised of mostly minority students (62% Black, 26% Caucasian, 9% Hispanic, 3% Asian, <1% Native American), and 78.5% of students were enrolled in free/reduced lunch programs. Since elective choices began in 6th grade in this district and state standardized tests were administered in 6th, 8th, and 10th grades, Kinney collected data for students in those three grades.

Kinney assembled a database that indicated whether each student in 6th, 8th, and 10th grade was enrolled in band, strings, or choir. This database also included the following information for each student:

  • reading achievement test score
  • math achievement test score
  • free/reduced lunch status (as an indicator of SES)
  • number of parents/guardians in the home
  • district mobility (whether the student had moved into the district in the past year),
  • school mobility (whether the student had transferred schools within the district in the past year)
  • ethnicity
  • sex

Kinney used these factors as variables in conducting a multinomial logistic regression to build a predictive model for initial 6th grade enrollment and 8th/10th grade enrollment/retention in band, orchestra, and choir.

What did the researcher find?

Statistical analyses revealed the following student characteristics to be significant predictors of whether a student was more or less likely to enroll in elective music classes at each grade level:

What does this mean for my classroom?

Results suggest that students with higher math achievement are more likely to participate in instrumental music in urban schools. However, the fact that this was consistent across the span of grade levels suggests “that higher achieving students are attracted to instrumental programs from the outset and that systematic differences between this population and the general school population remain relatively stable over time” (p. 36). This provides evidence against the common argument that participation in music raises standardized tests scores and instead implies the difference lies in who chooses to participate in instrumental music in the first place.

Band teachers should be aware that student SES can predict whether a student participates in band. Teachers should be conscious of the fact that SES can be a limiting factor for student enrollment and find ways to make access to band class accessible to students who might struggle to afford an instrument or other supplies.

High school band programs typically involve various before- and after-school requirements (e.g., marching band, pep band), which may be why students from two-parent/guardian homes were more likely to participate in band in high school. It is possible that single-parent/guardian homes encounter more challenges in navigating these extra activities. For example, a single parent who works the night shift may not be able to transport their child to or from before- or after-school activities. High school band teachers should be aware of these challenges and work to mitigate them so that all students have equal opportunity to participate.

Finally, teachers should be aware of the ways in which enrollment trends may reflect racial and gendered “norms” of who participates in band, orchestra, or choir. Only band reflected a more balanced enrollment of males and females, while females were two to three times as likely as males to enroll in orchestra or choir. While race/ethnicity trends varied across grade levels and ensemble types, the data implies that white students tend to more consistently enroll in ensemble classes. Teachers can be mindful of these trends in their recruiting efforts, and all music educators can work to defy racial/ethnic and gender stereotypes in the repertoire they choose, the materials they use, and the musicians who are represented in their classrooms. “Through deliberate, conscientious efforts to reach students often underserved by music ensemble offerings, teachers will no doubt create a more democratic, equitable, and viable elective choice for all” (p. 41).

RTRL.27: Who Enrolls in High School Music? (Elpus & Abril, 2019)



Elpus, K., & Abril, C. R. (2019). Who enrolls in high school music? A national profile of U.S. students, 2009-2013. Journal of Research in Music Education, 67(3), 323-338.

What did the researchers want to know?

What proportion of U.S. high school students enroll in ensemble and non-ensemble music courses, and what are the characteristics of students who enroll in ensemble music courses?

What did the researchers do?

Elpus and Abril utilized existing data on 25,210 high school students gathered via a nationally representative longitudinal study that followed students at 940 U.S. high schools. These data included demographic characteristics (e.g., sex, SES, race/ethnicity, prior academic achievement) as well as information from students’ high school transcripts, including whether they participated in a high school ensemble or non-ensemble music class. Elpus and Abril used bivariate and multivariate statistical analyses to identify demographic characteristics of music and nonmusic students and to investigate the unique impact of each characteristic.

What did the researchers find?

Of the students who graduated in 2013, 24% had enrolled in one or more music ensembles during at least one year of high school. Choir had the largest percentage of participation (13%), followed by band (11%) and orchestra (2%). In terms of non-ensemble courses, only 3% of students had enrolled in a guitar class, 3% in a piano class, and less than 1% in a music technology course.

The table below shows cross-tabulations for the entire sample, ensemble students, and each type of ensemble:

Characteristics of Music Ensemble Students and the Full Population
(Elpus & Abril, 2019)

Statistical analysis revealed that differences between instrumental and non-instrumental were statistically significant in every characteristic except birth-assigned assigned sex. In contrast, choir students did not differ significantly from non-choir students for any characteristic except for birth-assigned sex.

In order to investigate the unique impact of each demographic characteristic, Elpus and Abril used logistic regression, which is a statistical analysis that attempts to “control for” (i.e., remove the effects of) certain variables. These results showed that the probability of enrolling in an ensemble music class increases with SES as well as with prior academic achievement (as measured by a standardized algebra test). However, the latter was not true for Black/African American students, for whom “the likelihood of ensemble enrollment decreases as academic achievement increases” (p. 331).

It is also interesting to note that the effects of SES and prior academic achievement were not significant among choir students. When looking at each type of ensemble separately, SES only predicted enrollment among band vs. non-band students, while prior academic achievement only predicted enrollment among instrumental vs. non-instrumental students. With the exception of gender, choral students were more representative of the larger student population than band/orchestra students.

What does this mean for my classroom?

Elpus and Abril’s findings confirm that ensemble music courses are the most prevalent music offerings among U.S. high schools. We need to continue to think about how we might attract more students to participate in school music through non-ensemble courses. However, band and orchestra are disproportionately likely to enroll students of higher SES and higher prior academic achievement. Therefore, Elpus and Abril state, “the problem may not be in increasing the number of students in music courses (which is commonly heard in professional rhetoric) so much as ensuring that music courses are attracting a representative and wide cross-section of the general student population” (p. 334). As music educators, we can be aware of which types of students may tend to be less likely to enroll in ensemble music classes and actively work to recruit these students. 

Additionally, it is worth contemplating possible reasons why choir enrollment tends to be more reflective of the general population. Elpus and Abril suggest several possibilities:

  • The versatility of the singing voice and wide range of cultures and styles in choral music “might be more attractive to students from varying cultural backgrounds” (p. 335)
  • Choir may be “more readily available as an avenue of ensemble music-making for those high school students who, for any of a variety of reasons, did not begin to develop proficiency on a band or orchestra instrument in their earlier schooling” (p. 335).
  • “Choirs may require less time commitment and financial demands than band, which possibly makes them more attractive to students who have familial or work commitments and those of more limited financial means” (p. 335).

Based on these speculations, the following types of questions warrant reflection:

  • How might we increase the appeal and relevance of music course offerings to students from varying cultural backgrounds?
  • How might we provide avenues of entry into high school instrumental music classes for students who haven’t previously participated?
  • What time, financial, or other constraints may deter some students from participating in ensemble music classes? 
    • Are we being reasonable with the amount of time we expect from students outside of regular school hours?
      • Might our attendance and/or grading policies unfairly penalize students with work or family commitments?
      • Do we make the effort to truly listen to and take into consideration a student’s personal situation when they have a conflict?
    • What might be all the monetary costs—both obvious and hidden—to instrumental music participation (e.g., instrument rental/purchase, uniform fees, reeds and other supplies, private lessons)?
      • How realistic are these costs for students of lower SES? 
      • How might we help find ways to ease this financial burden and ensure that every student has the opportunity to participate and be successful in instrumental music?

RTRL.26: Singing Ability, Musical Self-concept, and Future Music Participation (Demorest, Kelley, & Pfordresher, 2017)

What did the researchers want to know?

What factors predict students’ decisions whether to participate in elective music classes?

What did the researchers do?

Demorest, Kelley, and Pfordresher surveyed 319 sixth-grade students (95% of the population) from five elementary schools that feed the same junior high in the Pacific Northwest. Students completed two questionnaires; one asked questions pertaining to their musical and family background, and the other measured their attitudes and beliefs about music participation, including their musical self-concept and perceived costs of participating in music. After students registered for their seventh-grade classes (since music study became elective in seventh grade in this school district), the researchers compared each student’s decision whether to continue participating in school music to their survey results.

The researchers conducted a second study in which they randomly sampled 55 of the 319 sixth graders to complete a singing assessment, in which students were asked to sing the song “Happy Birthday” as well as echo-sing various pitches, intervals, and patterns. Singing accuracy was compared to students’ decisions whether to participate in music and to their questionnaire responses.

What did the researchers find?

Students who elected to participate in a music class reported “significantly higher perceptions of musical self-concept than those who did not . . . ; were more likely to be influenced by peers in their participation choices . . . ; and believed more strongly that music is not a barrier to other activities” (p. 410). Further analysis revealed that musical self-concept, peer influence, and family musical engagement were the strongest predictors of students’ decisions whether to participate in music.

Statistical analysis showed no significant difference in singing accuracy between those who elected to participate in music and those who did not. Comparison to questionnaire responses revealed that “musical self-concept was the only unique predictor of singing accuracy” (p. 414).

What does this mean for my classroom?

Music teachers should be aware that musical self-concept is strongly related to students’ future musical participation and consider monitoring students’ musical self-concepts “to identify students who may be feeling less positive about their musicality. If caught early enough, perhaps teachers could provide opportunities for students to improve their self- perceptions of musicality. . . . If we as a profession are interested in expanding the number of children who choose to continue elective music instruction, we should continue to explore how singing skill and musical self-concept interact throughout a child’s early development and what experiences might encourage improvement in both attributes” (p. 417).

In Memoriam

Dr. Steven M. Demorest

August 24, 1959 – September 22, 2019

With gratitude for his many contributions to the music education profession

RTRL.20: School Music Participation and Lifelong Arts Engagement (Elpus, 2018)


Elpus, K. (2018). Music education promotes lifelong engagement with the arts. Psychology of Music, 46(2), 155-173.

What did the researcher want to know?

Does participation in school-based music education affect the likelihood that students will engage in music performance or attend musical events as adults?

What did the researcher do?

Elpus analyzed data from the U.S. National Endowment for the Arts 2012 Survey of Public Participation in the Arts. To examine musical creation/performance, he looked at respondents’ answers to items that asked whether they had played a musical instrument or performed/practiced any singing (alone or with others) in the previous 12 months. To examine music patronage, he examined respondents’ answers to items that asked whether they had attended live performance of classical music/opera or live jazz in the previous 12 months.

Elpus analyzed this data in relation to whether respondents participated in school music education (performance and/or appreciation). Elpus treated the study as quasi-experimental, using observable covariates to adjust for selection bias. In other words, he took into consideration the respondents’ answers to the following:

  • race/ethnicity
  • gender
  • education level
  • parent’s education level
  • household income (above or below $50,000)
  • region of the U.S. in which they lived

What did the researcher find?

According to descriptive statistics, 28% of respondents had participated in school-based music education. In the year prior to the survey, only 13% had attended live classical music/opera, 13% had played an instrument, and 10% had sung.

Those who had participated in school-based music education (performance or appreciation) were two to three times more likely to to play an instrument or sing as adults. Interestingly, there was no significant relationship between household income and musical creation/performance (after controlling for other variables). However, Elpus found that those with higher levels of education and those whose parents had attained higher levels of education were more likely to play an instrument, which corresponds with existing notions “that children of higher socioeconomic status (SES) parents are more likely to pursue instrumental music education, with parental education serving … as a blunt proxy for respondents’ family SES in childhood” (p. 165).

Those who participated in school-based music performance classes were 35% more likely to have attended classical music or opera performance than those who hadn’t participated in performance classes, while those who had taken music appreciation classes were 93% more likely to attend these events than those who hadn’t taken music appreciation classes. The likelihood that respondents had attended classical music/opera events increased significantly with education level; those holding bachelor’s degrees were over five times more likely to attend these performances than those with no high school diploma.

Compared to White respondents, African American adults were significantly less likely to report that they played an instrument or had attended classical music/opera but were 244% more likely to have attended live jazz.

What does this mean for my classroom?

If one of our aims as music educators is to foster in our students a lifelong connection with the arts, Elpus’s findings can reassure us that this goal is indeed being achieved for many. Furthermore, if we wish to increase lifelong engagement with the arts, we must be sure that all students have access to high-quality music education. 

However, the topics examined in this study focused on Euro-centric art music, with popular/vernacular forms of music largely ignored. We must continue to examine the kinds of musical experiences we offer in schools, who may or may not choose to participate in the courses offered, and how we might broaden our concept of school music to engage and better serve more students.