ABSTRACT 
 
 
 
 

Title of Dissertation: MUSIC TEACHER LABOR MARKETS: 
THREE STUDIES ON MARYLAND EARLY-
CAREER MUSIC TEACHERS 

  
 David Steven Miller, Doctor of Philosophy, 2023 
  
Dissertation directed by: Dr. Kenneth Elpus, School of Music 

 
 

In these three studies, I analyze data from the Maryland Longitudinal Data System 

(MLDS) to investigate how labor markets for early-career music teachers are related to the 

ongoing music teacher shortage in Maryland. In the first study, I explore the transition from 

postsecondary graduation into the public-school music teacher workforce. In the second study, I 

examine early-career music teachers' retention and attrition patterns. In the final study, I consider 

the impacts of the COVID-19 pandemic on these two critical stages of the music teacher 

pipeline. In the final chapter, I evaluate evidence from these three studies alongside the broader 

teacher labor market literature to provide policy recommendations that may ameliorate the 

worsening music teacher shortage in Maryland. 

Leveraging data from the MLDS, I used a variety of analytic strategies to describe the 

workforce transition of individuals who graduated from Maryland institutions of higher 

education with degrees in music education between 2008 and 2020 (N = 700). I used linear 

probability modeling to examine the individual factors associated with entry into the public-



   

school music teacher workforce in Maryland. I used descriptive statistics to examine the teaching 

responsibilities of new teachers and the characteristics of schools in which new teachers worked. 

Finally, I explored the characteristics of high schools from which the new teachers had graduated 

and how these characteristics compared to the schools in which new teachers were employed. 

Results indicated that approximately 60% of new graduates entered Maryland's public-school 

music teacher workforce. The majority (60%) found employment in elementary schools, and 

59% taught general music as part of the instructional responsibilities. Overall, new teachers 

tended to come from high schools with higher overall socioeconomic status, larger student 

enrollment counts, and student body demographics with higher proportions of White students 

compared to state averages. New music teachers in Maryland also taught in schools that were a 

farther distance from home compared to previous national estimates for all teachers. 

In the second study, I examined early-career music teachers' retention and attrition 

patterns between 2013 and 2020 (N = 6,430). On average, 74% of early-career music teachers 

retained their current position each year, 8% moved to a different school in Maryland, and 18% 

of left public school music teaching in Maryland. Teachers who moved schools tended to have 

different teaching responsibilities in their new position. About half of the teachers who moved 

schools changed the grade levels that they taught, and the vast majority of this change (77%) was 

to an older student population. On average, teachers tended to move to schools with higher 

average socioeconomic status and to schools whose demographic composition had a higher 

proportion of White students. I used discrete-time survival analysis to examine the teacher-, 

school-, and district-level characteristics associated with whether and when an individual would 

leave public-school music teaching in Maryland. Kaplan-Meier survivor functions showed that 

half of all music teachers left public-school music teaching in Maryland within their first five 



   

years of teaching. Attrition risk was greatest in years two and three. Other factors associated with 

attrition risk included sex, race, ethnicity, school grade level, aspects of school culture, the 

student body, school and district urbanicity, and the number of teachers employed by the school 

district. 

In the final study, I exploit the exogenous nature of the COVID-19 pandemic to explore 

its impact on the workforce transition and attrition of early-career music teachers in Maryland. 

Using an interrupted time series analysis, a quasi-experimental design that can yield causal 

insights, I demonstrate that attrition of early-career music teachers was likely not impacted by 

the pandemic. Analyses for the workforce transition were not as conclusive: There was a drop in 

the workforce entry rate in the first year of the pandemic, but the second year came with a sharp 

rebound to a much higher rate of workforce entry. Still, the number of provisionally-licensed 

music teachers, an indicator of shortage severity, has been increasing steadily for nearly a 

decade. Although media narratives have sounded the alarm on the COVID-19 pandemic’s 

potential impact on the teacher shortage, it would be disingenuous to suggest solving pandemic-

related problems would reverse the trend. 

I conclude this dissertation with a discussion of policy implications for Maryland local 

education agencies and music teacher preparation programs. Additionally, I discuss the unique 

substantive and methodological contributions to the broader music education literature. 

Substantively, these studies provide a novel framework for examining music teacher labor 

markets. Methodologically, these studies illustrate a blueprint for music education researchers to 

leverage the rich data available within the underutilized State Longitudinal Data Systems to 

explore issues of access, uptake, equity, and pipelines throughout K-12 and postsecondary music 

education.  



   

 
 
 
 
 
 
 
 
 
 

MUSIC TEACHER LABOR MARKETS: THREE STUDIES ON MARYLAND 
EARLY-CAREER MUSIC TEACHERS 

 
 
 

by 
 
 

David Steven Miller 
 
 
 
 
 

Dissertation submitted to the Faculty of the Graduate School of the 
University of Maryland, College Park, in partial fulfillment 

of the requirements for the degree of 
Doctor of Philosophy 

2023 
 

 
 
 
 
Advisory Committee: 
Dr. Kenneth Elpus, Chair 
Dr. Stephanie Prichard 
Dr. Michael Hewitt 
Dr. Robin Giebelhausen 
Dr. Adam Grisé 
Dr. David Blazar, Dean’s Representative 
 



ii 
 

Dedication 

To the teachers and mentors who guided me down this path, the students who inspired me to 

persist, and our tenacious pursuit of a better world. 

  



iii 
 

Acknowledgments 

I’d like to first thank my advisor, Ken Elpus, for his enduring support throughout this endeavor. I 

would not be the teacher, scholar, or person I am if it were not for your mentorship. To the rest 

of my committee, your feedback, questions, and guidance have been invaluable. Stephanie, 

Mike, Robin, and Adam, you inspire me every day. David, you have gone above and beyond in 

your mentorship and support throughout all of my work within and beyond the MLDS Center. 

Additionally, I am grateful for the assistance provided by the MLDS Center. I appreciate 

the feedback received from the MLDS Center and its stakeholder partners. All opinions are mine 

and do not represent the opinion of the MLDS Center or its partner agencies. 

To all of my doctoral colleagues, your companionship and support has been a consistent 

high point throughout my degree program. The world is a better place for having you all in it. 

Allison and Bri’Ann, you both have been with me throughout the entirety of my journey here. I 

cannot overstate how grateful I am for your friendship, and you both have been an endless source 

of inspiration, fun, and reflection. From the bottom of my heart, I thank you for bringing such 

unbridled passion and joy into my life. 

To my supportive partner, Nicoletta, thank you for being my sounding board and my 

support system. Your patience and help throughout this journey have been immeasurable. To my 

uncredited co-author, Clover, your diligence by my side, at my feet, and sleeping on the 

keyboard throughout my writing sessions have been a source of constant comfort. Together, the 

two of you made it possible to overcome every hurdle along the way. 

  



iv 
 

Table of Contents 

Dedication ...................................................................................................................................... ii 

Acknowledgments ........................................................................................................................ iii 

Table of Contents ......................................................................................................................... iv 

List of Tables ................................................................................................................................ ix 

List of Figures ................................................................................................................................ x 

Chapter 1. Introduction ............................................................................................................... 1 

1.0 Overview of the Document ................................................................................................... 1 

1.1 Identification of the problem under consideration ................................................................ 3 

1.1.01 Teacher Labor Markets ................................................................................................. 3 

1.1.02 Music Teachers Labor Markets .................................................................................... 4 

1.1.03 A Constrained Music Teacher Pipeline: Leaks and Filters .......................................... 5 

1.1.04 Great Resignation ....................................................................................................... 12 

1.1.05 A gap in the literature: policy-relevant empirical work ............................................. 12 

1.1.06 Maryland Context ....................................................................................................... 12 

1.2 Purpose and Research Questions ......................................................................................... 14 

1.2.1 Chapter 2: The workforce transition of music education graduates in Maryland ........ 15 

1.2.2 Chapter 3: Retention and attrition of early-career music teachers in Maryland: A 
survival analysis .................................................................................................................... 15 

1.2.3 Chapter 4: Has the COVID-19 pandemic exacerbated the Maryland music teacher 
shortage? Evidence from a statewide longitudinal data system ............................................ 16 

1.3 Research Design: Analysis of State Longitudinal Data Systems ........................................ 16 

1.4 Significance ......................................................................................................................... 18 

1.5 Scope and limitations of the study ...................................................................................... 20 

Chapter 2. The workforce transition of music education graduates in Maryland ............... 22 

2.0 Introduction ......................................................................................................................... 22 

2.1 Literature Review ................................................................................................................ 23 

2.1.00 General Education ...................................................................................................... 23 

2.1.01 Demographic characteristics of those pursuing careers in teaching ....................... 23 

2.1.02 Student teachers: demographics and career plans ................................................... 25 

2.1.03 Geographic preferences of teachers: local labor markets ....................................... 26 



v 
 

2.1.04 Student teachers’ entry into the teaching profession .............................................. 30 

2.1.05 Issues of Equity - Barriers to Entry ......................................................................... 32 

2.1.06 Summary of education literature ............................................................................. 35 

2.1.10 Music Education ......................................................................................................... 35 

2.1.11 Motivation, Preparation, and Aspiration for Music Teaching ................................ 36 

2.1.12 Licensure Practices and Candidate Demographics ................................................. 40 

2.1.13 Issues of Equity in Music Teacher Education ......................................................... 42 

2.1.14 Summary of music education literature .................................................................. 44 

2.1.2 Summary ....................................................................................................................... 45 

2.2 Method ................................................................................................................................ 46 

2.2.00 Empirical Strategy ...................................................................................................... 48 

2.2.01 Research Question 1: Entry into the profession ...................................................... 48 

2.2.02 Research Question 2: Characteristics of Employment ............................................ 50 

2.2.03 Research Question 2a: Congruence of Teaching Assignment and Home School .. 51 

2.3 Results ................................................................................................................................. 51 

2.3.01 Research Question 1: Entry into the Profession ......................................................... 52 

2.3.02 Research Question 2: Characteristics of Employment ............................................... 54 

2.3.03 Research Question 2a: Congruence of Teaching Assignment and Home School ...... 58 

2.4 Discussion ........................................................................................................................... 59 

2.4.01 Workforce Entry ......................................................................................................... 59 

2.4.02 Characteristics of Employment .................................................................................. 60 

2.4.03 Congruence of Teaching Assignment and Home School ........................................... 64 

2.4.04 Conclusion .................................................................................................................. 66 

Chapter 3. Retention and attrition of early-career music teachers in Maryland:  

A survival analysis ...................................................................................................................... 68 

3.0 Introduction ......................................................................................................................... 68 

3.1 Literature review ................................................................................................................. 69 

3.1.00 General Education ...................................................................................................... 69 

3.1.01 Teacher Characteristics ........................................................................................... 69 

3.1.02 Preservice Teacher Preparation ............................................................................... 74 

3.1.03 School Characteristics ............................................................................................. 77 



vi 
 

3.1.04 Improving Retention ............................................................................................... 81 

3.1.05 Those who left ......................................................................................................... 84 

3.1.06 General Education Summary .................................................................................. 85 

3.1.10 Music Teachers ........................................................................................................... 86 

3.1.11 Music Teacher Mobility and Burnout ..................................................................... 86 

3.1.12 Professional Networks and Collegial Support ........................................................ 90 

3.1.13 Music Teacher Summary ........................................................................................ 93 

3.1.2 Summary ....................................................................................................................... 94 

3.1.3 Conceptual Framework ................................................................................................. 94 

3.2 Method ................................................................................................................................ 96 

3.2.00 Empirical Strategy ...................................................................................................... 98 

3.2.01 Research Question 1: Mobility patterns of early-career teachers ........................... 98 

3.2.02 Research Question 1a: School characteristics of movers ....................................... 98 

3.2.03 Research Question 2: Survival Analysis ................................................................. 98 

3.3 Results ............................................................................................................................... 104 

3.3.01 RQ1: Mobility Patterns of Early Career Music Teachers ......................................... 104 

3.3.02 RQ1a: School Characteristics of Movers ................................................................. 105 

3.3.03 RQ2: Survival Analysis ............................................................................................ 106 

3.4 Discussion ......................................................................................................................... 114 

3.4.01 RQ1: Mobility Patterns of Early Career Teachers .................................................... 114 

3.4.03 RQ 2: Survival Analysis ........................................................................................... 115 

3.4.04 Conclusion ................................................................................................................ 119 

Chapter 4. Has the COVID-19 pandemic exacerbated the Maryland music teacher 
shortage? Evidence from a statewide longitudinal data system ........................................... 122 

4.0 Introduction ....................................................................................................................... 122 

4.1 Literature Review .............................................................................................................. 123 

4.1.00 COVID-19 and K-12 Teacher Experiences .............................................................. 123 

4.1.01 Teacher Burnout .................................................................................................... 124 

4.1.02 Teacher Attrition in the Pandemic ........................................................................ 126 

4.1.03 K-12 Music Teacher Experiences ......................................................................... 128 

4.1.04 COVID-19 and the early-career music teacher ..................................................... 131 



vii 
 

4.1.05 Summary ............................................................................................................... 133 

4.1.10 COVID-19 and Preservice Teacher Experiences ..................................................... 133 

4.1.11 General Education ................................................................................................. 133 

4.1.12 COVID-19 and preservice music teacher experiences ......................................... 135 

4.1.2 Teacher Shortages and Alternative Pathways ............................................................ 136 

4.1.3 Summary ..................................................................................................................... 137 

4.2 Method .............................................................................................................................. 138 

4.2.0 Empirical Strategy ...................................................................................................... 139 

4.3 Results ............................................................................................................................... 141 

4.3.1 RQ1a: Provisionally-licensed teachers ....................................................................... 141 

4.3.2 RQ1b: Attrition of early-career teachers .................................................................... 142 

4.3.3 RQ1c: Postsecondary graduates’ transition ................................................................ 144 

4.3.4 RQ1: How has the COVID-19 pandemic impacted the music teacher shortage in 
Maryland? ............................................................................................................................ 147 

4.4 Discussion ......................................................................................................................... 147 

4.4.01 Implications for Music Teachers and Teacher Educators ........................................ 148 

4.4.02 Implications for Music Education Research ............................................................. 150 

4.4.03 Conclusion ................................................................................................................ 152 

Chapter 5. Conclusion .............................................................................................................. 154 

5.1 Policy implications for Maryland ...................................................................................... 154 

5.1.10 Recommendations for Schools/LEAs ....................................................................... 155 

5.1.11 Cultivate mentoring relationships and supportive professional networks ............ 155 

5.1.12 Provide additional supports in low SES schools................................................... 156 

5.1.13 Cultivate a recruitment pipeline with music teacher preparation programs ......... 157 

5.1.2 Recommendations for Music Teacher Preparation Programs .................................... 158 

5.1.21 Music education recruitment and admission practices ......................................... 158 

5.2 Contributions to Music Education Scholarship ................................................................. 159 

5.2.1 Leveraging State Longitudinal Data Systems for music education research .......... 159 

5.2.2 Music Teacher Pipeline and Labor Markets ........................................................... 163 

5.3 Final Reflections ............................................................................................................... 164 

Appendix A ................................................................................................................................ 167 



viii 
 

Appendix B ................................................................................................................................. 168 

References .................................................................................................................................. 169 



ix 

List of Tables 

 

Table 2.1 Workforce entry of Music Education Graduates .......................................................... 52 

Table 2.2 Linear probability models for workforce entry ............................................................. 54 

Table 2.3 School Characteristics of newly employed music teachers .......................................... 55 

Table 2.4 Teaching Responsibilities for new music teachers ....................................................... 57 

Table 2.5 Distance Analyses ......................................................................................................... 57 

Table 2.6 Standardized Mean Differences between employed School and Home School ............ 59 

Table 3.1 Retention/Attrition of Early-Career Music Teachers ................................................. 104 

Table 3.2 Descriptive statistics for school characteristics of Movers ........................................ 106 

Table 3.3 Discrete-Time Survival Analysis ................................................................................. 110 

Table B.1 Logistic Regression model for workforce entry ......................................................... 168 

 

 



x 

List of Figures 

Figure 2.1 Kernel Density plot of distance between hometown and employment school compared 

to national median for all new teachers ....................................................................................... 58 

Figure 3.1. Conceptual framework for teacher turnover .............................................................. 95 

Figure 3.2 Survivor Function ...................................................................................................... 107 

Figure 3.3 SF by Gender ............................................................................................................. 107 

Figure 3.4 SF by School Level .................................................................................................... 108 

Figure 3.5 SF by Race/Ethnicity ................................................................................................. 108 

Figure 3.6 SF by Certification Type ............................................................................................ 109 

Figure 4.1 Provisionally Licensed Music Teachers .................................................................... 142 

Figure 4.2 Early-career music teacher attrition ITS .................................................................. 143 

Figure 4.3 Early-career music teacher attrition ITS (Teachers of Color) .................................. 143 

Figure 4.4 Workforce entry ITS .................................................................................................. 146 

Figure 4.5 Workforce entry ITS (Persons of Color) ................................................................... 146 

 

 



1 
 

Chapter 1. Introduction 

1.0 Overview of the Document 

For decades, many American schools have been unable to fill available teaching positions with 

qualified personnel; educational research suggests the shortage of teachers is a complex and 

multifaceted phenomenon. Early scholarship on the topic began to proliferate in the 1990s (e.g., 

Ingersoll, 1999; National Commission on Teacher and America’s Future, 1997) and continues to 

be addressed in contemporary research. In 1992, an average of 5.1% of teachers left the 

profession each year (Carver-Thomas & Darling-Hammond, 2019). Throughout the 1990s, the 

average rate of teachers leaving the profession did not differ significantly from similar 

professions, such as nurses and social workers (Harris & Adams, 2007), but high rates of 

retirement and growing student enrollment numbers accelerated the demand for new teachers 

(Ingersoll, 1999). By 2008, the rate of teachers leaving the profession each year grew to 8.4%; 

early-career educators had much higher rates of attrition compared to mid-career, delineating an 

issue with the retention of newly recruited teachers (Carver-Thomas & Darling-Hammond, 

2019). Throughout the 2010s, a combination of comparatively low teacher salaries following the 

2009 recession, falling enrollments in teacher preparation programs, and higher rates of teacher 

turnover/attrition exacerbated the shortage (García & Weiss, 2019). Clearly, the problem of the 

teacher shortage has grown over the past decades. Presently, concerns regarding how the 

COVID-19 pandemic and the “Great Resignation” (Parker & Horowitz, 2022) may compound 

the teacher shortage warrant additional considerations. The causes of a shortage in any given 

area are numerous but can be conceived of as leaks along the pipeline for teacher preparation. If 

there are too many leaks along the pipeline, there will not be enough qualified teachers, resulting 

in a shortage. 



2 
 

Music is one such area in which teacher shortages have been identified; as many as 32 

states and the District of Columbia officially designated music as a teacher shortage area 

between the 2011-2012 and 2020-2021 school years (Hash, 2021b), on par with shortages of 

qualified teachers in other subject areas. As recently as the 2021-2022 school year, the State of 

Maryland identified music for all grades PK-12 as a shortage area, underscoring a need for 

policy-relevant empirical work. Broadly, in this document I leverage data from the Maryland 

Longitudinal Data System (MLDS) to address the Maryland music teacher shortage by 

examining two critical stages of the music teacher pipeline: 1) the transition of postsecondary 

graduates into the teaching profession, and 2) the retention/attrition of early-career music 

teachers. The MLDS was established with funds from a federal grant initiative authorized under 

the federal Education Technical Assistance Act of 2002. The MLDS Center has a robust 

infrastructure for collecting, processing, and organizing census-level data for K-12 students and 

teachers, postsecondary students and faculty, and workforce. 

As this dissertation consists of three distinct, but thematically related, studies, I structure 

the document as follows: in the remainder of this chapter, I first contextualize the teacher 

shortage phenomenon as one facet of broader teacher labor markets. Next, I consider how music 

teachers may experience the labor market differently than non-music teachers and discuss how 

the music teacher pipeline is uniquely constrained. Then, I outline the purpose of this document, 

the rationale for addressing this problem with three related studies, and the significance of this 

work. Following, I list my primary research questions and provide a rationale for using the 

analysis of state longitudinal data as the basis for my research design. I then discuss the scope 

and limitations of the current studies. In Chapters 2, 3, and 4, I report three empirical studies, 

each with novel contributions to the research literature. In Chapter 5, I provide a synopsis of the 



3 
 

three studies and policy implications for their collective significance. In this way, I examine the 

historical patterns of the music teacher shortage while looking forward to the future. I conclude 

Chapter 5 by discussing the prospects for continued use of state longitudinal data systems for 

music education research. 

In discussing music education broadly and music teacher labor markets specifically, I 

consider how my positionality as a White male music educator and researcher in the academy 

frames my view on music education, social justice, and equity. My understanding is shaped by 

my experiences teaching in low-income schools and my avid support for public education. I have 

not personally experienced many of the issues of marginalization and exclusion discussed 

throughout this document; my perceptions are informed by my engagement with academic 

scholarship and personal interactions with friends, colleagues, mentors, and students. Thus, all 

that follows is but one framing that colors my interpretation of background, context, data, and 

results that are undoubtedly impacted by my positionality. 

1.1 Identification of the problem under consideration 

1.1.01 Teacher Labor Markets 

Four defining attributes of the broader teacher labor market provide an important contextual 

understanding of the current status of teacher shortages. First and foremost, the nature of the 

teacher labor market is inherently local (Boyd et al., 2005; Krieg et al., 2016; McHenry-Sorber & 

Campbell, 2019; Reininger, 2012). That is to say that more so than other professions, teachers 

tend to work and live in areas closer to where they grew up. Furthermore, teachers tend to work 

in schools and school districts similar to the ones they attended. This local nature necessitates 

that school districts experiencing teacher shortages must consider how to recruit and retain 

educators from within or near their district. Thus, potential pathways and strategies available to 



4 
 

local education agencies (LEAs) may be unique based on differences in proximity to teacher 

preparation programs and postsecondary attendance rates. Secondly, the teacher workforce has 

diversified somewhat, but the demographic composition remains disproportionately White and 

female (Redding & Baker, 2019; Redding & Nguyen, 2020; Reininger, 2012). Efforts to 

diversify the teacher workforce that consider the local nature of teacher labor markets have led a 

number of LEAs to adopt “grow your own” initiatives (Podolsky et al., 2019). In these 

initiatives, LEAs recruit current students to pursue careers as educators so that they may return to 

the district to teach upon completion of their teacher preparation program. Still, the homogeneity 

of the teacher workforce raises questions regarding barriers to equity along the teacher pipeline. 

Thirdly, the retention rate of teachers in their first five years of teaching is approximately 50%, 

although estimates from some locales have been slightly higher (Henry et al., 2012; Joiner & 

Edwards, 2008; Kelley, 2004; Smith & Ingersoll, 2004). Schools and districts have developed 

mentoring and induction programs to support beginning teachers, but the attrition from teachers 

in their earliest years persists. Finally, the growing teacher shortage issues have resulted in a 

multitude of alternative teacher certification pathways in attempts to assuage hiring difficulties 

and fill vacancies (Podolsky et al., 2016). When schools are unable to fill a vacancy with a 

qualified teacher, they must hire individuals on provisional credentials (also referred to as 

conditional licensure, emergency certification, or alternative pathways). The number of teachers 

hired on such licenses then provides a window into the severity of a shortage in any particular 

locale or content area (Goldhaber et al., 2021). 

1.1.02 Music Teachers Labor Markets 

There are several reasons why music teachers may experience labor markets differently 

than other teachers. Compared to the general teacher population, music teachers tend to have 



5 
 

fewer employment opportunities because LEAs employ a limited number of music teachers in 

their schools. Some schools may have only one or two music teachers, and others may split a 

single music teacher among multiple schools, effectively limiting the number of options for 

employment near a music teacher’s home. Compared to non-music teachers, music teachers 

receive certification in a broader range for age and content areas (May et al., 2017). For example, 

a new music teacher may be primarily interested in teaching high school band but accept a job 

teaching elementary general music. It is less common for a teacher certified in secondary math, 

for example, to teach 3rd grade, as those content areas and grade levels require different 

certifications. Thus, a music teacher may leave their current position for employment in a new 

school with opportunities in which they are more interested. At the secondary level, music 

becomes elective for students. Employment for these music teachers is predicated upon having a 

sufficiently large enough number of students enrolled in their courses. Otherwise, an LEA may 

reduce the full-time equivalence of a music teacher or assign the music teacher to multiple 

schools. Finally, music teachers may experience the market differently because, by the time they 

complete their teacher preparation program and are ready to transition into the profession, they 

have persisted through a pipeline with multiple constraints that were not imposed on the broader 

teacher profession (Grisé, 2019). 

1.1.03 A Constrained Music Teacher Pipeline: Leaks and Filters 

In the education field, the majority of aspirant teachers formally enter the pipeline for 

teacher preparation upon enrollment into a four-year teacher preparation program (Barth et al., 

2016), but there are several explicit points at which the music teacher preparation pipeline 

differs. Other content area specializations (e.g., math, science, English, social studies) are 

mandatory course requirements for high school graduation and college admissions, but music is 



6 
 

almost universally elective at the secondary level. Although students may informally engage 

with music outside of school in the form of playing in a band, singing-songwriting, etc., 

admission into a music teacher preparation program is predicated upon passing a formalized 

audition process, most often within a Western Art Music paradigm (Fitzpatrick et al., 2014; 

Grisé, 2019; Koza, 2008). The pathway to being a music teacher is then inextricably linked to 

first being a school music student; it is impossible to separate adolescent access and uptake to 

school-based ensemble music education and later becoming a qualified school music teacher. 

The resulting pool of potential teacher candidates for music then consists of a limited subset of 

individuals compared to other content areas. The pipeline differences between “aspirant music 

educators” (Grisé, 2019, p. 5) and their non-music counterparts necessitate a critical examination 

of all points in which a potential music teacher may transition into or out of the teacher 

preparation pathway. 

The pathway toward becoming a music teacher begins long before enrolling in a teacher 

preparation program, and systemic barriers to access and issues of inequity are entwined with the 

pipeline the whole way through (Elpus, 2015; Fitzpatrick et al., 2014; Grisé, 2019). The first 

stage of the music teacher pipeline in the United States begins with enrollment in an ensemble-

based music program, often at the elementary or middle school level, and has additional filters 

along the following stages: 1) application, acceptance, and matriculation into a postsecondary 

institution, 2) successful audition/placement into a music program, 3) persistence through degree 

completion, 4) entering the teaching profession, and 5) retention in the profession. 

At and between each stage of the pipeline, there are what I refer to as leaks and filters. 

Leaks indicate an indiscriminate process resulting in an individual self-selecting out of the 

pipeline. Filters, on the other hand, represent structural and systemic barriers to equity; some of 



7 
 

these filters are covert, while others are overt. I characterize as “overt” those filters that are 

readily visible: perhaps a school offers fewer music opportunities for students, students lack the 

financial resources to participate in school music programs, or students are denied admission to a 

postsecondary institution based on academics, musical audition, or financial strain. Conversely, 

“covert” filters are more subtle to the external observer, despite being readily felt by the students 

being filtered out of the pipeline. Consider how leaks and filters influence the following two 

hypothetical students: 

Emily participates in school orchestra, works hard, but cannot afford private lessons. 

After a trial lesson with a university violin professor, the professor gently lets Emily know 

they are not yet up to the performance standards for admission and recommends taking 

lessons, practicing intensely, and acquiring a nicer violin before a formal audition.  

Joshua is from a low-income family and spent much of their life learning programming 

skills to support future career aspirations. In high school, they excelled as a saxophonist 

and regularly participated in auditioned honors ensembles. When talking with their 

parents about career aspirations, they floated the idea of music education. Joshua’s 

parents dismissed the idea outright; they want Joshua to stick to their much more 

lucrative plan of computer programming and software engineering so Joshua can 

support them financially.  

In Emily’s case, the clearly defined overt filter of failing an audition never transpired because 

she was covertly filtered before ever registering to audition. For Joshua, the overt filter of 

financial resources did not prevent them from participating in band, but financial considerations 

covertly filtered Joshua out of the pipeline because of family support. Discussing issues of the 

music teacher shortage requires scrupulous examination of these pipeline filters; through 



8 
 

judicious reflection, one may discern the inequitable outcomes these filters may impose upon 

individuals as one seeks to cultivate more equitable processes. 

Potentially aspirant music teachers first enter the music teacher pipeline, or are filtered 

out, at the point in which music becomes elective. For most students in the United States, this 

occurs at the secondary level, where the most common courses available are band, choir, or 

orchestra (Elpus, 2017). In some extreme cases, the time at which music becomes elective may 

be as early as fourth grade, at which point whether or not a student will continue in ensemble 

music courses through high school may be determined. One observable result of the elective 

nature of music education in the upper elementary and secondary grades is a disparity in the 

representation of students by race and ethnicity (Elpus, 2022b; Elpus & Abril, 2011; 2019; 

Miller, 202) Consistently, demographic analyses of music students have demonstrated that those 

who elect music study are systematically different from those who do not elect music study. 

Broadly speaking, music students tend to be more White and more affluent than their non-music 

counterparts; these disparities are especially prevalent when comparing band and orchestra 

students to the general student population. Exacerbating inequity is the “significant racial 

inequality abyss” (Robinson & Hendricks, 2018, p. 30) of American music education; in addition 

to disproportionately White music teachers/teacher candidates, music education opportunities, by 

and large, center the White experience. These demographic analyses demonstrably underscore 

clear inequitable filtering at the earliest stage of the music teacher pipeline. 

The next three stages, matriculation into a postsecondary institution, successful audition 

into a postsecondary music program, and persistence through degree completion, are potentially 

the moments where the inequitable filtering is most overt. Marginalized populations face 

additional challenges in postsecondary attendance and completion in the forms of academic, 



9 
 

social, and economic capital (Fitzpatrick et al., 2014). Within music programs, systematic issues 

of inequity in the audition process, and consequently the pre-audition lives of aspirant music 

teachers, are abundant. Audition preparation can be a long and arduous process, sometimes 

beginning well over a year prior to the audition. Often, K-12 music experiences alone do not 

adequately prepare students for a successful audition, and students without access to private 

lessons are disadvantaged in this regard (Koza, 2008). Upon entry into the music education 

degree program, students enroll in a rigorous academic program consisting of music theory, 

history, lessons/ensembles and other performance requirements, music education coursework, 

and general education graduation requirements. Often, many of these courses are sequential in 

nature with a chain of prerequisites that must be passed before progressing to the next level. 

Thus, failing to receive an adequate passing grade in one course may set back graduation by an 

entire year or dissuade one from persisting in the degree entirely. 

Upon successfully completing a music education degree and obtaining teaching licensure, 

the next stage in the pipeline is entrance into the music teaching profession. Although most states 

license music teachers for all music grades PK-12 (May et al., 2017), new music teachers may 

have preferences for specific content areas and/or grade levels for their first teaching placement 

(Hellman, 2008; Kelly, 2003). Currently, the vast majority of music teacher responsibilities fall 

into one of four categories: general music, band, choir, or orchestra (Elpus, 2017). Even when 

other music courses, such as guitar, piano, or modern band are offered at a school, the teaching 

responsibilities for such courses tend to be added to the responsibilities of a teacher who is 

responsible for one or more of the primary areas. As LEAs employ a limited number of music 

teachers in their schools, new music teachers may have fewer employment options that match 

their desired grade level and content area. If there are no positions of interest open to a graduate, 



10 
 

or the graduate is not hired for their top choices, the graduate is faced with two options: seek 

employment in less desirable positions, or delay/suspend entry into the teacher workforce. Thus, 

while the music teacher shortage in Maryland and elsewhere indicate the availability of teacher 

positions, the potential mismatch between the nature of vacant positions and the career 

goals/interests of new music teachers may leave these vacant positions unfilled. Pipeline filtering 

at this stage quietly influences if and how graduates enter the workforce.  

Sue recently graduated with her music education degree and is interested in teaching 

high school orchestra. She applied for the few vacancies in the districts around her 

hometown, but they were ultimately filled by other candidates. There were several other 

elementary general music vacancies that went unfilled, but Sue decided to delay entry 

into the teaching profession rather than apply for jobs that she did not find appealing. 

Although this filtering may first appear to be an indiscriminate leak along the music 

teacher pipeline, broader systems of inequity persist absent any individual’s decision. Orchestra, 

for example, is less likely to be offered in smaller schools and schools with a higher proportion 

of students receiving free/reduced lunch (Elpus, 2022a). In an instance where a student from a 

small high school with a high proportion of students receiving free/reduced lunch school 

managed to enroll in a school orchestra program and persist along the music teacher pipeline 

through this point, they are now faced with fewer opportunities that match their desired position 

compared to those from large, affluent, suburban schools. At a surface-level examination, Sue 

may have been characterized as too picky in her job expectations, that she self-selected out of the 

music teacher pipeline. However, deeper reflection on the broader structural inequities along the 

music teacher pipeline revealed the otherwise covert filter that prevented her entry. 



11 
 

The final stage of the pipeline, retention of music teachers in the teaching profession, is 

of paramount importance to mitigate the teacher shortage. One of the largest contributing factors 

to the rising demand for teachers nationally is the attrition of early-career teachers (Carver-

Thomas & Darling-Hammond, 2019), a trend mirrored by music teachers (Hancock, 2008). 

Retaining early-career music teachers can then have the most immediate impact on the number 

of music teacher vacancies. A music teacher may leave the profession for a multitude of reasons, 

such as lack of administrative support or low teacher self-efficacy (Hancock, 2008), but retention 

can be improved via mentorship and collegial networks (Conway, 2002). Policy levers aimed at 

keeping current early-career music teachers in the profession can impact the teacher shortage 

with almost immediate effects. 

Potential filters that push early-career music teachers out of the profession can remain 

covert, their operation obscured behind a web of entangled factors. Teacher gender, race, and 

ethnicity are generally not related to turnover decisions (Nguyen et al., 2019). Instead, several 

school-level factors related to job satisfaction, such as overall school climate and administrative 

support, are more strongly associated with a teacher’s decision to leave or remain in a school 

(Nguyen et al., 2019; Nguyen & Springer, 2021). From an equity perspective, it may first appear 

that the systems that filter out new teachers are indiscriminate, with individuals varying along 

gender and racial/ethnic lines no more or less likely to move or leave than others. However, 

inequitable systems are just that: systems that permeate different structures, often invisibly or 

implicitly. Teachers of color are more likely to teach in schools with higher turnover rates and to 

teach in school districts with lower potential salary earnings (Carver-Thomas & Darling-

Hammond, 2019). Policy levers to foster retention of early-career teachers should then 

acknowledge these barriers because “effectively retaining teachers is crucial to make sure there 



12 
 

are enough well-prepared and committed teachers” (Carver-Thomas & Darling-Hammond, 2019, 

p. 19) for all schools and students. 

1.1.04 Great Resignation 

The pandemic shocked social, economic, and education systems throughout the United 

States and the rest of the world. Crisis narratives sounding the alarm on teacher shortages have 

dominated the media (e.g., ABC News, 2022; Natandson, 2022), but emotionally compelling 

narratives cannot replace empirical evidence when considering policy. With an eye toward the 

future, the extent to which the teacher shortage has contributed towards or has been shielded 

from the ‘Great Resignation’ warrants investigation. What has been true regarding the teacher 

shortages across the country historically needs to be re-examined so that the profession may plan 

for the coming years. 

1.1.05 A gap in the literature: policy-relevant empirical work 

To date, there have been no empirical studies examining the individual workforce 

transition of music education graduates and retention/attrition of early-career music teachers 

using state-level administrative data. A recent call for such scholarship (Austin, 2022) 

acknowledged the capacity of such research to advance the current understanding of potential 

music teacher shortages. An important nuance to the issue of teacher shortages is that shortages 

are demonstrably localized (Thompson, 2022), and thus examinations of local contexts, such as 

an individual state, provide the most actionable evidence from a policy perspective. Although the 

results of the three studies comprising this dissertation may not be wholly generalizable to other 

states, they provide a methodological blueprint for how such investigations may transpire in 

other locales. 

1.1.06 Maryland Context 



13 
 

Among states currently experiencing a shortage of music teachers, Maryland offers a 

particularly interesting case in which to examine the final two stages of the music teacher 

pipeline. Maryland is one of the most educated states in the country 

(https://worldpopulationreview.com/state-rankings/most-educated-states), there is broad support 

for education, and teacher salaries rank in the top 10 of the country (see 

https://worldpopulationreview.com/state-rankings/best-states-for-teachers). As recently as 2021, 

the State of Maryland passed landmark legislation to increase education funding by $3.8 billion 

each year for the next decade, including opportunities for increased salary earnings among 

teachers, to strengthen the education system (Blueprint for Maryland’s Future, 2021). Despite 

having in place many of the policies commonly advocated to promote teacher recruitment and 

retention, Maryland still has experienced a shortage of music teachers. Consequently, an 

examination of the final two pipeline stages in Maryland would yield pertinent insights into these 

critical periods within a context that is largely supportive of music teachers, disentangled from 

some of the factors related to pipeline attrition outside of an aspirant or current teacher’s locus of 

control. 

Among music teacher positions particularly, several distinctive characteristics of 

Maryland contextualize the findings from these studies. First, most LEAs in Maryland are large 

county-wide systems. In total, there are five large suburban LEAs that serve over 60% of the 

student population positioned along and around the Baltimore-Washington D.C. corridor, 18 

rural county districts, and one large urban district. An important consideration about large county 

districts compared to smaller independent school districts is that teachers may have greater 

flexibility to transfer schools within a district without resigning from their current position. Prior 

research at the high school level has indicated that compared to national averages, Maryland 



14 
 

schools and LEAs offer more diverse music courses at a greater number of schools, particularly 

in the suburban districts (Miller, 2023). Furthermore, teachers at these high schools frequently 

teach more than one content area specialization, such as band and orchestra, orchestra and guitar, 

chorus and piano, etc. Elementary students in many Maryland schools have the option for 

elective ensemble music enrollment, often taught by itinerant specialists.  

Finally, aspects of Maryland’s state longitudinal data system, the MLDS, provide several 

rich analytical opportunities that are not necessarily available in other states’ SLDSs. The MLDS 

data from MSDE contains course rostering information, which is not currently available in every 

state (Bloom-Weltman et al., 2021). Course rostering data allowed me to identify the exact 

teaching responsibilities of music teachers for the years from which course data were available 

(starting in 2013), rather than solely examining the school of employment and licensure area. 

Consequently, the music teachers examined in these studies were all teachers who actively taught 

music courses. The longevity of the MLDS provides another reason for which to study 

Maryland; postsecondary and teacher workforce data are available from as early as 2008. As the 

federal government first awarded grants for SLDS development beginning in 2005, Maryland 

has coverage spanning a broader period than many other states.  

1.2 Purpose and Research Questions 

The purpose of these studies was to explore the workforce transition of Maryland postsecondary 

music education graduates, the retention/attrition patterns of early-career music teachers, and the 

extent to which both were impacted by the COVID-19 pandemic. Fully understanding these 

dimensions of the teacher shortage necessitates understanding the systemic barriers to equity at 

every point. A shortage of music teachers necessarily indicates that K-12 students may be denied 

a high-quality music education due to societal inequities far beyond their sphere of agency. As 



15 
 

previously discussed, the music teacher pipeline has a multitude of filtering points, starting with 

access to music education. The music teacher shortage then can exacerbate inequities through a 

recursive cycle: the most marginalized school communities are more likely to be impacted by 

shortages (Gitomer et al., 2019) and denied a quality music education, thus students from these 

communities are filtered out of the pipeline at the earliest stages. Because teachers prefer to work 

close to their hometown (Boyd et al., 2005; Reininger, 2012), it can be difficult for these school 

communities to recruit new teachers; the recursive cycle thus continues. By examining the final 

two stages of the music teacher pipeline historically and with an eye toward the future, I 

illuminate points of equity and inequity at the stages where policy can impact the most 

immediate change. The following research questions guided these inquiries. 

1.2.1 Chapter 2: The workforce transition of music education graduates in Maryland 

RQ1: What proportion of music education graduates from Maryland postsecondary 

institutions enter the teacher workforce? 

RQ1a: What teacher-level characteristics are associated with entry into the 

teaching profession? 

RQ1b: What are the teaching responsibilities (i.e., band, choir, orchestra, guitar, 

general music, and the number of students enrolled) and characteristics of 

schools/LEAs (i.e., size, grade level, school demographics, urbanicity, etc.) in 

which new music teacher graduates obtain their first teaching assignment? 

RQ1c: To what extent are the schools of new music teachers’ first teaching 

assignments similar to their attended high school? 

1.2.2 Chapter 3: Retention and attrition of early-career music teachers in Maryland: A 

survival analysis  



16 
 

RQ2: What are the retention and attrition patterns of early-career music teachers in 

Maryland public schools? 

RQ2a: For those who move schools, what are the characteristics of the schools 

they leave and the schools to which they move? 

RQ2b: What teacher-, school-, and LEA-level factors are associated with whether 

and when an early-career music teacher leaves teaching in Maryland? 

1.2.3 Chapter 4: Has the COVID-19 pandemic exacerbated the Maryland music teacher 

shortage? Evidence from a statewide longitudinal data system 

RQ3: How has the COVID-19 pandemic impacted the music teacher shortage in 

Maryland? 

RQ3a: How has the COVID-19 pandemic impacted the workforce transition of 

new music education graduates in Maryland? 

RQ3b: How has the COVID-19 pandemic impacted the retention and attrition 

patterns of early-career music teachers in Maryland? 

RQ3c: How has the number of music teachers who are not certified in music 

changed since the start of the pandemic? 

1.3 Research Design: Analysis of State Longitudinal Data Systems 

To answer these questions, I leveraged data from the Maryland Longitudinal Data System 

(MLDS). Large government data sources, such as SLDS, are underutilized in music education 

research, but “analyses of large-scale government datasets can have many benefits” for music 

education research (Miksza, 2022, p. 363). I limited analyses to data within a singular state 

because broad national data obscure the variation within states, national-scale data have limited 

generalizability for substantive policy at the state level (Goldhaber & Holden, 2020), and the 



17 
 

nature of teacher labor markets are inherently local (McHenry-Sorber & Campbell, 2019; 

Reininger, 2012). As such, the data collected under the auspices of an SLDS are a rich source of 

information that can yield policy-relevant results of particular value to the state from which the 

data were collected. 

Although exact regulations and requirements may vary by state, researchers may gain 

access to SLDS data by submitting a formal data/research request. SLDSs contain sensitive data, 

and access to data may require researchers to submit to background checks and complete security 

training. In some instances, an SLDS may provide a qualified researcher access to the raw data 

themselves, provided that the researcher satisfies all background and training requirements. If 

researchers have plans to engage regularly with the SLDS for ongoing research projects, the 

organization may establish a staff researcher position. Data and research requests must usually fit 

within an organization’s research agenda, legislative mandate, or other sets of requirements. In 

many cases, obligations to an SLDS must be completed prior to work on further developed 

projects, such as manuscripts for publication or conference presentations.  

For my ongoing work with several MLDS research projects, I have an established 

relationship as an MLDS staff researcher, which required a federal background check, 

fingerprinting, and ongoing data security training. In this capacity, I collaborate with researchers 

and stakeholders across the education field in the State of Maryland, leveraging MLDS data for a 

variety of research projects. As part of my ongoing engagement, I have a secure virtual machine 

through which I pull data tables from the operational data store (ODS), write coding scripts to 

clean and analyze data, and prepare all work for suppression review (the MLDS Center’s 

processes of reviewing all analytical results to ensure compliance with data confidentiality to 

limit deductive disclosure risk). Still, all new research projects must 1) go through the formal 



18 
 

proposal process, and 2) comply with MLDS Center’s legislative mandate requiring that research 

proposals be cross-sector and cross-agency. Written research proposals to the MLDS are first 

submitted to the internal research team for review, followed by a formal presentation and 

feedback. Next, project proposals are presented to the whole research and policy advisory board, 

after which the board may approve or reject a research proposal. Cross-sector research 

constitutes research that uses data spanning across major transitions, such as K-12 to 

postsecondary or postsecondary to workforce. Cross-agency research constitutes research that 

uses data collected and reported under the responsibilities of different agencies, such as the 

Maryland State Department of Education, the Maryland Higher Education Commission, or the 

Department of Labor. At the completion of each project, I submit the MLDS Center Product, 

which may be a research presentation, research report, policy brief, or another artifact beneficial 

to the State of Maryland. All further developed work must be reviewed prior to public 

dissemination. 

1.4 Significance 

The results of these studies yielded substantive policy implications for music teacher education 

in the State of Maryland and unique substantive and methodological contributions to the broader 

music education research field. First, understanding the workforce transition of recent music 

education graduates provides insight into the other most immediately impactful tool to address 

the music teacher shortage. By addressing how many graduates find employment, the factors 

associated with employment, and the characteristics of schools in which they find employment, 

teacher preparation programs and schools/LEAs can work towards equitable placement 

outcomes. Secondly, understanding the labor mobility patterns of early-career music teachers 

provides schools, LEAs, and teacher preparation programs with invaluable insight into whether 



19 
 

and when a new teacher may move schools or leave public school teaching in the state. By 

identifying teacher-, K-12 school-, and LEA-level factors associated with such decisions, 

targeted policy interventions to address these factors can be explored to improve retention. 

Finally, the COVID-19 pandemic has pervaded nearly every aspect of life and spurred an 

unprecedented level of labor market movement known as the Great Resignation (Parker & 

Horowitz, 2022). Although narratives of teacher flight have dominated sensational news media 

(Thompson, 2022), the evidence from the present studies suggests that early-career music 

teachers have yet to contribute to such movement. Importantly, the present studies are limited in 

scope to early-career music teachers; more empirical research is needed to quantify the extent to 

which the teacher shortage and other aspects of music education have or have not been impacted 

by the pandemic. 

Methodologically, this dissertation contributes to the music education scholarship field in 

several ways. Researchers have previously studied these issues using a variety of methods, such 

as survey, qualitative inquiry, and secondary analysis. Although these approaches have distinct 

benefits, they generally come with the tradeoff of limited capacity to inform policy. The use of 

SLDSs, while inherently limited based on data available within, provide an additional 

methodological approach with which to probe a variety of questions of broad interest to the 

music education research community. Currently, music education researchers have not taken 

advantage of the prolific data available within an SLDS. Earlier, I detailed an overview of the 

generalized process for gaining access to SLDS data. In each study, I document the processes for 

working with the data for analytic research, providing a blueprint for future research. The 

analytical techniques used within this document provide salient and timely contributions to the 

music education scholarship literature as well. Linear probability modeling provides an intuitive 



20 
 

interpretation for analyses with binary outcomes. Survival analysis techniques and the 

interrupted time series design are underutilized in music education research but may be 

employed in future research studies, particularly when leveraging an SLDS as the data source. In 

Chapter 5, I discuss how music education researchers may leverage SLDS data for future 

research. 

1.5 Scope and limitations of the study 

This study is delimited to the context and years for which relevant data are available, the State of 

Maryland between 2008 and 2021. General trends from previous years or other locales may be 

similar to those found in the current study. Although the results from this study may inform 

future practices and policy in Maryland and elsewhere, readers should consider how other 

contexts may be systematically different from Maryland. Two key limitations also must be 

considered. First, analyses are limited to the data observable in the MLDS. The quantitative 

nature of large-scale data analyses obscures the nuance and lived experiences of individuals on 

their pathway towards and away from music teaching. Secondly, the analyses of the music 

teacher pipeline within this document are limited to just the final stages of the pathway, after 

many potentially aspirant music teachers have been filtered out. As such, several systemic 

barriers along the music teacher pathway remain unexamined. To fully excavate the music 

teacher pipeline and remove the filters requires monumental work by all those in the music 

education field: researchers, teachers, teacher educators, and policymakers alike. Cultivating a 

truly equitable system of music education requires tenacity, perseverance, inward reflection, 

outward action, and coming to terms with the colonizing past and present of music education so 

that the profession may decolonize practices into the future. This dissertation is but one link in a 

long chain of scholarship past and future towards this end, one that I hope invites other scholars 



21 
 

to leverage the ideas and techniques contained within to pursue a more equitable music education 

for all students.  



22 
 

Chapter 2. The workforce transition of music education graduates in Maryland 

 2.0 Introduction 

Policy briefs, media reports, and lived experiences of those in K-12 education all point toward 

the same issue: school districts have increasing difficulty in hiring enough qualified teachers to 

fill their vacant positions (García & Weiss, 2019; Sutcher et al., 2016). Generally referred to as a 

teacher shortage, much of the empirical work on this topic has historically focused on labor 

market patterns of teachers currently employed (i.e., whether teachers stay in the same school, 

move schools but stay in the profession, or leave the profession entirely). In the past decade, 

however, an increased amount of scholarship on the role of teacher preparation programs has 

illuminated a multitude of factors that are related to decisions to pursue education degrees and 

enter the teacher workforce, including teacher candidate demographics, academic achievement, 

hometown geography, student teaching experiences specifically and teacher preparation 

programs broadly, as well as barriers to entry for the teaching profession, highlighting issues of 

equity among potential teachers. 

Although there has been a growing body of evidence regarding the teacher workforce 

generally, there is little empirical work regarding music education graduates’ entry into the 

teacher workforce. There are several potential reasons why music teachers may experience the 

teacher job market in systematically different ways compared to the general education 

workforce, such as the elective nature of music courses and the comparatively few number of 

music teachers employed by any particular school or district. These potential differences 

underscore the need for empirical research focusing on the transition of music education 

graduates into the teacher workforce (Austin, 2022). Furthermore, the labor market for teachers 

is local (Reininger, 2012), and thus the use of national data has substantive limitations for 



23 
 

informing policy at the state level (Goldhaber & Holden, 2020). Thus, the purpose of this study 

was to investigate factors associated with post-graduation employment in K-12 settings of those 

graduating with music education degrees from teacher preparation programs in the state of 

Maryland.  

2.1 Literature Review 

In this literature review, I first provide an overview of the extant literature regarding the 

transition into the teaching profession in the general education literature. Specifically, I 

summarize the literature on demographic characteristics of those who pursue careers in teaching, 

the impact of hometown and student teaching experiences on entry into the profession, and 

issues of equity for aspiring professional educators. Then, I discuss music education research 

related to the characteristics of music education majors and licensure candidates, including 

demographics, motivations for teaching, and where they want to teach. Finally, I discuss music 

teacher preparation programs, state licensure practices, and issues of equity in music education 

regarding barriers to entry for undergraduate programs and barriers to licensure/transition into 

the workforce. 

2.1.00 General Education 

2.1.01 Demographic characteristics of those pursuing careers in teaching 

Redding and Nguyen (2020) analyzed data from seven waves of the National Center for 

Education Statistics (NCES) School and Staffing Survey (SASS) between the 1988-1989 and 

2012-2013 academic years, as well as each available Teacher Follow-Up Survey (TFS), to 

examine changes in demographics of beginning teachers. Descriptive statistics showed the 

gender composition of beginning teachers changed slightly between 1988 and 2012 with a slight 

decrease from 26% male teachers in 1988 to 23% male teachers in 2012. Additionally, the 



24 
 

proportion of new teachers identifying as White was down to 80% in 2012 compared to 87% in 

1988, but trends in racial/ethnic changes had stagnated since 2000. Moreover, the increase in 

representation of teachers of color was largely driven by increases in the number of Hispanic 

teachers. Finally, the academic background of new teachers has also changed. A larger portion of 

new teachers had a graduate degree (22% in 2012 compared to 9% in 1988), more were fully 

certified (96% in 2012 compared to 80% in 2000), and more attended highly selective colleges 

(35% in 2012 compared to 12% in 1988). Regarding the schools in which first-year teachers 

worked, the authors found that in recent years, new teachers were more likely to be placed in 

schools with higher proportions of students receiving free/reduced-price lunch (FRL) and high 

proportions of students of color. However, after considering broader demographic shifts within 

the K-12 student population, differences within student populations between new and 

experienced teachers were diminished in 2012 after about a decade of observed differences 

between 2000 and 2008. 

Redding and Baker (2019) analyzed data from the 2008-2012 cohort of the Baccalaureate 

and Beyond Longitudinal Study to examine racial and ethnic differences in those who pursue 

degrees in education. This nationally representative survey was administered by the NCES to 

those who graduated with an undergraduate degree within the 2007-2008 academic year in three 

waves: the first wave was during the final undergraduate year, the second wave was in the year 

immediately following, and the final wave surveyed participants four years after graduation. 

Their dependent variables of interest were whether a graduate majored in education and whether 

a graduate had ever worked as a teacher, either in a full-time, supportive, itinerant, or substitute 

capacity. Results of their analysis indicated that White college graduates were nearly twice as 

likely to have majored in education compared to Black, Latinx, and other persons of color when 



25 
 

controlling for individual demographic, academic, socioeconomic, and institutional differences. 

Furthermore, even when controlling for state-level heterogeneity through fixed effects, 

racial/ethnic gaps in entry to the teaching field persisted: those identifying as Black had a 6.4 

percentage point lower probability of entering teaching than those identifying as White, and 

Latinx graduates had a 2.9 percentage point lower probability of entering teaching than those 

identifying as White. However, when controlling for degrees in education, substantive 

differences between racial/ethnic groups were mostly diminished, suggesting that some of the 

barriers to equity within the education field begin prior to college graduation, such as 

matriculation into college or major selection within an institution. In line with expectations, 

holding a degree in education was the strongest predictor of entry into the teaching field, with 

education graduates 65.6 percentage points more likely to join the teacher workforce than those 

without education degrees. 

2.1.02 Student teachers: demographics and career plans 

In an analysis of prospective teachers in a large urban district, Ronfeldt et al. (2013) 

surveyed candidates’ pre- and post-student teaching to examine how student teaching preparation 

and teacher characteristics impacted student teachers’ career plans and self-perceptions of 

instructional quality (N = 1,002). The authors examined four student teacher cohorts from Fall 

2008 to Spring 2010 and linked survey responses to district-level administrative data for each 

school. A large majority (81%) of candidates identified as female, 67% were White, and their 

average age was about 31 years old. The higher average age was possibly due to a large minority 

(40%) seeking a graduate degree. Overall, teacher characteristics were most closely related to 

career plans. For example, those identifying as White expressed lower preferences to work with 

underserved populations compared to Hispanic/Latinx candidates. Candidates identifying as 



26 
 

Black were about one-third as likely to plan on teaching for their whole career compared to 

White candidates. Candidates who attended high school in the same district as their student 

teaching placement had odds three times higher for planning to stay in that district for their 

whole career compared to those who attended high school in a different district. Finally, higher 

measures of GPA were associated with lower total planned years for teaching. Overall, school 

characteristics of student teaching placement did “not predict perceptions of preparedness, 

teacher efficacy, or career plans” (p. 328). A notable exception, however, was that candidates 

who student taught in schools with higher proportions of limited English proficiency students 

“predicted stronger preferences to teach underserved students” (p. 331). Finally, candidates with 

better perceptions of cooperating teacher quality and more autonomy in their teaching 

experiences had stronger measures of teaching efficacy at the end of their student teaching. 

In additional analyses of the same data, Shirrell and Reininger (2017) examined the 

extent to which student teachers’ planned persistence in teaching changed throughout their 

internship as a function of school working conditions. Overall, 61% of student teachers did not 

change their planned length of teaching throughout their internship experience. When using 

teacher stability rates as an indicator of working conditions, student teaching in a school with the 

highest rates of teacher turnover predicted a lower total number of years for which an intern 

planned to teach. Furthermore, approximately 10% of student teachers in this sample had 

initially planned to teach in the same district upon graduation but, upon finishing student 

teaching in schools with challenging working conditions, indicated they would never plan on 

teaching in that district. 

2.1.03 Geographic preferences of teachers: local labor markets 



27 
 

 In a seminal study, Boyd et al. (2005) linked data for every new teacher hired in a New 

York State public school between 1998-2002 to information about the hometown, academic 

qualifications, and preparing institutions (N = 33,465). Using descriptive statistics and logistic 

regression, the authors found that teachers demonstrated a strong preference for teaching close to 

their hometown and locales most similar to their hometown. Based on zip code centroid distance, 

teachers were twice as likely to teach within five miles of their hometown compared to 20 miles 

away from their hometown. They were three times as likely to teach within 25 miles of their 

hometown compared to 80 miles away. Distance from preparing institutions was also a factor, 

but it was not nearly as strong a predictor as the distance from hometown. Even when controlling 

for distance (as well as other characteristics), teachers also had a strong preference to teach in 

locales most similar to their hometown. This effect was strongest for those who grew up in 

suburban areas: a teacher who grew up in a suburb was nearly five times as likely to work in that 

suburb compared to the urban portion closer to the city center, and they were about three times as 

likely to work in their home suburban region compared to a suburb from a different metro area. 

The preference for working close to home was not a unique characteristic of teachers in 

New York state, but across all college graduates pursuing a career in teaching nationally 

(Reininger, 2012). Using the National Education Longitudinal Study of 1988 (NELS:88/00, a 

nationally representative two-stage stratified cluster sample of the 1988 8th grade cohort), 

Reininger (2012) analyzed the geographic mobility of new teachers as compared with college 

graduates pursuing workforce employment elsewhere. Additionally, Reininger linked data from 

the CCD, a public-use repository of school- and district-level data, to provide additional 

information about the schools that respondents attended between 1988 and 1992. Among college 

graduates, teachers moved substantially less far away from home compared to non-teachers: the 



28 
 

median distance moved by teachers was 13 miles compared to the 54-mile median distance of 

non-teachers. This relationship held across all urbanicities; the median distance moved for 

teachers was consistently less than the median distance for non-teachers. Furthermore, 60% of 

teachers were living local to their hometown (defined as within a 20-mile radius), whereas only 

42% of non-teachers were local. With respect to issues of teacher shortages, particularly in hard-

to-staff schools, perhaps the most salient consequence of this study is that when conditioning on 

the number of high school students who earn a bachelor’s degree, high schools with a higher 

percentage of FRL and students of color would produce as many teachers as schools with more 

affluent and White students. Given teachers’ preference to work close to home, and the smaller 

portion of high school students from these areas obtaining undergraduate degrees, “the lack of 

local supply [of teachers] may be perpetuating the cycle in these disadvantaged schools” 

(Reininger, 2012, p. 141). 

Whereas previous geographic studies limited analyses to hometown geography and first 

teaching placement, Krieg et al. (2016) also considered the geographic proximity of student 

teaching as a factor. Analyzing the teaching placements of 8,527 student teachers from six 

teacher preparation institutions in the state of Washington, Krieg and colleagues found that while 

proximity to hometown was a factor in first teaching placement, proximity to student teaching 

placement was a more powerful predictor of first teaching placement. Whereas 54.3% of newly 

hired teachers taught within 25 miles of their home district, about 75% of all hired student 

teachers taught within 25 miles of their student teaching placement. Further, interns with higher 

levels of academic qualifications, as measured by college GPA, were more likely to be placed in 

more advantaged schools/school districts. Although the relationships discussed are only 

associative as the analysis was descriptive in nature, the pattern of “student teaching placements” 



29 
 

may ostensibly “contribute to inequitable distribution of teacher” (Krieg et al., 2016, p. 365) 

quality across various schools or districts. 

In an extension of this line of inquiry, Goldhaber et al. (2021) connected district-level 

teacher shortages to the geographic proximity of student teachers from 15 teacher education 

preparation programs in the state of Washington. To construct their analytic data set, Goldhaber 

and colleagues linked newly hired teacher data from the Washington State’s Office of 

Superintendent of Public Instruction to student teacher data from the above-mentioned preparing 

institutions from 2009-2017 (N = 38,948). First, they used the portion of new hires made on 

emergency credentials as a proxy for staffing shortages. Then, they specified a naive logistic 

regression model of the bivariate relationship between the probability of staffing a position with 

emergency credentials to the proportion of district teachers who hosted a student teacher the 

prior year. However, this model was potentially biased by spillover effects: districts adjacent to 

or nearby others that hosted a higher proportion of student teachers would potentially benefit 

from their proximity. Thus, they defined spatial-weighting matrices to account for weighted 

proximity to nearby districts based on the number of student teachers and the reciprocal of the 

distance between districts. Results indicated that for each one-percentage-point increase in the 

proportion of teachers who hosted a student teacher within that district, there was a .22 

percentage-point decrease in the percentage of teachers newly hired on emergency credentials, 

equivalent to a marginal effect of approximately 20%. When accounting for geospatial distances, 

a one-percentage-point increase in the proportion of teachers hosting a student teacher near a 

district was associated with a .45 percentage-point decrease in the percentage of teachers newly 

hired on emergency credentials. Importantly, the within-district effect of student teachers was 

attenuated when accounting for district proximity, suggesting that student teaching placements 



30 
 

within and nearby are predictive of future staffing shortages. These results were robust to 

controls for proximity to teacher education programs. 

National aggregate data on characteristics of the aspirant/first-year teacher workforce 

provide an important window into the past and present status of the teaching profession at large. 

Taken together, these nationally representative samples of new teachers (Redding & Nguyen, 

2020) and recent college graduates (Redding & Baker, 2019; Reininger, 2012), underscored by a 

local examination of teacher workforce entry (Boyd et al., 2005; Goldhaber et al., 2021; Krieg et 

al., 2016) have revealed two critical factors. First, while the demographic composition of the 

teacher workforce is diversifying, it is not keeping pace with changes in student populations. 

Secondly, prospective teachers’ hometown geography and urbanicity, their geographical 

preferences, and other personal/school characteristics that are unique to the career decisions of 

teachers pose a specific set of challenges: across the country, these amount to systemic issues 

nationwide regarding equity within the teacher pipeline prior to any decision to major in 

education, which I discuss in more detail in section 2.1.5. Importantly, student teacher intentions 

do not necessarily translate into actual workforce entry for a variety of reasons. To better 

understand the future of the teaching profession, I now examine the transition between student 

teaching and workforce entry. As the aggregate data may obscure trends within states and given 

the local nature of the teacher labor market (McHenry-Sorber & Campbell, 2019; Reininger, 

2012), it is important to understand the numerous factors associated with the if and how teacher 

candidates may enter the profession at the district and state levels. 

2.1.04 Student teachers’ entry into the teaching profession 

Goldhaber et al. (2014) analyzed a unique longitudinal data set from six teacher 

preparation programs in the state of Washington, following 8,080 teacher candidates who 



31 
 

completed student teaching between 1998 and 2009. Of teacher candidates in their analytic 

sample, 5,674 were observed as teachers in public schools, teachers in private schools, or 

otherwise employed in non-teacher roles in public school settings by 2011, the final observed 

year in the data set. Their results suggested that almost 30% of student teachers did not find 

employment in Washington Schools, consistent with national estimates. Because observations 

were right-censored, there is a chance that some student teachers delayed entry into the teaching 

profession for several reasons, such as family or graduate school, but would eventually return. 

Student teachers may also have taken positions out of state, which would not be observable in 

the authors’ data. Goldhaber et al. used split-population models to estimate the unique impact of 

measured covariates on both the timing and probability of entry into the profession. Regarding 

characteristics of teacher candidates, younger candidates were more likely to be observed in a 

public-school teaching job (10-year increase associated with a 2.5 percentage point decrease in 

the probability of hire after 1 year, 4.0 percentage point decrease of hiring after 5 years), teacher 

candidates of color were less likely to be observed in a public-school teaching job (72.5% 

probability for White candidates, 70% probability for candidates of color, although differences 

were not statistically significant), and candidates endorsed in STEM fields and special education 

were also more likely to be hired compared to elementary teachers and secondary teachers in 

other fields. Characteristics of cooperating teachers were generally not predictive of entry into 

the profession, but characteristics of the schools in which candidates completed their student 

teaching were. Completing student teaching in a suburban school or a school with higher 

turnover rates (particularly for those candidates with higher academic credentials) led to an 

increased probability of hiring. In fact, 15% of all observed hires were in the school where 



32 
 

student teachers completed their internship, particularly in cases of high academic credentials 

and high turnover rates. 

In a later analysis of the same data, Goldhaber et al. (2017b) found that student teaching 

placements may impact not only entry into the workforce but also teacher effectiveness upon 

entry. Specifically, they found that teachers who had completed their student teaching in a school 

with similar characteristics to the school of their first teaching placement were, on average, more 

effective (as measured by Valued-added models) than teachers who had completed their student 

teaching internship in a school with vastly different characteristics. Although one explanation for 

this result is that being prepared in a similar context provides an easier pathway to transfer the 

skills developed in their internship into their teaching, the endogenous nature of student teaching 

assignments and entry into the teacher workforce requires an abundance of caution. For example, 

it is also possible that teachers who perceived themselves as ineffective during their student 

teaching actively sought different contexts upon workforce entry. Similarly, those who perceived 

themselves as highly effective actively sought similar contexts. In this case, measures of teacher 

effectiveness would be potentially independent of the degree to which school placements 

matched. However, if the advantage of teaching in a school with matched characteristics holds, 

there would be “important ramifications in terms of equity” (Goldhaber et al., 2017b, p. 353) 

within student teaching placements. The authors found that teachers were much more likely to 

complete their internships in schools with lower proportions of FRL and underrepresented 

minority students, necessitating that schools with higher proportions of such students would be 

less able to hire teachers with internship experiences in such schools. 

2.1.05 Issues of Equity - Barriers to Entry 



33 
 

Diversity and representation in the teacher workforce should not just be lofty ideals, but 

issues of deep reflection, critical thought, and a call to action. When students of color comprise 

about 52% of the student body, but teachers of color comprise only 20% of the workforce 

(MacFarland et al., 2018), those within the educational system must work to rectify current 

practices and policies that have disenfranchised others. Indeed, there are multiple “structural, 

institutional, and environmental mechanisms that uphold colonial, white supremacist ideals and 

work against efforts to recruit and retain [teachers of color]” (Carter Andrews et al., 2019, p. 7). 

One final barrier to entry for those pursuing careers in teaching prior to workforce entry is a 

licensure performance exam, such as the educative Teacher Performance Assessment (edTPA), 

which has been adopted in at least some capacity in 41 states and the District of Columbia (see 

https://www.edtpa.org/policy_and_accreditation). A growing body of evidence suggests that 

these performance exams may disproportionately impact prospective teachers of color, serving as 

an exclusionary gatekeeping barrier for entry into the profession (Carter Andrews et al., 2019).  

The edTPA portfolio consists of video-recorded classroom teaching, artifacts of planning, 

instruction, and assessment, and reflective writing, often completed during the student teaching 

internship. The student teacher then submits the portfolio for evaluation. In an examination of the 

edTPA technical documentation, Gitomer et al. (2021) assessed the extent to which the 

assessment, rubric, and scoring practices/policies met a robust threshold of validity and 

reliability necessary for its high-stakes function. The authors found that despite consisting of 27 

unique subject-area assessments, the technical documentation only reported aggregate 

information regarding validation tests. Further, much of the validation work took place in 

controlled lab settings “that differ substantially from operational conditions that currently exist in 

the field” (Gitomer et al., 2021, p. 10). This runs counter to the tenet of validity “that evidence 



34 
 

does not automatically generalize beyond the kinds of contexts and conditions within which the 

validating evidence was collected” (p. 10). Summarily, the psychometric validity of clinical lab 

results does not translate to real educational settings, and the validity of one subject domain does 

not translate to validity amongst all 27 unique assessments. Gitomer and colleagues reported 

issues of reliability as well: most portfolios are assessed by only a single rater, and the 

documentation subsequently reports misuses of kappa and alpha indices of reliability because 

reliability calculations did not consider the correlation of measures from a single rater. There was 

also a large degree of variation between subject areas, raising issues of uniform cut-scores for 

passing. Perhaps the most pressing issue of equity, however, comes from scoring disparities 

between demographic groups: on average, Black/African American candidates scored lower and 

closer to the pass/fail cut-score threshold compared to White candidates, with the difference 

amounting to .39 standard deviations. Gitomer and colleagues concluded that the use of edTPA 

is ill-suited for high-stakes, pass/fail demarcations for assessments of teaching readiness due to a 

host of technical and empirical issues. 

Issues of equity in the edTPA transcend its development and documentation: Goldhaber 

et al. (2017a) analyzed the extent to which edTPA scores were predictive of employment and 

teacher effectiveness (as measured by value-added measures) using data from Washington 

State’s Professional Educator Standards Board and administrative data from the Office of the 

Superintendent of Public Instruction. The primary analytic sample consisted of candidates who 

took the edTPA in the 2013-2014 school year (N = 2,362). When using edTPA as a standardized 

continuous measure, a one standard deviation increase in edTPA score was associated with a 5.9 

percentage point increase in the probability of employment. Results for employment probability 

were likely strong because passing the edTPA was consequential for licensure. That is, failing to 



35 
 

receive a high enough score necessarily precluded a candidate from obtaining a teaching license. 

Strikingly, however, was the difference in pass rates between Hispanic and non-Hispanic White 

candidates: while only 3.7% of non-Hispanic candidates failed after January 2014, 13.7% of 

Hispanic candidates failed to pass. Further, when used as a pass/fail screener, edTPA was not 

predictive of student math outcomes, but it was predictive of student reading outcomes. One 

potential explanation is that as an assessment, edTPA scores are largely predicated on a 

candidate’s ability to write a clear narrative for the evaluator. When considered alongside issues 

of validity and reliability (Gitomer et al., 2021), edTPA outcomes are potentially an indicator of 

one’s reading/writing ability in academic English more so than a measure of teacher readiness, 

raising further doubt about its equitable use as a screening measure for teacher readiness. 

2.1.06 Summary of education literature 

A plethora of empirical work has demonstrated that entry into the teaching profession is 

impacted by personal, geographical, and institutional factors. The demographic composition of 

the teacher workforce has diversified slightly, but it is not yet representative of the student 

population (MacFarland et al., 2018; Redding & Nguyen, 2020). Transitions from postsecondary 

institutions into the teaching profession are greatly impacted by the student teaching experience, 

both in terms of geographic location and characteristics of their host school (Goldhaber et al., 

2014 Krieg et al., 2016; Shirrell & Reininger, 2017). Furthermore, individuals from marginalized 

populations also face potentially inequitable gatekeeping in the form of licensure examinations 

(Carter Andrews et al., 2019; Gitomer et al., 2021). 

2.1.10 Music Education 

While several similarities between educators broadly and music teachers specifically may 

exist, the music teacher pathway is distinct in the manner in which one enters the profession. In 



36 
 

this section, I examine the literature on recent licensure practices in the United States and the 

characteristics of music teacher licensure candidates. Then, I discuss research on issues of equity 

to contextualize some of the reasons why the population of music teacher candidates differs from 

the general teaching population. Finally, I review the literature on music education students’ 

motivations and aspirations for teaching music. Altogether, this literature provides a contextual 

framework to understand how the pathway for music teachers may be unique and warrant further 

empirical research. 

2.1.11 Motivation, Preparation, and Aspiration for Music Teaching 

Although the pathway to becoming a music teacher begins long before entry into a music 

education program, the first formal entry point is the audition into the music/music education 

program. Rickels et al. (2013) surveyed prospective music education majors about their career 

influences (N = 328). Participants completed a survey on their audition day at one of eight 

participating institutions varying in geographic location, size, and structure. Survey items 

consisted of six-point Likert-type responses ranging from zero (not at all influential) to five 

(extremely influential) regarding experiences and individuals that influenced their interest in 

music education programs. The highest rated experiences were performances with a school 

music group (M = 4.42, SD = .78), membership in honor performance ensembles (M = 4.37, SD 

= 1.01), and leadership positions in their school music program (M = 4.35, SD = .85). The 

highest rated individuals were their high school music teacher (M = 4.5, SD = 1.08) and private 

lessons instructor (M = 3.97, SD = 1.66). Clearly, participants perceived their high school music 

experiences as highly influential. 

One of the central goals for music education curricular and fieldwork experiences is to 

prepare preservice music teachers for their potential entry into the teaching profession. Schmidt 



37 
 

(2013) reviewed the literature regarding the relationship between such experiences and PMTs’ 

beliefs on how it would impact their entry into the profession. Importantly, many of PMTs’ 

beliefs about teaching and the type of teacher they want to be are grounded in their years of 

experience as a student. However, their beliefs may also be shaped by preservice preparation. 

Still, disparities between undergraduate coursework, observations, and student teaching 

experiences in the teaching methods, beliefs, and theories of teaching may be sources of 

frustration for these aspiring educators. This vexation is especially salient when considering that 

many undergraduate students were motivated to pursue careers in music because of their 

experiences with school music programs, particularly at the high school level (Rickels et al., 

2013). 

Bartolome (2017) followed a cohort of music education students from their sophomore 

year through their entry into the profession immediately af