Sage Journals: Discover world-class research

Abstract

We use longitudinal data across public postsecondary institutions in Tennessee to examine the stages at which potential Black educators disproportionately exit the teacher pipeline. Black and White bachelor’s graduates declare and complete teaching-related majors at similar rates, suggesting comparable levels of initial interest in teaching careers; moreover, newly certified Black college graduates are more likely than their White peers to gain employment in the state. However, in between those points, Black undergraduate program completers are substantially less likely to become licensed. Additionally, Black graduate students are less likely to complete master’s degrees in teaching-related majors and receive teaching licenses, though newly certified Black master’s graduates are again more likely to gain employment than their White peers. These results suggest that the largest loss of potential Black teachers occurs at the end of teacher preparation, which can be explained partially but not fully by inequities in certification exam passage rates.

Keywords

diversity equity teacher education/development Black education descriptive analysis longitudinal studies

Enrollments in teacher certification programs have steadily declined across the United States over the past decade, a particularly troubling trend given that the overall demand for teachers is expected to increase in response to retirements and attrition (Sutcher et al., 2016). As a result, teacher preparation programs and education leaders are considering new and different ways to recruit and train preservice teachers (PSTs), including, for example, increasing attention to Grow-Your-Own, residency, and other nontraditional programs (Gist, 2019; Guha et al., 2016). To best direct these efforts at bolstering the teacher workforce, it is critical to understand the present pipeline into teaching and especially the stages at which prospective teachers may drop out.¹ Furthermore, with growing evidence of the academic and nonacademic benefits of having teachers of color (ToCs) for all students, and especially for students of color (Blazar, 2021; Cherng & Halpin, 2016; Dee, 2005; Gershenson et al., 2022), it is even more critical to understand where and how we are losing potential ToCs.

There is extensive evidence that ToCs uniquely experience and progress through different stages of the pipeline into teaching, often resulting in systemic racial and ethnic disparities rooted in the long history of racism and oppression in the United States (see Frank, 2019, for a review of this literature). Carter Andrews et al. (2019) have thus called for deeper investigation into the “challenges associated with racially and ethnically diversifying the teaching workforce” (p. 6), with a specific attention to the push-out and keep-out effects that particular milestones along the pipeline may have on potential ToCs. For example, some have argued that teacher certification requirements screen out potential ToCs interested in teaching by gatekeeping licensure behind tests that have little to do with actual teaching practice (Petchauer, 2012). At the same time, overall perceptions of and preferences toward the teaching profession may—possibly because of these inequities—differ for ToCs in ways that result in their collective choice to pursue other career fields more often (Christian et al., 2024). Up to this point, mostly qualitative research has examined the unique, and often inequitable, experiences of potential ToCs who either continue or exit the teaching pipeline at a given stage. However, far less work has tried to understand the process as a whole. While each narrative of how ToCs encounter and often overcome these obstacles is important and can become lost in aggregate analyses (Frank, 2019), a statewide analysis of differences by race in how potential teachers navigate all of the different stages in the teacher pipeline promises to highlight where exactly systemic inequities may be greatest and where efforts at their reduction might be best directed.

This study, therefore, conceptualizes the teacher pipeline as a set of discrete stages and examines how potential teachers move through each transition. It is, to our knowledge, among the first statewide analyses of racial disparities across multiple stages of two distinct but connected pathways into teaching: (a) a bachelor’s pathway, which we operationalize as beginning by indicating interest in a teaching career as undergraduates (through major declaration), continuing through certification, and finally ending with employment as public school teachers; and (b) a master’s pathway for students who may choose to become teachers later in their lives. Note that we are focused on pathways into teaching, and as such, we end our pipeline with initial employment as a teacher; racial inequities in where teachers work and for how long, though equally important and worth interrogating, are beyond the scope of this article. By providing a bird’s eye view of the teacher pipeline, we offer insights for policymakers interested in addressing inequities in the recruitment, preparation, and employment of new ToCs and for researchers who seek to identify new fruitful avenues for future study (including Black teachers’ specific school settings as well as their patterns of subsequent mobility and attrition from the teacher workforce). In particular, we identify the specific stages in the pipeline that demonstrate the greatest racial gaps in order to better target future interventions and research for addressing these disparities.

Literature Review

Despite recent calls for its diversification (Grissom et al., 2015; Hughes et al., 2020) grounded in emerging evidence that doing so would lead to improved student outcomes (Blazar, 2021; Cherng & Halpin, 2016; Redding, 2019), the teaching workforce is still far less racially and ethnically diverse than the student population. We organize our review of the literature around what we currently know about the three main stages of the teacher pipeline examined in this study: (a) declaring an education major, (b) receiving a teacher license, and (c) gaining employment as a teacher. We review literature on current trends for all (prospective) teachers as well as on potential disparities in stage completion between (prospective) White teachers and ToCs, focusing on Black teachers whenever possible.

Majoring in Education

For most prospective teachers, the first step in the pipeline entails entering a teacher preparation program, which, at least at the undergraduate level, usually involves declaring an education major. Even at this initial juncture, the prospective teaching workforce is disproportionately White across multiple points of comparison. The U.S. Department of Education (USDOE, 2016) estimated that 27% of all education majors in 2012 identified as potential ToCs (12% as Black), compared to 38% (14% Black) of postsecondary students in all other majors and 49% (16% Black) of all K–12 students nationwide. Similarly, among those who graduated, only 5% of all bachelor’s degrees conferred to Black students in 2012 were in education, less than the 8% of White students.

In one of the only studies focusing on college applicants’ interest in education, Bartanen and Kwok (2022) reported that Black applicants at one institution were significantly less likely than their White peers to express an interest in education, a gap that widened both after programs completed their admissions process and after admitted applicants enrolled. This study suggests, then, that disparities in the pipeline of prospective teachers exist even before entering a teacher preparation program, often reflecting differences in individuals’ preferences, but that the admissions and recruitment practices of schools of education within colleges and universities may potentially exacerbate them further, implying that schools of education can do more to attract and recruit racially and ethnically diverse postsecondary students.

Research has found similar racial and ethnic differences in preferences for education when looking at actual teacher preparation program (TPP) enrollment instead of major choice. This is an important distinction, as not all education majors lead to teacher certification (e.g., career counselors or school leaders still receive an education major), and many eventual teachers received undergraduate majors outside of education. In their review of TPP admission statistics with a specific focus on candidate diversity, Drake et al. (2021) reported that only 7% of all TPPs in the United States enrolled a student population at least as diverse as the overall student population at their institutions; if the demographics of admitted TPP cohorts reflected those of their institution’s overall student populations, they estimate that roughly 80,000 more ToCs would be prepared nationwide.

Teacher Licensure

The next step in the typical teacher pipeline involves acquiring a license. This step is a virtually universal requirement for becoming a public school teacher in the United States. While accessing accurate nationwide statistics on the racial composition of licensed TPP completers is difficult, as Title II of the Higher Education Act does not require reporting of the racial and ethnic composition of graduates (Carver-Thomas, 2018), some individual states like Tennessee do provide details on the diversity of TPP completers as part of their teacher preparation report card (State Collaborative on Reforming Education, 2016). The most recent report cards show that, on average, 16% of TPP completers identify as ToCs, with program percentages varying from 0% to 35% across the state (Tennessee State Board of Education, 2022). As 37% of the K–12 population in Tennessee were students of color in 2017–2018 (Tennessee Department of Education, 2018), this suggests that only the most diverse programs in the state are certifying prospective teachers who reflect the racial and ethnic diversity of K–12 students in the state.

Literature on teacher certification exam passing rates, arguably a good proxy for licensure, has shown similar patterns to TPP enrollment and completion along racial and ethnic lines. These exams usually test for literacy, computational, and other skills assumed to be necessary for teaching and, more recently, for pedagogical practices through a variety of assessment procedures, including multiple choice tests, essays, and performance portfolios. Over the years, multiple studies have documented the deterrent effects of teacher licensure requirements on the progression of preservice ToCs in the teacher pipeline (Bennett et al., 2006; Goldhaber & Hansen, 2010; Greenberg Motamedi et al., 2021). In a historical analysis of Praxis passing rates, Fenwick (2021) found that Black and Latino test takers were about twice as likely as White test takers to fail a Praxis test, requiring these ToCs to retake exams, delaying their teaching licensure, and resulting in additional financial burdens. Interestingly, similar racial disparities in pass rates are present in certification exams for other professions (Klein & Bolus, 1997; Yeo et al., 2020), despite only mixed evidence of their predictive capacity for professional success (Anderson, 2020; Gitomer et al., 2015; Goldhaber & Hansen, 2010). A more recent analysis of the initial basics skills certification exam passage rates in Washington found significantly lower passage rates among Hispanic and non-Hispanic candidates of color (approximately 70%) than among White candidates (over 80%) (Greenberg Motamedi et al., 2021). Among the candidates who took at least one of the three Washington certification tests, fewer Hispanic (61%) and non-Hispanic (63%) candidates of color became certified when compared to White (66%) candidates. Together, these findings suggest that teacher licensure requirements might push out potential ToCs by serving as additional barriers to certification.

Employment as a Teacher

Once licensed, the final step is to gain employment. Only a few studies have reported on the difference in career choices and employment rates between White TPP completers and TPP completers of color, finding mixed evidence that may depend on how these identities are operationalized as well as on labor market context. For example, Bardelli and Ronfeldt (2021) reported that recent Black TPP completers in Tennessee were about 1.6 times as likely as their White peers to be employed, but TPP completers who identified as Asian/Pacific Islanders and Hispanic were less likely (by 0.5 and 0.8 times, respectively) to be employed as their White peers. In Washington State, Goldhaber et al. (2014) reported on the employment outcomes of TPP completers between 1998 and 2010 and noted that completers of color were collectively almost half as likely to be hired as public school teachers than their White peers in the same cohorts. A more recent analysis of the teacher pipeline in Washington, though, found no meaningful ethnic or racial differences among newly certified teachers in terms of hiring rates (Greenberg Motamedi et al., 2021).

Contributions

This article makes several contributions to the emerging literature on the pipeline into teaching. While each of the studies reviewed so far explored a single step in the teacher pipeline in depth, none followed the same cohorts of college students through bachelor’s degree completion, teacher certification, and workforce outcomes. In this study, we observe the same group of students throughout the entirety of the teacher pipeline, allowing us to observe how they move through each stage and where they exit.

Prior studies also have used majoring in education as the primary signal for interest in teaching, which likely misses the mark because many who declare education majors never intend to teach, while some who declare noneducation majors (e.g., math, English) actually do (Cowan et al., 2016). A contribution of the present study is to identify and especially focus on those education majors that are most likely to predict becoming a teacher (which we refer to as teaching majors) to investigate whether similar racial disparities exist among them. This focuses our analysis on the group of college graduates that is most likely to pursue a teaching career. At the same time, we also separately explore the graduate pipeline into teaching for those completing noneducation undergraduate majors to more comprehensively understand if and how this alternative pathway contributes to observed racial disparities in the teacher workforce.

Furthermore, little is known about which career paths are chosen by prospective ToCs who do not get hired as public school teachers. We therefore make a novel contribution to the literature by following these graduates into the labor market and observing the careers in which they end up employed in place of teaching.

Finally, collapsing all non-White college students into a single category (e.g., “students of color”), as is common in many studies, risks masking the racialized experiences specific to individual identities, leading to results that are difficult to interpret and, likely, less interesting. As Black students are the largest non-White sample of both K–12 and college students in our dataset, we focus on the Black teacher pipeline in this article.² Additionally, given extensive evidence that Black students have better academic and nonacademic outcomes when taught by Black teachers (see Redding, 2019, for a review of the evidence), it is critical to understand where we might be losing potential Black educators to inform how might increase their representation.

More formally, we ask the following research questions: (a) What percentage of Black bachelor’s completers progress along various stages of the teaching pipeline—declaring teaching-related majors, becoming licensed, and gaining employment as a teacher? (b) How do these rates of progression compare to those of White students at the same stages? (c) How do the patterns we observe among Black and White undergraduates compare to the patterns we observe among master’s graduates?

Methods

Data

Our study employs three separate sources of data from across multiple of Tennessee’s statewide longitudinal data systems.

First, we use college enrollment and graduation data from the Tennessee Higher Education Commission. This dataset includes information at the term level for all students enrolled in a public institution (i.e., community colleges and state universities) and a select group of private institutions. It contains data on declared majors, overall credits attempted and completed, and GPA. Although the omission of some private institutions may exclude a small proportion of the state’s prospective teachers, this dataset comprises the majority of all postsecondary students statewide.

Importantly, we restrict our sample to only those undergraduate and graduate students who successfully received a degree. We primarily do this in order to ensure consistent coverage across datasets; specifically, we keep only those postsecondary students who completed their degrees and would therefore have become eligible for certification and hiring during the years in which we have licensing and employment records (2002 through 2016). For those without graduation year information, it is difficult, given variation in the length of programs across the state, to determine the years in which these students would have sought licensure or employment had they obtained their degrees and would require assumptions about equivalent times to program completion for Black and White students. Still, as a robustness check, we make efforts to explore initial interest in a teaching career among all undergraduate students regardless of eventual degree attainment and find that this does not appear to be a stage where Black prospective teachers disproportionately exit the pipeline.³

Onto this dataset, we merge teacher licensing and employment records from the Tennessee Education Research Alliance. These data provide licensing information for all teachers in Tennessee, including the type of license received and all associated endorsements. Teacher employment records include the county and school buildings in which teachers work and their teaching assignment categories. We are able to link 12.7% of college graduates to teacher licenses and 9.6% to teacher employment data as public school teachers, indicating the proportions of the overall college graduate population that became certified and found a Tennessee public school teaching job respectively.

Finally, we merge data from the Tennessee Unemployment Insurance (UI) program onto our dataset. These data include quarter-level employment information for employees of firms that participate in the state’s UI program. From these data, we are able to identify the employment sector description for the prospective teachers in our sample who do not end up employed as public school teachers in the state. Our match rate ranges between 75% and 80% depending on the subsample of graduates we consider.

It is worth noting that progression to each successive stage of the teacher pipeline is functionally operationalized by the existence of a match in each of these datasets. For example, a prospective teacher is indicated as getting hired if they match to the employment dataset and as failing to find a job if they do not. Consequently, we are not able to distinguish among licensed graduates who sought employment and failed, those who chose at this juncture not to pursue teaching as a career, and those who found employment (either as a teacher or in some other sector) outside the State of Tennessee.

We acknowledge that this may inherently overstate the true rates at which both Black and White prospective teachers exit each stage of the broader teaching profession (e.g., by omitting from our estimate of employment rates any licensed graduates who were hired as teachers in schools outside of Tennessee or in private schools). In addition, any relative differences in preferences for non-Tennessee public school jobs among Black and White graduates may influence and inform the racial disparities we observe in these rates; we explore this possibility through exploratory analyses (see end of Results) differentiating among in-state and out-of-state students. However, if we more narrowly conceptualize the teacher pipeline as a path towards a job in Tennessee’s public school system, this is less of a data limitation and more a contribution toward overall attrition, in which the state successfully trains these individuals to become teachers but is unable to find them employment within its public schools.

Categorizing Majors

As a preliminary step, we first categorize postsecondary majors into groups based on the likelihood that a graduate with a given major becomes employed as a teacher in the state. This set of decisions allows us to reduce the hundreds of individual majors that are awarded down to a smaller set of categories.

We employ a combination of empirical methods and subjective classification to form these categories. More technically, we first use a cross-validated least absolute shrinkage and selection operator (LASSO) linear regression to identify all majors not predictive of a teaching career. A LASSO regression is a statistical learning model that selects the subset of variables (i.e., college majors) that maximize out-of-sample predictive power for an outcome of interest (i.e., entry into teaching), excluding all variables with little to no predictive power. Appendix Table A1 Column 2 reports the LASSO adjusted coefficients from this regression—which can be interpreted as the adjusted fraction of graduates in each major that become employed as a teacher in the state—for four-digit Classification of Instructional Programs (CIP) major codes.

Intuitively, the LASSO helps us set aside the many majors not predictive of a teaching career and allows us to focus our hand classification only on those majors in which a reasonable proportion of graduates become employed in public schools. In the next step, we use the results of the LASSO regression to construct a set of decision rules for the deterministic classification of majors into four groups. We create two groups within the set of majors with coefficients shrunk to zero (i.e., are unrelated to teaching)— nonteaching education and general majors —and two within those that have positive coefficients (i.e., predict a teaching career)—teaching majors and high school specialist majors.

We define nonteaching education majors as all majors with CIP codes that begin with 13 but do not appear to function as direct pathways into teaching (i.e., have LASSO regression coefficients that are shrunk to zero). These include degrees for education counseling, curriculum and instruction, or school administration, all of which are significantly more common at the graduate level and among students who have previously found employment as teachers.

Additionally, general majors include all majors with CIP codes awarded in the state that do not begin with 13 and are also not likely to result in a teaching career. These include, for example, majors in engineering, telecommunications, or computer science.

On the other hand, we define teaching majors as those most empirically (i.e., with the largest LASSO regression coefficients) and theoretically likely to result in a teaching career. To be classified as a teaching major, we require that the LASSO coefficient associated with a given major exceeds in magnitude that of Teacher Education – Specific Subject Areas, the major with a CIP code beginning with 13 least predictive of a career in teaching without a coefficient shrunk to zero (and therefore classified as a nonteaching education major.) As a result, most of these majors are categorized under education in the CIP system (e.g., teacher education majors in special education and teaching); notably, however, this classification also includes majors in multidisciplinary studies.

Finally, high school specialist majors include those majors with CIP codes that do not begin with a 13 and are less likely to result in a teaching career than the teaching majors but for which we still regularly observe graduates being employed as teachers in the state (i.e., each of these majors has a nonzero LASSO coefficient). These majors include subjects that are usually taught in high school and that require a subject-matter specialization, like English language arts, mathematics, social studies, and science.

Although we estimate these models on Black and White teachers together, we explore whether we obtain different coefficients when running our LASSO regressions on separate subsamples by race (see Appendix Table A1). At a coarse level, we see qualitatively similar groupings of majors among both Black and White students. However, a closer look reveals some interesting patterns. First, most coefficients are smaller for Black college students, suggesting that Black college students are less likely to progress through the entire pipeline to become teachers than their White peers regardless of their major of choice. Interestingly, this appears to be more the case among teaching majors in specific grade levels or subject areas than among high school specialist majors, where the coefficients across subsamples tend to be more comparable. However, this is not the case among graduates majoring in special education, where a higher proportion of Black than White students become teachers.

Samples

We begin with a sample of 287,510 Black and White graduates of Tennessee higher education institutions between 2002 and 2016 for whom we have any information on major declaration (see the solid light gray box in Figure 1). Across all 15 years, 17.5% of students in our sample identified as Black, while 82.5% identified as White (see Table 1). Our full sample consists of predominantly in-state students (85.9%) and a majority of female students (58.3%).

Figure 1.

Teacher pipeline(s) conceptual model.

Table 1

Sample Descriptive Statistics by Individual and Institutional Characteristics

	Full sample	BA pipeline sample	MA pipeline sample
Panel A—Individual characteristics
Race
Black	50,395 (17.5%)	5,792 (17.5%)	6,020 (16.1%)
White	237,115 (82.5%)	27,235 (82.5%)	31,396 (84.0%)
Gender
Female	167,590 (58.3%)	26,724 (80.9%)	24,036 (64.2%)
Male	119,920 (41.7%)	6,303 (19.1%)	13,380 (35.8%)
Residency status
In-state	246,960 (85.9%)	30,130 (91.2%)	34,588 (92.4%)
Out-of-state	39,793 (13.8%)	2,866 (8.7%)	2,701 (7.2%)
Foreign	649 (0.2%)	18 (<0.1%)	124 (0.3%)
Unknown	108 (<0.1%)	13 (<0.1%)	3 (<0.1%)
N	287,510	33,027	37,416
Panel B—Institutional characteristics (BA)
Public	238,064 (82.8%)	26,725 (80.9%)	31,519 (84.2%)
Private	49,446 (17.2%)	6,302 (19.1%)	5,897 (15.8%)
HBCU	17,216 (6.0%)	1,086 (3.3%)	2,107 (5.6%)
Non-HBCU	270,294 (94.0%)	31,941 (96.7%)	35,309 (94.4%)
Institution system
Tennessee Board of Regents	163,306 (56.8%)	21,399 (64.8%)	21,493 (57.4%)
University of Tennessee system	74,758 (26.0%)	5,326 (16.1%)	10,026 (26.8%)
Independent colleges and universities	49,446 (17.2%)	6,302 (19.1%)	5,897 (15.8%)
N	287,510	33,027	37,416
Panel C—Institutional characteristics (MA)
Public	34,786 (79.7%)	4,471 (71.1%)	29,870 (79.8%)
Private	8,852 (20.3%)	1,821 (28.9%)	7,546 (20.2%)
HBCU	2,784 (6.4%)	198 (3.2%)	2,323 (6.2%)
Non-HBCU	40,854 (93.6%)	6,094 (96.8%)	35,093 (93.8%)
Institution system
Tennessee Board of Regents	23,202 (53.2%)	3,707 (58.9%)	20,041 (53.6%)
University of Tennessee system	11,584 (26.6%)	764 (12.1%)	9,829 (26.3%)
Independent colleges and universities	8,852 (20.3%)	1,821 (29.0%)	7,546 (20.2%)
N	43,638	6,292	37,416

As prior work has shown differences between Black and White college applicants in their interest in a teaching career (Bartanen & Kwok, 2022), we first explore the extent to which these graduates showed interest in a teaching career by declaring a teaching major anytime during an undergraduate degree. While declaring a teaching major is not a necessary and sufficient condition for interest in teaching, we believe that it still identifies a group of college students who were motivated to pursue teaching as a career seriously enough to major accordingly.

In the second stage of our analysis, we focus on the subsample of bachelor’s degree holders who completed a teaching major (see the light dotted box in Figure 1 below). Among our full sample of 287,510 graduates, a total of 33,027 (11.5%; N_b = 5,792, N_w = 27,235) students earned a bachelor’s degree in a teaching major. This BA sample includes predominantly female (80.9%) and in-state (91.2%) students. By examining this BA sample, we can follow the group of college graduates who maintained their interest in a teaching career through the steps required to become a certified teacher in a public school.

This choice, however, also leaves out potential teachers who do not follow a traditional undergraduate teaching preparation pathway, such as teachers who receive their bachelor’s degree in a specialized field and then become teachers through a master’s degree program or an alternative route to teaching. Therefore, in subsequent analyses, we also explore a third sample of students that includes college graduates with degrees in any undergraduate majors who enroll in a master’s degree, excluding those who had already received a teaching license prior to earning their master’s degree (see the dark dotted box in Figure 1 below). This includes 37,416 students (13% of our full sample of 287,510 students; N_b = 6,020, N_w = 31.396) who completed a master’s program in any major during our observation period. In terms of their master’s program majors, 12.8% are teaching majors, 13.9% are high school specialist majors, 18.6% are nonteaching education majors, and the rest (54.7%) are general majors unlikely to lead to teaching.

Across these three samples, roughly 80% of bachelor’s and master’s completers attended public institutions, while approximately 6% attended HBCUs (Historically Black Colleges and Universities). Over half of all students (56.8%) graduated from colleges and universities that are part of the Board of Regents (TBR) network, followed by about one-quarter (26%) from University of Tennessee-system schools and 17.2% from independent colleges and universities. In the undergraduate pipeline sample in particular, we observe a lower proportion of students enrolled in University of Tennessee schools.

Modeling the Teacher Pipeline

Our data give us the opportunity to model the teacher pipeline as a set of sequential milestones, starting from students’ very first postsecondary experiences to their eventual employment as teachers in the state. The steps in the pipeline that we can observe include declaring a major, receiving a postsecondary degree, earning a license, and finding employment as a teacher. We then divide students’ postsecondary experiences into bachelor’s and master’s programs to further model the sequential relationship between degrees.

We present our conceptual model of the pipeline to teaching in Figure 1 below, which tracks students from enrollment in an undergraduate institution through entrance into the teaching workforce. At each juncture, students may either progress towards becoming a teacher (along a white arrow) or exit from the pipeline entirely (as indicated by a striped arrow). This figure also describes specific samples used in our analyses, through gradient shading and, when appropriate, by noting analytic sample sizes. Additional values of N for both Black and White teachers at each of these stages of the pipeline can be found in subsequent results. Also note that the boxes of this model are not drawn to scale; instead, our Sankey diagrams (Figures 2 –5) present visually proportional indications of Black and White teachers’ progression through the undergraduate and graduate pipelines.

Figure 2.

Teaching major pipeline for Black undergraduates.

Figure 3.

Teaching major pipeline for White undergraduates.

Figure 4.

Master’s degree teacher pipeline for Black graduates.

Figure 5.

Master’s degree teacher pipeline for White graduates.

As another preliminary step, we test whether modeling these pipelines to teaching separately for Black and White students is appropriate for our data and sample using a structural equations model (SEM). We fit two nested models: one that constrains the path coefficients between the two groups to be the same and one that leaves the path coefficients unconstrained. We use a Wald likelihood ratio test to compare the fit of these two models. The results of this test suggest that the teacher pipelines are significantly different for Black and White college graduates overall, (χ²[73] = 2,767.44, p < .001). Therefore, for the rest of the article, we focus on both describing the teacher pipeline and examining persistence at each stage separately for Black and White graduates as a way to better understand where disparities exist.

Statistical Inference

Throughout the Results section, we report Pearson’s chi-squared test of independence values and their associated p-values. These statistics generally test for whether two categorical variables are independent from each other; here, they explore whether observed differences between Black and White graduates’ teacher pipeline outcomes are statistically significant. For each test, the null hypothesis states that there is no difference between the two groups on a given outcome; a significant p-value rejects this null hypothesis.

Results

Preference Towards Teaching

We consider the declaration of a teaching major for at least one semester the first step into the teaching pipeline. For our full sample (see the solid light gray box in Figure 1), we report the breakdown of undergraduate majors that college completers declare in Table 2 separately by race. We find that just shy of 18% of all college graduates (N_b = 8,939, N_w = 42,528) for whom we have term major data declared a teaching major for at least one semester during our observation period, regardless of whether they identify as Black or White (joint row χ²[3] = 0.9, p = .825). Moreover, we find no differences in the likelihood of sustained interest in a teaching career between Black and White college graduates, as 11.5% of students belonging to each group (N_b = 5,792, N_w = 27,235) receive a bachelor’s degree in a teaching major during our observation period (joint row χ²[4] = 0.0, p > .999).

Table 2

Undergraduate Preferences Towards a Teaching Major

	Black		White
Panel A: Ever declared major
General	20,046	39.8%	105,005	44.3%
High school specialist	20,215	40.1%	86,330	36.4%
Nonteaching education	1,195	2.4%	3,252	1.4%
Teaching	8,939	17.7%	42,528	17.9%
Total	50,395	100.0%	237,115	100.0%
Panel B: Major received
No information	3,463	6.9%	21,532	9.1%
General	21,958	43.6%	105,788	44.6%
High school specialist	19,181	38.1%	82,520	34.8%
Nonteaching education	1	0.0%	40	0.0%
Teaching	5,792	11.5%	27,235	11.5%
Total	50,395	100.0%	237,115	100.0%

Outcomes for College Graduates With a Teaching Major

Since Black and White college graduates in Tennessee appear to have shown similar preferences towards a teaching major, differences in their likelihoods of becoming a teacher likely originate later in the pipeline. Consequently, we explore differences in the teacher licensing program completion rates of Black and White teaching major graduates (our second sample; see the light dotted box in Figure 1). We then follow licensed teachers and observe if they become employed as teachers in Tennessee public schools.

Figure 2 displays the teacher pipeline for Black college students that receive a bachelor’s degree in a teaching major (far left blue bar). We find that only 19.6% of the students in this group immediately receive a teaching license in Tennessee (N = 1,137) and 12.1% of them first pursue a master’s degree (N = 700). Of the licensed teachers in this group, we observe 84.0% of them becoming employed in a public school in the state as a teacher of record (N = 1,285).

Figure 3 displays this teacher pipeline for White college students who receive a bachelor’s degree in a teaching major. We find that White teaching major graduates have more than double the likelihood of becoming licensed teachers in the state (50.7%, N = 13,819) and roughly double the likelihood of pursuing a master’s degree (20.5%, N = 5,592) when compared to Black teaching major graduates, while the rest leave the teacher pipeline likely to pursue employment out of state, in a private school, or in a different sector altogether (28.7%, N = 7,824). Of the licensed teachers in this group, we observe 78.2% of them being employed in a public school in the state as a teacher of record (N = 14,507).

This first set of results shows a stark difference by race in the likelihood of leaving the teacher pipeline after college graduation. Black college students who receive a bachelor’s degree in a teaching major (19.6%) are 31.1 percentage points (or 61.4%) less likely than their White peers (50.7%) to receive a teaching license in the state (joint row χ²[2] = 1,020.7, p < .001). On the other hand, we observe that Black college students (84.0%) are 5.8 percentage points (or 6.9%) more likely than their White peers (78.2%) to be employed as a teacher, conditional on becoming licensed in the state (joint row χ²[1] = 6.0, p < .05).

Two main takeaways emerge from these results that together highlight the period between graduation and before seeking employment as the stage most marked by differential progression rates among Black and White prospective teachers. First, we find a clear and sizable difference in the Black teacher pipeline after college graduation, as we observe significant disparities in the likelihoods of becoming licensed and of pursuing a master’s degree between Black and White college graduates with a teaching degree. Second, we find that Black college graduates are actually more likely to become employed as teachers, conditional on receiving their teaching licenses.

Employment for Graduates Who Do Not Become Teachers

We follow graduates with bachelors’ degrees in teaching majors who do not become Tennessee public school teachers—regardless of licensure and master’s completion—into the workforce in order to understand where they find employment instead (see the solid black box at the top of Figure 1). We are able to match roughly 75% of these graduates to employment insurance data from the Tennessee Department of Labor. The graduates who remain unmatched either moved out of Tennessee after completing their degree or work in an industry that does not qualify for the employment insurance program (i.e., being self-employed). We report their employment industry in Table 3.

Table 3

Employment Sectors for Bachelor’s Graduates With Teaching Majors but Not Teaching

	Black		White
Healthcare and social services	788	17.5%	1,392	10.9%
Retail trade	638	14.2%	1,675	13.2%
Elementary and secondary schools	460	10.2%	1,790	14.1%
Professional and business services	367	8.1%	841	6.6%
Accommodation and food services	344	7.6%	1,373	10.8%
Other educational services	296	6.6%	803	6.3%
Transportation and warehousing	213	4.7%	177	1.4%
Finance and insurance	198	4.4%	399	3.1%
Other services	123	2.7%	451	3.5%
Manufacturing	111	2.5%	346	2.7%
Wholesale trade	82	1.8%	293	2.3%
Information	65	1.4%	153	1.2%
Arts, entertainment, and recreation	53	1.2%	270	2.1%
Real estate and rental and leasing	43	1.0%	137	1.1%
Other careers	21	0.5%	178	1.4%
Unmatched	705	15.6%	2,450	19.2%
Total	4,507	100.0%	12,728	100.0%

Note. This table aggregates data from the Tennessee Unemployment Insurance program to match bachelor’s degree graduates in a teaching field to their employment sector if we do not observe them becoming employed as a teacher in a Tennessee public school. Employment sectors follow North American Industry Classification System (NAICS) conventions.

Overall, we find that Black and White college graduates with teaching major degrees follow different nonteaching employment patterns (overall χ²[15] = 454.43, p < .001). Looking at individual employment industries, Black teaching major graduates are 6.6 percentage points (or 37.7%) more likely to be employed in health care and social services (17.5%) than their White peers (10.9%, joint row χ²[15] = 112.8, p < .001) and 3.9 percentage points (or 27.7%) less likely to be employed in an elementary or secondary school (10.2% vs. 14.1%, joint row χ²[15] = 37.9, p < .001).⁴ Rates of employment in other industries—including being missing from our dataset—are similar between Black and White graduates. These results suggest that Black college students who complete a teaching degree are more likely to not only exit the teacher pipeline but also leave the broader field of education altogether.

Role of Master’s Degrees as a New Entry Point

Graduate programs could serve as a way to recruit students back into the teacher pipeline, as master’s programs in teaching provide an additional opportunity for college graduates to complete the requirements for teacher licensure. This could be, for example, an additional entry point for high school subject specialists who might have completed a bachelor’s degree in a specialized field (e.g., English language literature) and want to become teachers. Thus, we explore whether the master’s degree pipeline differs between Black and White prospective teachers. We are especially interested in whether certification at the master’s level provides opportunities for the recruitment into teaching of Black college graduates with degrees in fields other than teaching, as we observe Black college students graduating in high school specialist majors at higher rates than their White peers. We report these pipelines in Figures 4 and 5 for Black and White master’s degree graduates respectively (our third sample; see the dark dotted box in Figure 1).

Figure 4 displays how Black master’s program completers enter (and exit) the teacher pipeline through graduate degrees in education. We observe 7.4% of all Black students in this group complete a master’s degree in a teaching major (N = 447) and 19.6% complete a master’s degree in a nonteaching education major (N = 1,178). We also find that 18.8% of all master’s degree graduates receive a teaching license (N = 1,132), 90.6% of whom ever end up employed as a teacher in a Tennessee public school (N = 1,025).

Figure 5 displays the same outcomes for White master’s graduates. We observe 13.8% of the students in this group complete a master’s degree in a teaching major (N = 4,325) and 18.4% complete a master’s degree in a nonteaching education major (N = 5,787). Altogether, 27.1% of all White master’s degree graduates receive a teaching license (N = 8,508), with 86.8% of these certified teachers ever becoming employed as a teacher in a Tennessee public school (N = 7,384).

These analyses reveal a disparity in licensure rates between Black and White master’s program graduates similar to—albeit smaller than—the one for undergraduate students; moreover, unlike in undergraduate programs, we also find an earlier gap in the likelihoods of graduation from a program with a teaching major. Specifically, Black master’s degree graduates (7.4%) are 6.4 percentage points (or about 46.0%) less likely than their White peers (13.8%) to complete a master’s degree in a teaching field (joint row χ²[3] = 159.7, p < .001) and 5.9 percentage points (or 44.3%) less likely to receive a teaching license (4.7% compared to 10.6%, joint row χ²[7] = 181.9, p < .001). Moreover, other master’s degree pathways into licensure contribute a similar proportion of licensed teachers for Black and White master’s students. Together, these results suggest that—for Black students who completed their bachelor’s outside of a teaching field—master’s degrees do not serve as a channel for correcting the racial disparities in entry into teaching observed at the undergraduate level and in fact may even exacerbate the underrepresentation of Black teachers in the state.

Finally, we find that Black master’s students who become licensed teachers (90.6%) are 3.8 percentage points (or 4.2%) more likely than their White peers (86.8%) to be employed as public school teachers (joint row χ²[1] = 1.6, p = .206). This result, though not statistically significant, is consistent with the employment rate pattern observed for undergraduate students.

Employment for Graduates Who Do Not Become Teachers

In a similar analysis as above, we follow master’s graduates with demonstrated interest in teaching but who do not become employed as public school teachers into the workforce. This sample includes all master’s students who became certified but not employed regardless of their major, as well as unlicensed graduates whose master’s degree majors were in teaching and nonteaching education (but not high school specialist or general) fields (see the solid medium gray box at the bottom of Figure 1). We report their employment industry in Table 4.

Table 4

Employment Sectors for Master’s Graduates in Teaching Majors but Not Teaching

	Black		White
Elementary and secondary schools	130	17.9%	753	20.7%
Other educational services	129	17.8%	586	16.1%
Healthcare and social services	120	16.5%	508	13.9%
Retail trade	76	10.5%	326	9.0%
Professional and business services	49	6.7%	142	3.9%
Finance and insurance	23	3.2%	55	1.5%
Accommodation and food services	21	2.9%	273	7.5%
Other services	15	2.1%	103	2.8%
Other careers	52	7.2%	277	7.6%
Unmatched	111	15.3%	619	17.0%
Total	726	100.0%	3,642	100.0%

Note. This table aggregates data from the Tennessee Unemployment Insurance program to match master’s degree graduates in education to their employment industry if we do not observe them being employed as a teacher in a Tennessee public school. Employment sectors follow North American Industry Classification System (NAICS) conventions.

While we find that these Black and White master’s graduates follow overall different employment patterns (overall χ²[9] = 49.79, p < .001), we do not observe significant differences in their employment in elementary and secondary schools (17.9% vs. 20.7%, joint row χ²[9] = 2.3, p = .986) or other educational services (17.8% vs. 16.1%, joint row χ²[9] = 1.0, p = .999). These results do not suggest any differential exit patterns between Black and White master’s students from the broader field of education at this point of the teacher pipeline.

Illustrative Exploratory Analyses

Overall, this study was designed to provide a bird’s eye view of the teacher pipelines for Black and White students in order to identify the stages at which the former disproportionately exit. We believe that the next step in this work involves more deeply interrogating what might explain observed disparities at each specific stage and, related, whether these disparities exist to the same extent among specific subgroups of prospective teachers or at specific kinds of programs. Though such analyses largely fall beyond the scope of the current study, we offer three illustrative examples of how subsequent study might uncover additional complexities within the Black teacher pipeline. Each of these analyses takes a different explanatory approach: one more closely investigates the mechanisms at play in the disparities occurring at a single stage (in this case, focusing on the stage of certification where we find racial disparities to be greatest); a second explores whether and how these disparities differ when we focus on a subset of programs (in this case, HBCUs, which have a strong reputation for serving Black students); and a third examines how patterns change when looking into specific subgroups of Black and White students (in this case, in- versus out-of-state students, as we find a higher proportion of out-of-state Black students). We select three compelling lines of inquiry with strong theoretical rationales but also underscore that there are countless other explanations and analytic choices that merit future research (as we elaborate in our discussion).

Praxis Test-Taking and Licensure Rates

In analyses above, we find that, among teaching majors, Black graduates are disproportionally less likely to ultimately become certified as teachers. Three possible explanations are likely to account for the racial differences observed at this stage, all related to the requirements for teacher licensure in the state: differences in (a) completing approved coursework, (b) completing student teaching, and (c) passing the teacher certification examination requirements. Given that prior literature has demonstrated racial disparities in certification passage rates (Bennett et al., 2006; Goldhaber & Hansen, 2010), we suspect that this is the most likely explanation.

Unfortunately, our data do not allow us to conclusively pinpoint the exact mechanism behind the disparity in licensure rates; however, we can observe certification exam-taking behavior for a limited subsample of the college graduates in this article (N = 11,775 students who took the Praxis test between 2012 and 2016). Among these test-takers (N_b = 1,343, N_w = 9,850), Black college graduates (70.6%) are 37.2 percentage points (or 52.7%) more likely than their White peers (33.4%) to fail a Praxis test (joint row χ²[1] = 432.7, p < .001) and about 22.6 percentage points (or 23.5%) less likely to eventually pass a Praxis test (73.6% compared to 96.3%, joint row χ²[1] = 64.7, p < .001). In other words, failure to meet teacher certification examination requirements could explain about 40% of the difference we observe in exits between Black and White college students from completing a teaching major to receiving certification; however, the majority of observed disparities still remains unexplained.⁵

Historically Black Colleges and Universities

Another possible explanation for racial disparities in licensure rates might involve the fact that the higher education landscape comprises mostly primarily White institutions (PWIs) that often present structural, social, and other challenges for Black students not faced by their White peers (Carter Andrews et al., 2019; Kohli, 2009). Given their strong reputation for supporting Black students (Gasman et al., 2017), we investigate whether the rates of progression through each stage of the pipeline are higher for Black alumni of HBCUs than for Black students graduating from other schools. In these analyses (see Appendix Tables A4 through A7), we found that the percentage of Black bachelor’s graduates of HBCUs progressing through at each stage of the pipeline is lower, on average, than the proportion of Black graduates of non-HBCUs. For example, we find that Black HBCU graduates with bachelor’s teaching majors (11.1%) are 10.1 percentage points (or 47.6%) less likely than their peers at non-HBCUs (21.2%) to earn teacher licenses (joint row χ²[2] = 40.1, p < .001) and 6.5 percentage points (or 49.6%) less likely to pursue a master’s degree (6.6% compared to 13.1%, joint row χ²[2] = 26.3, p < .001). However, licensed Black graduates of HBCUs find employment as a teacher at similar rates as Black graduates of non-HBCUs (80.9% compared to 84.3%, joint row χ²[1] = 0.1, p = .752).

These trends were true also for our MA sample, though differences among Black HBCU and non-HBCU graduates were somewhat less pronounced. Of primary interest, the certification exam (Praxis) passage rates among Black students in HBCUs were lower than passage rates for Black peers at non-HBCUs; we find that Black master’s graduates of HBCUs (61.8%) are 13.6 percentage points (or 18.0%) less likely than their peers at non-HBCUs (75.4%) to pass a Praxis test (joint row χ²[1] = 3.7, p = .054).

Conversely, the small group of White bachelor’s graduates at HBCUs demonstrated higher rates of earning teaching licenses (69.8% compared to 50.6%, joint row χ²[2] = 13.1, p < .01) and higher but statistically comparable rates of transitioning into teaching careers (82.3% compared to 78.2%, joint row χ²[1] = 0.3, p = .584) compared to White students at non-HBCUs; rates of progression for White master’s students graduating from HBCUs and non-HBCUs were similar. Overall, these analyses provide no indication that HBCUs are doing a better job than PWIs at supporting the progression of Black students along the pipeline.

In- Versus Out-of-State Students

One additional potential explanation for the disparity in licensure rates among Black and White teachers stems from the nature of our statewide data. Graduates who intend to teach in states other than Tennessee may not choose to report the results of their Praxis tests to the Tennessee Department of Education; if this preference disproportionately exists among Black students, then our lack of out-of-state licensure data may overstate any racial disparities in the likelihood of earning certification. Given that a higher proportion of Black students in our full (20.5%), undergraduate (12.4%), and graduate (11.9%) samples initially resided outside of Tennessee than among their White peers (12.7%, 8.0%, and 6.7%, respectively), we explore whether racial disparities in licensure that we observed above persist even when restricting to samples of in-state students, who we assume would be equally likely among both Black and White populations to pursue certification and seek employment in Tennessee.

In these analyses (see Appendix Tables A8 through A10), we find that, as expected, out-of-state students are less likely to show up in our data as licensed and employed regardless of race, perhaps due to their desire to return to their home states to teach. However, the higher proportion of out-of-state students among our Black subsamples explains at most a small portion of the disparities we document in their progression through the teaching pipeline. As an example, among Black undergraduates who graduated with teaching majors, in-state students (21.1%, N = 1,070) were more than twice as likely as out-of-state students (9.3%, N = 67) to obtain a teaching license (joint row χ²[2] = 44.3, p < .001), whereas the certification rates of in- (51.0%, N = 12,780) and out-of-state (47.7%, N = 1,039) students were much more similar among White students (joint row χ²[2] = 4.4, p = .111). These results do suggest that the higher proportion of out-of-state Black students may be partially responsible for the large gap in licensure rates among Black and White prospective teachers; at the same time, even among in-state students only, the disparity remains quite large (29.9 percentage points) and nearly identical to that for the full sample (31.1 percentage points), given the small total proportion of out-of-state students and their still qualitatively similar racial disparity (38.4 percentage points). Since the patterns of disparities among in-state students at all other stages of the pipeline look very similar to the patterns of all students, we do not believe the higher proportion of out-of-state Black students explains licensure disparities between Black and White graduates with teaching majors. Moreover, whether these students found employment as teachers in another state (or, similarly, in a private school) or not, their exit from this final stage of the pipeline contributes equally to the racial disparities and underrepresentation of Black teachers in Tennessee’s public school system.

Discussion

Our study is among the first to follow the same group of college students from a statewide sample through the steps required to become a teacher while documenting differences between Black and White potential teachers at each step. We focus on college graduates who ever declared a teaching major to focus on those who showed interest in teaching as a career and took the first step into the teacher pipeline. We then follow these students through completing this major, receiving a teacher license, and finally becoming employed as a teacher in a public school. We discuss our four main findings below, as well as the results of our analysis of master’s students.

First, initial interest in the teaching profession—operationalized by declaring and completing a teaching major—is similar between Black and White students completing bachelor’s programs in undergraduate institutions across Tennessee. Roughly 18% of all undergraduate program completers ever declared a teaching major during their program, and about 11.5% receive a teaching major for their bachelor’s degree. Notably, and distinct from previous studies, these rates are not different for Black or White graduates. These distinctions may stem from our focus on a statewide sample rather than the entire nation (USDOE, 2016) or a single institution (Bartanen & Kwok, 2022); they may also derive from our operationalization of interest in teaching as declaring a teaching-related major, rather than declaring any education major (USDOE, 2016) or expressing interest on a college application (Bartanen & Kwok, 2022). Regardless of the explanation, our results suggest that Black and White college graduates in Tennessee are equally interested in a career in teaching early in the pipeline; moreover, given equal degrees of “attrition” from initial interest in a teaching major to degree completion among both Black and White graduates, our findings do not seem to indicate that, at least collectively across the state, programs are worse at supporting and retaining interested prospective Black teachers than their White peers on the path to graduation.

After college graduation, however, we find stark differences in progression along the pipeline for Black and White college students. Among undergraduates who receive a teaching major, Black students are much less likely (by 31 percentage points) to become licensed in the state than their White peers. This disparity in licensure rates appears to be the single largest contributor to the underrepresentation of Black teachers in the workforce among the pipeline stages we observe in Tennessee. These certification disparities are somewhat consistent, though more extreme, than those observed in Washington where fewer Hispanic (61%) and non-Hispanic candidates of color (63%) became certified as compared to White (66%) candidates (Greenberg Motamedi et al., 2021). Two of the potential mechanisms we illustratively explore above explain some of the racial disparities we observe in licensure rates: (a) Black graduates are more likely than their White peers to fail certification exams, and (b) Black graduates disproportionately come from outside of Tennessee and therefore may be less likely to report their licensure scores to our state longitudinal data system. However, as these explanations seem to account for no more than half of the observed disparities in licensure rates, more research is needed to interrogate the mechanisms behind remaining racial disparities.

An additional explanation could involve Black students completing other steps toward licensure that surpass what is necessary for graduation—for example, specific certification program coursework or student teaching requirements—at lower rates than their White peers. Consistent with this explanation, some qualitative studies document the racialized experiences that Black student teachers face during their teacher preparation, particularly at PWIs, which might lead them to being pushed out of a teaching career (Carter Andrews et al., 2019; Frank, 2019). However, in our third exploratory analysis, we find that licensure rates are lower among Black students in HBCUs than Black students at non-HBCUs; assuming that these kinds of experiences are less prevalent at HBCUs, these results do not provide any suggestive evidence that racialized experiences would explain any of the remaining disparities in licensure rates.

Focusing on the employment differences of teacher candidates who receive licenses, Black licensed teachers are actually more likely than their White peers to become employed as public school teachers in the state. This finding again contrasts with prior research such as Goldhaber et al. (2014) and Bartanen and Kwok (2022) that found recently certified White teachers to have greater employment rates than ToCs in Washington and Texas, respectively. Although we are again not able to disentangle potential explanations for this contrast, this result helps pinpoint certification as the stage at which Black students most disproportionately exit the teaching pipeline by largely eliminating the hiring process as a major source of racial disparities.

Finally, among those who receive a bachelor’s degree in a teaching major but do not become employed as teachers, Black students are more likely than their White peers to seek employment in healthcare and other social services. Similarly, potential Black teachers who exit the pipeline after graduation are less likely to remain involved in education (e.g., in noninstructional roles or in non-public-school settings) than their White peers, implying a more significant and permanent departure from teaching. This finding merits further study, with particular effort toward ascertaining the extent to which this divergent result is attributable to the individual decisions and preferences of graduates for other fields of work or the institutional barriers that may restrict their possibilities of entering the educational workforce broadly (or the interplay among the two).

We additionally explore the extent to which master’s degrees might provide an alternative pathway into the teaching profession for those who did not complete a bachelor’s in a teaching major. A graduate degree in education could provide a later opportunity to recruit new people in the teaching profession from majors outside of teaching fields, as well as career changers; therefore, we wondered specifically if graduate programs might provide prospective Black educators another route into teaching that could offset the racial disparities we observe at the undergraduate level. However, we find that Black master’s degree students are less likely to both pursue an education degree and receive a teacher license than their White peers in graduate institutions. This new entry point into teaching not only fails to close the racial gap in licensing rates observed at the undergraduate level but may actually exacerbate it.

Though our study is only capable of describing—and not necessarily understanding the underlying reasons for—racial disparities in the teaching pipeline, it does begin to at least suggest possible avenues for addressing these inequities through policy. For example, a high-level policy initiative focused on the undergraduate period between declaring a teaching-related major and getting certified offers the most promise for ensuring similar rates of progression among Black and White students. However, more research is needed to understand which dimensions of that period (e.g., certification exams, student teaching completion) contribute most to observed inequities in order to better inform such a policy. The illustrative exploratory analyses we present in this article exemplify the multiple directions in which this inquiry could head, including examining in greater focus the mechanisms that constitute a given stage, exploring whether certain kinds of programs do a better job of supporting the progression of prospective Black teachers along the pipeline (and why), or determining whether disparities differ for subgroups of Black and White prospective educators who share other common characteristics (e.g., endorsement area).

That said, a broader question policymakers need to consider is whether “similar rates of progression” are enough. Given the overrepresentation of White teachers in the workforce, a strong case can be made that more should be done to increase the representation of Black students at all phases of the pipeline, especially given the overwhelming evidence that Black and Brown teachers matter for all students and especially for students of color. To that end, alternative teacher preparation models designed to recruit prospective ToCs may prove to serve as critical entries to teaching that help address this overrepresentation, though whether these nontraditional pathways more effectively address the racial disparities in the educator workforce also merits further study. Additionally, by narrowing our understanding of disparities in the teaching profession to the steps between undergraduate major declaration and initial employment, we offer little insight into how the equally crucial stages that occur after teachers find their first jobs that may shape racial disparities in the teaching profession, including inequities in school settings, mobility, and attrition, which we believe is a critical area in need of further study.

Our work in this article is limited by the availability of various data sources, each of which suggests potential fruitful avenues for future research. Access to more detailed course-taking data and teacher licensure exam attempts could allow a richer description of the steps involved in receiving a teacher license after graduation, in turn helping us better understand racial disparities we observe and possible policy levers for addressing them. Pairing these quantitative analyses with qualitative data about Black teacher candidate decision-making at this point of the teacher certification process could also help identify program features either push ToCs out of the pipeline or influence them to consider pursuing other careers and exit before they become licensed teachers, as well as possible ways to increase the representation of Black students who enroll in teaching-related majors and teacher certification programs early on. Finally, our results about the differences in career choices for Black and White students who complete a teaching major raise important questions about which factors lead to racial and ethnic differences in completing various pipeline stages, and especially gaining certification, where we observe the largest disparities. We believe that further exploration of the reasons behind Black and White students’ career choices could be a fruitful avenue for future research.

Footnotes

Appendix

Table A10

Outcomes for In-State Versus Out-of-State Students Who Receive an MA

		In-state		Out-of-state
		Black	White	Black	White
Panel A—Recruitment of master’s degree students
General (MA)		2,984[56.3%; 3.4]	15,829[54.1%; 0.6]	389[54.3%; 2.3]	1,270[60.1%; 0.8]
HS specialist (MA)		833[15.7%; 19.8]	3,824[13.2%; 3.6]	189[26.4%; 17.8]	361[17.1%; 6.0]
Nonteaching educ. (MA)		1,075[20.3%; 2.6]	5,599[19.1%; 0.5]	103[14.4%; 11.7]	188[8.9%; 4.0]
Teaching (MA)		412[7.8%; 106.6]	4,032[13.8%; 19.3]	35[4.9%; 27.8]	293[13.9%; 9.4]
N		5,304	29,284	716	2,112
Panel B—Teacher licensing
General (MA)	Teacher license	49[0.9%; 0.2]	249[0.9%; 0.0]	3[0.4%; 0.0]	8[0.4%; 0.0]
General (MA)	Exit	2,935[55.3%; 3.2]	15,580[53.2%; 0.6]	386[53.9%; 2.3]	1,262[59.8%; 0.8]
HS specialist (MA)	Teacher license	66[1.2%; 15]	624[2.1%; 2.7]	8[1.1%; 0.8]	33[1.6%; 0.2]
HS specialist (MA)	Exit	767[14.5%; 41.4]	3,200[10.9%; 7.5]	181[25.3%; 21.1]	328[15.5%; 7.1]
Nonteaching educ. (MA)	Teacher license	673[12.7%; 7.2]	4,192[14.3%; 1.3]	52[7.3%; 6.9]	91[4.3%; 2.3]
Nonteaching educ. (MA)	Exit	402[7.6%; 56]	1,407[4.8%; 10.1]	51[7.1%; 4.9]	97[4.6%; 1.7]
Teaching (MA)	Teacher license	272[5.1%; 119.8]	3,133[10.7%; 21.7]	9[1.3%; 31.1]	178[8.4%; 10.5]
Teaching (MA)	Exit	140[2.6%; 2.3]	899[3.1%; 0.4]	26[3.6%; 2.6]	115[5.5%; 0.9]
N		5,304	29,284	716	2,112
Panel C—Employment as a teacher
Teacher license	Employed as teacher	963[90.9%; 0.8]	7,215[88.0%; 0.1]	62[86.1%; 7.8]	169[54.5%; 1.8]
Teacher license	Exit	97[9.1%; 5.7]	983[12.0%; 0.7]	10[13.9%; 12]	141[45.5%; 2.8]
N		1,060	8,198	72	310

Note. Column percentages and contribution to total Pearson’s chi square in brackets. Total Pearson’s chi square for Panel A: χ²(3) = 156.3, p < .001 (in-state); χ²(2) = 79.7 p < .001 (out-of-state). Total Pearson’s chi square for Panel B: χ²(7) = 289.5, p < .001 (in-state); χ²(7) = 93.0, p < .001 (out-of-state). Total Pearson’s chi square for Panel C: χ²(1) = 7.3, p < .01 (in-state); χ²(1) = 24.4, p < .001 (out-of-state).

Acknowledgements

This project would not have been possible without the partnership, support, and data provided by the Tennessee Department of Education (TDOE). Notwithstanding any TDOE data or involvement in the creation of this research product, the TDOE does not guarantee the accuracy of this work or endorse the findings. Any errors are the sole responsibility of the author(s).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Matthew Ronfeldt and Jae Eun Choi received financial support from the Institute of Education Sciences (IES), U.S. Department of Education (PR/Awards R305A220054). Matthew Truwit and Emanuele Bardelli received pre-doctoral support from the Institute of Education Sciences (IES), U.S. Department of Education (PR/Awards R305B150012 and R305B200011).

Open Practices

Both code and details on how to apply for the data used in this study are available at .

ORCID iDs

Emanuele Bardelli

Matthew Truwit

Matthew Ronfeldt

Notes

Authors

EMANUELE BARDELLI is executive director of information and evaluation at Santa Rosa City Schools; ebardelli@srcs.k12.ca.us. His research focuses on the development, implementation, and evaluation of local educational policies to advance equity.

MATTHEW TRUWIT is a graduate student pursuing his doctorate in quantitative research methods in education and a master’s degree in statistics at the University of Michigan; mtruwit@umich.edu. His research evaluates and interrogates the consequences of educational policy on student development and teacher quality.

JAE EUN CHOI is a graduate student pursuing her doctorate in teaching and teacher education at the University of Michigan; jaeeunc@umich.edu. Her research involves the evaluation of educational policies aimed at improving teacher quality and promoting educational equity.

MATTHEW RONFELDT is a professor of educational studies at the University of Michigan School of Education, 610 E. University Ave., Ann Arbor, MI 48109; ronfeldt@umich.edu. His research aims to improve teaching quality by focusing on preservice and inservice teacher education, teacher labor markets, the organizational contexts of schools, and the assessment of teaching and teacher preparation.

References

Anderson

K. A.

(2020). A national study of the differential impact of novice teacher certification on teacher traits and race-based mathematics achievement. Journal of Teacher Education, 71(2), 247–260. https://doi.org/10.1177/0022487119849564

Bardelli

Ronfeldt

(2021). Workforce outcomes of Tennessee program completers in high needs subject areas. American Journal of Education, 128(1), 59–93.

Bartanen

Kwok

(2022). From interest to entry: The teacher pipeline from college application to initial employment (EdWorkingPaper: 22-535). Annenberg Institute at Brown University. https://doi.org/10.26300/hqn6-k452

Bennett

C. I.

McWhorter

L. M.

Kuykendall

J. A.

(2006). Will I ever teach? Latino and African American students’ perspectives on PRAXIS I. American Educational Research Journal, 43(3), 531–575. https://doi.org/10.3102/00028312043003531

Blazar

(2021). Teachers of color, culturally responsive teaching, and student outcomes: Experimental evidence from the random assignment of teachers to classes (EdWorkingPaper No. 21–501). Annenberg Institute for School Reform at Brown University. https://doi.org/10.26300/jym0-wz02

Carter Andrews

D. J.

Castro

Cho

C. L.

Petchauer

Richmond

Floden

. (2019). Changing the narrative on diversifying the teaching workforce: A look at historical and contemporary factors that inform recruitment and retention of teachers of color. Journal of Teacher Education, 70(1), 6–12. https://doi.org/10.1177/0022487118812418

Carver-Thomas

(2018). Diversifying the teaching profession: How to recruit and retain teachers of color. Learning Policy Institute. https://doi.org/10.54300/559.310

Cherng

H.-Y. S.

Halpin

P. F.

(2016). The importance of minority teachers: Student perceptions of minority versus White teachers. Educational Researcher, 45(7), 407–420. https://doi.org/10.3102/0013189X16671718

Christian

Ronfeldt

Zafar

(2024). College students and career aspirations: Nudging student interest in teaching (EdWorkingPaper: 24-999). Annenberg Institute at Brown University. https://doi.org/10.26300/0rb6-5r94

10.

Cowan

Goldhaber

Hayes

Theobald

(2016). Missing elements in the discussion of teacher shortages. Educational Researcher, 45(8), 460–462. https://doi.org/10.3102/0013189X16679145

11.

Dee

T. S.

(2005). A teacher like me: Does race, ethnicity, or gender matter? American Economic Review, 95(2), 158–165.

12.

Drake

Ellis

Moorer

Walsh

(2021). Program diversity and admissions. National Council on Teacher Quality. www.nctq.org/publications/Teacher-Prep-Review:-Program-Diversity-and-Admissions-2021

13.

Fenwick

(2021). The history, current use, and impact of entrance and licensure examinations cut scores on the teacher-of-color pipeline: A structural racism analysis (p. 27). Consortium for Research-Based and Equitable Assessments, American Association of Colleges for Teacher Education (AACTE).

14.

Frank

T. J.

(2019). Using critical race Theory to unpack the Black mathematics teacher pipeline. In Davis

Jett

C. C.

(Eds.), Critical race theory in mathematics education (pp. 98–122). Routledge.

15.

Gasman

Samayoa

A. C.

Ginsberg

(2017). Minority serving institutions: Incubators for teachers of color. The Teacher Educator, 52(2), 84–98. https://doi.org/10.1080/08878730.2017.1294923

16.

Gershenson

Hart

C. M.

Hyman

Lindsay

C. A.

Papageorge

N. W.

(2022). The long-run impacts of same-race teachers. American Economic Journal: Economic Policy, 14(4), 300–342.

17.

Gist

C. D.

(2019). Special issue: Grow your own programs and teachers of color: Taking inventory of an emerging field. Teacher Education Quarterly, 46(1), 5–8.

18.

Gitomer

D. H.

Phelps

Weren

B. H.

Howell

Croft

A. J.

(2015). Evidence on the validity of content knowledge for teaching assessments. In Kane

T. J.

Kerr

K. A.

Pianta

R. C.

(Eds.), Designing teacher evaluation systems: New guidance from the measures of effective teaching project (pp. 493–528). Jossey-Bass.

19.

Goldhaber

Hansen

(2010). Race, gender, and teacher testing: How informative A tool is teacher licensure testing? American Educational Research Journal, 47(1), 218–251. https://doi.org/10.3102/0002831209348970

20.

Goldhaber

Krieg

J. M.

Theobald

(2014). Knocking on the door to the teaching profession? Modeling the entry of prospective teachers into the workforce. Economics of Education Review, 43, 106–124. https://doi.org/10.1016/j.econedurev.2014.10.003

21.

Greenberg Motamedi

Yoon

S. Y.

Hanson

. (2021). Pathways to teaching: Teacher diversity, testing, certification, and employment in Washington State (REL 2021-094). Regional Educational Laboratory Northwest.

22.

Grissom

J. A.

Kern

E. C.

Rodriguez

L. A.

(2015). The “representative bureaucracy” in education: Educator workforce diversity, policy outputs, and outcomes for disadvantaged students. Educational Researcher, 44(3), 185–192. https://doi.org/10.3102/0013189x15580102

23.

Guha

Hyler

M. E.

Darling-Hammond

(2016). The teacher residency: An innovative model for preparing teachers. Learning Policy Institute. https://learningpolicyinstitute.org/sites/default/files/product-files/Teacher_Residency_Innovative_Model_Preparing_Teachers_REPORT.pdf

24.

Hughes

Bailey

C. M.

Warren

P. Y.

Stewart

E. A.

(2020). “Value in diversity”: School racial and ethnic composition, teacher diversity, and school punishment. Social Science Research, 92, 102481. https://doi.org/10.1016/j.ssresearch.2020.102481

25.

Kizilcec

R. F.

Baker

R. B.

Bruch

Cortes

K. E.

Hamilton

L. T.

Lang

D. N.

Pardos

Z. A.

Thompson

M. E.

Stevens

M. L.

(2023). From pipelines to pathways in the study of academic progress. Science, 380(6643), 344–347.

26.

Klein

S. P.

Bolus

(1997). The size and source of differences in Bar exam passing rates among racial and ethnic groups. The Bar Examiner.

27.

Kohli

(2009). Critical race reflections: Valuing the experiences of teachers of color in teacher education. Race Ethnicity and Education, 12(2), 235–251. https://doi.org/10.1080/13613320902995491

28.

Petchauer

(2012). Teacher licensure exams and Black teacher candidates: Toward new theory and promising practice. Journal of Negro Education, 81(3), 252–267. https://doi.org/10.7709/jnegroeducation.81.3.0252

29.

Redding

(2019). A teacher like me: A review of the effect of student–teacher racial/ethnic matching on teacher perceptions of students and student academic and behavioral outcomes. Review of Educational Research, 89(4), 499–535. https://doi.org/10.3102/0034654319853545

30.

State Collaborative on Reforming Education. (2016). Prepared for day one: Improving the effectiveness of early-career teaching. State Collaborative on Reforming Education. https://tnscore.org/wp-content/uploads/2018/09/Prepared-From-Day-One_TeacherPrepPolicyReport2016.pdf

31.

Sutcher

Darling-Hammond

Carver-Thomas

(2016). A coming crisis in teaching? Teacher supply, demand, and shortages in the U.S. Learning Policy Institute. http://learningpolicyinstitute.org/sites/default/files/product-files/A_Coming_Crisis_in_Teaching_REPORT.pdf

32.

Tennessee Department of Education. (2018). Teacher and administrator racial and ethnic diversity in Tennessee. https://www.tn.gov/content/dam/tn/education/reports/rpt_teacher_admin_diversity.pdf

33.

Tennessee State Board of Education. (2022). Teacher preparation. https://teacherprepreportcard.tn.gov/teacher-prep

34.

U.S. Department of Education (USDOE). (2016). The state of racial diversity in the educator workforce. https://www2.ed.gov/rschstat/eval/highered/racial-diversity/state-racial-diversity-workforce.pdf

35.

Yeo

H. L.

Dolan

P. T.

Mao

Sosa

J. A.

(2020). Association of demographic and program factors with American Board of Surgery qualifying and certifying examinations pass rates. JAMA Surgery, 155(1), 22–30. https://doi.org/10.1001/jamasurg.2019.4081

The Black Teacher Pipeline: Insights From a State Longitudinal Data System

Abstract

Keywords

Literature Review

Majoring in Education

Teacher Licensure

Employment as a Teacher

Contributions

Methods

Data

Categorizing Majors

Samples

Modeling the Teacher Pipeline

Statistical Inference

Results

Preference Towards Teaching

Outcomes for College Graduates With a Teaching Major

Employment for Graduates Who Do Not Become Teachers

Role of Master’s Degrees as a New Entry Point

Employment for Graduates Who Do Not Become Teachers

Illustrative Exploratory Analyses

Praxis Test-Taking and Licensure Rates

Historically Black Colleges and Universities

In- Versus Out-of-State Students

Discussion

Footnotes

Appendix

Acknowledgements

Declaration of Conflicting Interests

Funding

Open Practices

ORCID iDs

Notes

Authors

References