Recruitment, employment, retention and the minority teacher shortage Special Issue Understanding and Solving Teacher Shortages: Policy Strategies for a Strong Profession education policy analysis archives A peer-reviewed, independent, open access, multilingual journal Arizona State University Volume 27 Number 37 April 8, 2019 ISSN 1068-2341 Recruitment, Employment, Retention and the Minority Teacher Shortage1 Richard Ingersoll University of Pennsylvania Henry May University of Delaware & Gregory Collins University of Pennsylvania United States Citation: Ingersoll, R., May, H., & Collins, G. (2019). Recruitment, employment, retention and the minority teacher shortage. Education Policy Analysis Archives, 27(37). http://dx.doi.org/10.14507/epaa.27.3714 This article is part of the special issue, Understanding and Solving Teacher Shortages: Policy Strategies for a Strong Profession, guested edited by Linda Darling- Hammond and Anne Podolsky. 1 This article draws extensively from an earlier working report (Ingersoll, May & Collins, 2017). The research from which this article is drawn was supported by the Sandler Foundation, the Albert Shanker Institute, and the Sally Hewlett and the Flora Family Foundation. Journal website: http://epaa.asu.edu/ojs/ Facebook: /EPAAA Twitter: @epaa_aape Manuscript received: 2/9/2018 Revisions received: 4/11/2018 Accepted: 3/24/2019 http://dx.doi.org/10.14507/epaa.27.3714 Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 2 Abstract: This study examines and compares the recruitment, employment, and retention of minority and nonminority school teachers over the quarter century from the late 1980s to 2013. Our objective is to empirically ground the ongoing debate regarding minority teacher shortages and changes in the minority teaching force. The data we analyze are from the National Center for Education Statistics’ nationally representative Schools and Staffing Survey (SASS) and its longitudinal supplement, the Teacher Follow -up Survey (TFS). Our data analyses document the persistence of a gap between the percentage of minority students and the percentage of minority teachers in the US . But the data also show that this gap is not due to a failure to recruit new minority teachers . In the two decades since the late 1980s, the number of minority teachers almost doubled, outpacing growth in both the number of White teachers and the number of minority students. Minority teachers are also overwhelmingly employed in public schools serving high - poverty, high-minority and urban communities. Hence, the data suggest that widespread efforts over the past several decades to recruit more minority teachers and employ them in disadvantaged schools have been very successful. But, these efforts have also been undermined because minority teachers have significantly higher turnover than White teachers and this is strongly tied to poor working conditions in their schools. Keywords: teacher quality; recruitment; retention; minority teachers Reclutamiento, empleo, retención y escasez de profesores minoritarios Resumen: El presente estudio analizar y comparar el reclutamiento, el empleo y la retención de los profesores minoritarios y los escolares nonminority a partir de finales de los años 1980 a 2013. Analizamos los datos de las Schools and Staffing Survey (SAS S) y su suplemento longitudinal, el Teacher Follow-up Survey (TFS). Nuestros análisis documentan la persistencia de una brecha entre el porcentaje de alumnos y el porcentaje de profesores en los Estados Unidos. Los datos también muestran que esta laguna no se debe a la falta de reclutamiento de nuevos profesores minoritarios. Desde el final de los años 80, el número de profesores minoritarios casi se ha duplicado, superando el número de profesores blancos y el número de otros alumnos. Los profesores minorit arios también son empleados en escuelas públicas que atienden a comunidades de alta pobreza, minorías y urbanas. Así, los datos sugieren que los esfuerzos para reclutar más profesores minoritarios y emplearlos en escuelas carentes tuvieron mucho éxito. Est os esfuerzos también se vieron perjudicados porque tienen una rotación significativamente mayor que los profesores y esta situación está fuertemente ligada a condiciones de trabajo precarias en sus escuelas. Palabras clave: calidad del profesor; contratación; retención; maestros Recrutamento, emprego, retenção e escassez de professores minoritários Resumo: O presente estudo analisar e comparar o recrutamento, emprego e retenção de professores minoritários e escolares nonminority partir do final dos anos 19 80 a 2013. Analisamos os dados da Schools and Staffing Survey (SASS) e seu suplemento longitudinal, o Teacher Follow-up Survey (TFS). Nossas análises documentam a persistência de uma lacuna entre o percentual de alunos e o percentual de professores nos EUA. Os dados também mostram que essa lacuna não se deve à falta de recrutamento de novos professores minoritários. Desde o final dos anos 80, o número de professores minoritários quase duplicou, superando o número de professores brancos e o número de outros alunos. Professores minoritários também são empregados em escolas públicas que atendem comunidades de alta pobreza, minorias e urbanas. Assim, os dados sugerem que Recruitment, Employment, Retention and the Minority Teacher Shortage 3 os esforços para recrutar mais professores minoritários e empregá -los em escolas carentes tiveram muito sucesso. Esses esforços também foram prejudicados porque eles têm uma rotatividade significativamente maior do que os professores e essa situação está fortemente ligada a condições de trabalho precárias em suas escolas. Palavras-chave: qualidade do professor; recrutamento; retenção; professores Introduction Over the past several decades, a shortage of minority school teachers has been an issue of national importance. Numerous scholars and commentators have argued that there is a growing mismatch between the degree of racial/ethnic diversity in the nation’s student population and the degree of diversity in the nation’s elementary and secondary teaching force (for reviews, see Albert Shanker Institute, 2015; Quiocho & Rios, 2000; Torres et al., 2004; Villegas & Irvine 2010; Villegas & Lucas, 2004; Villegas, Strom & Lucas, 2012; Zumwalt & Craig, 2005). Typically, scholars and commentators have held that as the nation’s population, and in turn the nation’s student body, has grown more diverse, the teaching force has not kept pace. Some go further—arguing that the teaching force has changed in the opposite direction, becoming even less diverse and more homogeneously White (e.g., Rogers-Ard et al., 2013; Lewis & Toldson, 2013; Villegas, Strom & Lucas, 2012). Commentators and researchers make three related arguments for why this mismatch is detrimental and why increasing the racial/ethnic diversity of the teaching force would be beneficial. The first focuses on demographic parity. This argument holds that minority teachers are important as role models for both minority and nonminority students. The underlying assumption is that the racial/ethnic makeup of the teaching force should reflect that of the student population, and that of the larger society. With increasing racial/ethnic diversity in the larger society, proponents hold, there is accordingly a growing need for more minority teachers as role models in schools (e.g., Albert Shanker Institute, 2015; Banks, 1995; Carnegie Forum on Education and the Economy, 1986; Cochran-Smith, 2004; Dilworth, 1992; Kirby et al., 1999; Lewis & Toldson, 2013). A second related argument focuses on what is often called “cultural synchronicity” (Irvine, 1988, 1989). This view holds that minority students benefit from being taught by minority teachers, because minority teachers are likely to have “insider knowledge” due to similar life experiences and cultural backgrounds. The assumption is that synchronicity is a valuable resource in teaching and learning (Achinstein & Aguirre, 2008; Foster, 1994; Gandara & Maxwell-Jolley, 2000; Haycock, 2001; Valencia, 2002). Proponents of this view cite a growing number of empirical studies showing that minority teachers have a positive impact on various outcomes for both minority and non- minority students (for reviews, see Villegas & Irvine, 2010; Villegas & Lucas, 2004). A third related argument concerns teacher shortages in disadvantaged schools. Minority teachers not only are likely to be well suited to teach minority students, this view holds, but they are also likely to be motivated by a “humanistic commitment” to making a difference in the lives of disadvantaged students. In turn, this argument holds, minority teachers are more likely than nonminority candidates to seek employment in schools serving predominantly minority student populations, often in low-income, urban school districts (e.g., Foster, 1997; Haberman, 1996; Ladson-Billings, 1995; Murnane et al., 1991; Quiocho & Rios, 2000). Research has shown that these same kinds of schools—urban, poor public schools serving minority students—disproportionately suffer from general teacher shortages (e.g., Liu et al., 2008). Hence, diversification of the teaching force in this view is a solution to the more general problem of teacher shortages in disadvantaged schools. Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 4 As a result of these various factors—a lack of minority teacher role models, insufficient cultural synchronicity between teachers and minority students, and a general dearth of qualified teachers in disadvantaged schools—commentators and researchers have concluded that the minority teacher shortage has resulted in unequal access to adequately qualified teachers and, hence, to quality teaching, in poor, urban public schools serving minority students. Unequal access to educational resources, such as qualified teachers, has long been considered a primary cause of the stratification of educational opportunity and, in turn, the achievement gap and, ultimately, unequal occupational outcomes for disadvantaged populations (e.g., Dreeben & Gamoran, 1986; Oakes, 1985, 1990; Rosenbaum, 1976; Wilson, 1996). Researchers have argued that there are several reasons for the continuing insufficient employment of minorities in teaching (for reviews, see Zumwalt & Craig, 2005; Villegas & Irvine, 2010; Lewis, & Toldson, 2013). Scholars have held that jobs in teaching were relatively more open to minority candidates than jobs in many other lines of work, at least through much of the past half century. However, one consequence of the 1954 Brown v. Board of Education Supreme Court decision to integrate schools, educational historians have held, was that large numbers of Black and African- American educators, in particular, were uprooted and displaced in mid-century, leading to a sharp decrease in the number of minority teachers (Fultz, 2004; Tillman, 2004; White, 2016). A dearth of minority teachers has persisted in subsequent decades, researchers have held, largely because of an inadequate labor supply pipeline into the teaching occupation. One prominent factor, such researchers hold, has been that minority student underachievement in elementary and secondary education has resulted in fewer minority students entering the postsecondary level, and lower graduation rates for those who do enter higher education (e.g., Banks, 1995). In turn, as career and employment options available to minorities have broadened in recent years, a decreasing share of this shrinking number of minority college graduates have entered teaching. In addition, researchers hold, when minority candidates do seek to enter teaching, the growth of occupational entry tests, coupled with lower pass rates on these tests by minority teaching candidates, has meant that fewer minority candidates are successful. The prevailing policy response to these minority teacher staffing problems has been to attempt to increase the supply pipeline of minority teachers (see, e.g., Albert Shanker Institute, 2015; Feistritzer, 1997; Hirsch, Koppich, & Knapp, 2001; Liu et al., 2008; Rice, Roellke, Sparks, & Kolbe, 2008; Villegas, Strom & Lucas, 2012). Over the past several decades, organizations such as the Education Commission of the States, the American Association of Colleges of Teacher Education, and the National Collaborative on Diversity in the Teaching Force have advocated for and implemented a wide range of initiatives designed to recruit minority candidates into teaching. Beginning in the late 1980s, the Ford Foundation, the DeWitt Wallace-Readers’ Digest Fund, and other foundations committed substantial funding to recruiting and preparing minority teachers. These efforts have included future educator programs in high schools, partnerships between community colleges with higher minority student enrollments and four-year colleges with teacher education programs, career ladders for paraprofessionals already in the school system, and alternative certification programs (e.g., Clewell & Villegas, 2001; Lau et al., 2007; Shen, 1998; Zeichner, 1996; Zeichner & Gore, 1990). Many of these initiatives have been designed to recruit minority teachers to teach in schools serving predominantly minority student populations, often in low-income, urban school districts. Some of these initiatives have been designed to recruit male minority teachers, in particular—often considered the group in shortest supply (e.g., Lewis, 2006; Lewis, & Toldson, 2013; Norton, 2005; Rogers-Ard et al. 2013). By the later 2000s, over half of the states had minority teacher recruitment policies (Villegas & Irvine, 2010). Given the importance of this issue and these questions, not surprisingly there has been a large and growing body of empirical research evaluating the significance of the racial/ethnic Recruitment, Employment, Retention and the Minority Teacher Shortage 5 composition of the teaching force, especially its relationship to student growth and achievement. Much of this work focuses on the degree of match or mismatch between the race/ethnicity of students and that of their teachers, and to what extent this match is tied to various student achievement outcomes (for reviews, Achinstein et al., 2010; Albert Shanker Institute, 2015; Villegas & Irvine, 2010; Villegas & Lucas, 2004; Villegas, Strom & Lucas, 2012). In contrast, there has been a surprisingly limited amount of empirical investigation of the basic levels, trends, and distribution of the demographic characteristics of the teaching force. In particular, there has been little original empirical examination, especially using nationally representative data, of how the racial/ethnic character of the teaching force has changed over recent decades, to what extent there is—or is not—sufficient employment of minorities in teaching, and the sources of minority teacher staffing problems. Underlying most of the commentary and policy on this issue has been the assumption, largely untested, that minority teacher staffing problems are rooted in the front end of the teacher supply pipeline. The assumption has been that an inadequate initial supply, coupled with barriers to entry, are the main reasons that insufficient numbers of minority teachers are employed. Thus, attention has tended to focus on identifying obstacles to recruiting minority candidates into teaching and, in turn, developing strategies to overcome these obstacles (Albert Shanker Institute, 2015; Villegas & Irvine, 2010; Villegas & Lucas, 2004; Rogers-Ard et al., 2013). In contrast, little attention has been paid to where minority teachers tend to be employed, what happens to minority teachers once they are employed, or to the role of the employing organizations in teacher staffing problems. There has been some research on the magnitude and factors behind the departures of minority teachers from schools (e.g. Bristol, 2018; Carver-Thomas & Darling-Hammond, 2019; Grissom & Keiser, 2011). However, relatively little attention has been paid to the exit end of the pipeline and the role of teacher turnover in minority teacher shortages and staffing problems. In general, as recent reviews have concluded, empirical research on minority teacher turnover has been limited, has had mixed findings, and has been inadequate to help policy address the magnitude, determinants, and consequences of minority teacher turnover, or to understand the implications of retention and turnover for shortages (Achinstein et al., 2010; Albert Shanker Institute, 2015). This study seeks to address these gaps. The Study This study uses nationally representative data to empirically ground the debate over minority teacher shortages and changes in the minority teaching force. We examine trends in the recruitment, employment, and retention of minority teachers to address several sets of research questions: Has the Number of Minority Teachers Changed? In recent decades, what changes have there been in the numbers of minority students and numbers of minority teachers in the school system, and how does this compare with nonminority students and teachers? Is there more or less racial/ethnic diversity in the teaching force? Where Are Minority Teachers Employed? What is the distribution of teachers across the school system by their race/ethnicity? In which types of schools are minority teachers employed? Are minority teachers more likely than nonminority teachers to be employed in schools serving high-poverty, urban, and high-minority student populations? Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 6 How High Is Minority Teacher Turnover In recent decades, what have been the rates of minority teacher turnover? How do these compare to nonminority teachers? What Are the Sources of Minority Teacher Turnover? What are the reasons behind the turnover of teachers, and does this differ by their race/ethnicity? What role do retirement, school demographic characteristics, and school organizational conditions play in the turnover of minority teachers, and how does this compare with nonminority teachers? What Is the Role of Minority Teacher Attrition in the Staffing Problems of Schools and in the Minority Teacher Shortage? What is the overall magnitude of minority teacher attrition—teachers leaving teaching altogether? How have minority teachers’ exit rates from teaching compared to their entry rates into teaching? If minority teacher attrition rates had been lower in recent decades, would it have made any significant difference in the growth in the total number of minority teachers employed? In the next section, we describe our data sources and define key terms and measures. In the following sections, we present the results of our data analyses sequentially for each of our five research questions. We then conclude by discussing the implications of our findings for understanding and addressing the minority teacher shortage. Data, Measures, and Methods Data The data for this study come from the National Center for Education Statistics’ (NCES) nationally representative Schools and Staffing Survey (SASS) and its supplement, the Teacher Follow- Up Survey (TFS). This is the largest and most comprehensive data source available on the staffing, occupational, and organizational aspects of elementary and secondary schools. The U.S. Census Bureau collects the SASS data for NCES from a random sample of schools stratified by state, public/private sector, and school level (for information on SASS, see NCES, 2005). There have been seven SASS cycles to date: 1987–88; 1990–91; 1993–94; 1999–00; 2003–04; 2007–08; 2011–12. Each cycle of SASS includes separate (but linked) questionnaires for school and district administrators and for a random sample of teachers in each school. After 12 months, the same schools are again contacted, and all those in the original teacher sample who had departed from their school are given a second questionnaire to obtain information on their departures. The TFS comprises this latter group, along with a representative sample of those who stayed in their teaching jobs. Unlike most previous data sources on teacher turnover, the TFS is large, comprehensive, and nationally representative, and it includes the reasons teachers themselves give for their departures, along with a wide range of information on the characteristics and conditions of the schools that employ teachers. It also is unusual in that it does not focus solely on a particular subset of separations but includes all types of departures. (For information on the TFS, see Chandler et al., 2004.) Our analysis uses data from all seven cycles of SASS/TFS to address our questions. This analysis uses data weighted to compensate for the over- and undersampling of the complex stratified survey design. Each observation is weighted by the inverse of its probability of selection in order to obtain unbiased estimates of the national population of schools and teachers in the year of the survey. Recruitment, Employment, Retention and the Minority Teacher Shortage 7 Measures and Methods Throughout this study, our definitions of minority teachers and nonminority teachers are based on Census Bureau classifications of race/ethnicity. “Nonminority” refers to those identified as “White, non-Hispanic.” We use these two terms interchangeably. “Minority” includes those identified as: Black/African American; native Hawaiian/Pacific/Islander or Asian; Native American/Indian/Alaska Native; Hispanic/Latino; and those of multiple races. “Hispanic/Latino” refers to ethnicity and includes those of all races. It is important to recognize that over half of those identifying as Hispanic also identify as White. Hence, the term “person of color” is not synonymous with minority, and, for clarity, we will not use the former term. Our classification of minority teachers and nonminority teachers is based on the SASS teacher-respondent’s self- identification of their race/ethnicity in the SASS questionnaires. Our data analyses involve three different methods and stages. In the first stage, we estimate mostly descriptive statistics to address our first four research questions. In the second stage, we follow up with a detailed multiple logistic regression analysis of the predictors of turnover to further address the fourth research question. In a third stage, we address our fifth research question by undertaking simulations of the minority teaching force under hypothetical minority teacher attrition scenarios. Next, we describe these stages of our analysis. Stage 1 In the first stage, we use all seven cycles of SASS/TFS to examine data on trends in the relative numbers of minority and nonminority students and minority and nonminority teachers, data on differences in the types of schools in which minority and nonminority teachers are employed, and data on trends in the turnover rates of minority and nonminority teachers. Research on teacher turnover has often focused solely on those leaving the occupation altogether, here referred to as teacher attrition, and has often de-emphasized those who transfer or move to different teaching jobs in other schools, here referred to as teacher migration. The logic is that teacher migration is a less significant form of turnover because it does not increase or decrease the overall supply of teachers, as do retirements and career changes and, hence, does not contribute to overall shortages. From a systemic level of analysis, this is correct. However, from the perspective of schools, employee migration is as relevant as employee attrition. The premise underlying our perspective is that whether those departing are moving to a similar job in another organization or leaving the occupation altogether, their departures similarly impact and are impacted by the organization. For this same reason, the distinction between attrition and migration is rarely noted in the larger literature on employee turnover, and research on other occupations and organizations almost always includes both (see, e.g., Price, 1977). In our analysis, we examine migration and attrition, both together and separately. For our fourth question, we examine the reasons teachers themselves give for their turnover, drawn from sets of items in the most recent (2012-13) TFS that asked teacher-respondents to indicate the importance of various factors for their departures. Self-report data such as these are useful because those departing are, of course, often in the best position to know why they are leaving. But such data are also based on subjective attributions by those who departed, introducing possible attribution bias. Stage 2 To address these limits, we follow up in a second stage of our analysis with a logistic regression analysis that examines the association of teacher turnover with individual, school, and organizational predictors. For this part of our analysis, we utilize the 2003–04 SASS and the 2004–05 TFS. The 2004–05 TFS has the advantage of having a larger sample size than the more recent 2008– 09 and 2012–13 cycles of the TFS. While levels of our variables change over time, we have found in our extensive analyses of all of the cycles of SASS/TFS that the associations between turnover and Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 8 particular teacher or school characteristics change little over time. The 2003–04 SASS sample comprises 43,358 nonminority and 7,865 minority elementary and secondary teachers. The 2004–05 TFS sample comprises 6,118 nonminority and 1,311 minority elementary and secondary teachers. The TFS includes only about 12% of teachers from the original SASS sample. To increase the sample size for our regression analyses, we combined the TFS measure of turnover with another measure of turnover collected from school principals—the 2004 Teacher Status variable—for the entire SASS teacher sample, increasing our effective sample size from about 7,500 to 51,000 teachers.2 In the regression models, the dependent variable—teacher turnover—is dichotomous based on whether each teacher remained with the school or either moved to another school or left teaching within one year after the 2003–04 SASS administration. We cumulatively examine three groups of predictors of turnover: teacher characteristics, school characteristics, and organizational conditions. Table 1 defines these variables. Table 2 provides mean teacher characteristics, school characteristics, and organizational conditions associated with the teachers in the combined SASS/TFS sample. Following previous research on teacher turnover, in the regression models we include control variables for two key individual teacher characteristics: gender and age. Because it has been found to have a U-shaped relationship to turnover (Ingersoll, 2001), we transform age into a three- category set of dummy variables—younger (less than 30), middle-aged (31–50), and older (greater than 50). Following previous research on school organization (e.g., Bryk et al., 1990; Chubb & Moe, 1990; Coleman & Hoffer, 1987; Ingersoll, 2001; Ingersoll & May 2012), in the regression models we include, as independent variables, school characteristics typically found to be important in this literature: school level and school size. To examine the role of school demographic characteristics, we also include measures of whether the school is urban, rural, or suburban, the proportion of each school’s student population at or below the poverty level (i.e., eligible for free or reduced-price lunch), the proportion of each school’s student population that is minority, and the proportion of the school faculty that is minority. Because these demographic factors are often highly 2 The Teacher Status variable has some error in distinguishing between migration (movers) and attrition (leavers). Essentially, school principals tend to overreport the number of leavers because teachers who quit their jobs often do not inform their previous schools that they have moved to another school. However, this measure is quite accurate in correctly identifying who is and is not still working at the original school. By comparing individual teacher’s values for the Teacher Status variable from SASS with confirmed final turnover from the TFS, we found the Teacher Status variable was about 93% accurate in distinguishing teachers who had departed from those who had not. More specifically, the Teacher Status variable from the SASS accurately identified 90% of confirmed leavers (i.e., 2,385 out of 2,650) as having left the teaching occupation. However, the Teacher Status variable classified 29% of confirmed movers (i.e., 559 out of 1,911) as having left the teaching occupation, and an additional 1% of confirmed movers (i.e., 18 out of 1,911) as stayers. When no distinction is made between movers and leavers, the Teacher Status variable was 92% sensitive (i.e., 4,471 out of 4,886 teachers identified as departing did, in fact, move from or leave their teaching jobs), and Teacher Status was 96% specific (i.e., 2,442 out of 2,532 teachers identified as not turning over those who did, in fact, stay in their teaching jobs). This translates to an overall accuracy rate of 93% (i.e., 6,913 out of 7,418). In our merger of the SASS and TFS measures, we corrected the Teacher Status measure using TFS data making the former approximately 96% accurate. Applying the sensitivity and specificity rates above to the uncorrected ATTRIT data (i.e., 40,563 stayers and 3,064 movers/leavers) and assuming 100% accuracy for those teachers included in the TFS data (i.e., 2,864 stayers and 4,565 movers/leavers), we end up with an overall accuracy rate of 96% (i.e., [(40,563 x .96) + (3,064 x .92) + (2,864 x 1.00) + (4,565 x 1.00)] / 51,056 = 0.96). Recruitment, Employment, Retention and the Minority Teacher Shortage 9 Table 1 Definitions of Measures Utilized in the Regression Analysis Teacher Turnover: a dichotomous variable where 1 = not teaching in same school as last year and 0 = stayer/currently teaching in same school. Teacher Characteristics Young: a dichotomous variable where 1 = teacher less than 30 years of age and 0 = other teachers. Old: a dichotomous variable where 1 = teacher older than 50 years of age and 0 = other teachers. Male: a dichotomous variable where 1 = male teacher and 0 = female teacher. School Characteristics Rural: a dichotomous variable where 1 = rural and 0 = suburban or urban. Suburban: a dichotomous variable where 1 = suburban and 0 = rural or urban. Secondary Level: a dichotomous variable where 1 = junior or senior secondary and 0 = elementary or middle or combined (K-12). Size: student enrollment of school. Poverty Enrollment: percentage of students eligible for the federal free or reduced-price lunch program for students from families below poverty level Minority Enrollment: percentage of minority students Minority Faculty: percentage of minority teachers Organizational Characteristics/Conditions Highest Salary: for districts with a salary schedule for teachers, normal yearly base salary highest step, or if no district salary schedule, the highest teacher yearly base salary, as reported by school administrators. Student Discipline Problems: on a scale of 1 = never happens to 5 = happens daily, the school mean of teachers’ reports for 8 kinds of student discipline problems: disruptive behavior; absenteeism; physical conflicts among students; robbery; vandalism; weapon possession; physical abuse of teachers; verbal abuse of teachers. School Leadership Support: on a scale of 1 = strongly disagree to 4 = strongly agree, the school mean of teachers’ reports for 4 items: principal communicates expectations; administration is supportive; principal enforces rules for student discipline; principal communicates objectives; staff are recognized for job well done. School Resources: on a scale of 1 = strongly disagree to 4 = strongly agree, the school mean of teachers’ reports for one item: necessary materials such as textbooks, supplies and copy machines are available as needed by the staff. School-wide Faculty Influence: on a scale of 1 = none to 4 = a great deal, the school mean of collective faculty influence over 7 areas: student performance standards; curriculum; content of in-service programs; evaluating teachers; hiring teachers; school discipline policy; deciding spending of budget. Classroom Teacher Autonomy: on a scale of 1 = none to 4 = a great deal, the school mean of individual teacher’s control over 6 areas: selecting textbooks and other instructional materials; selecting content, topics and skills to be taught; selecting teaching techniques; evaluating and grading students; determining the amount of homework to be assigned; disciplining students. Student-Discipline-Focused Professional Development: on a scale of 1 = not receive or not useful to 4 = very useful, the school mean of teachers’ reports of the usefulness of any professional development activities that focused on student discipline and management in the classroom. Subject-Content-Focused Professional Development: on a scale of 1 = not receive or not useful to 4 = very useful, the school mean of teachers’ reports of the usefulness of any professional development activities that focused on the content of the subjects they taught. Note: We used factor analysis (with varimax rotation method) to evaluate our indices of student discipline problems, school leadership, faculty influence and teacher autonomy. We considered item loadings of at least .4 necessary for inclusion in a factor. No items loaded on more than one factor. Each factor had high internal consistency (a > .7). The measures of student discipline problems, leadership, resources, faculty influence, teacher autonomy and professional development are all school means of the reports of the total SASS teacher sample for each school and not limited to the reports of those in the smaller TFS sample. Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 10 Table 2 Descriptive Statistics for Independent Variables Utilized in Regression Analysis Proportion Categorical Predictor Variables All Teachers Non-minority Minority Teacher Characteristics Young .17 .16 .17 Old .30 .31 .25 Male .25 .25 .24 Minority .17 School Characteristics Rural .19 .21 .11 Suburban .52 .55 .39 Secondary .30 .30 .28 Mean (Std. Dev.) Continuous Predictor Variables All Teachers Non-minority Minority School Characteristics School Size (in 100s) 8.04 (6.07) 7.87 (5.94) 8.9 (6.59) Poverty Enrollment (in 10s) 4.12 (2.93) 3.7 (2.73) 6.22 (2.99) Minority Enrollment (in 10s) 4.12 (2.93) 3.54 (3.3) 7.49 (2.98) Minority Faculty 4.12 (2.93) 10.1 (6.04) 42.83 (30.62 Organizational Characteristics/Conditions Highest Salary (in 10,000s) 6.08 (1.30) 6.06 (1.33) 6.20 (1.18) Student Discipline Problems (scale 1-5) 2.29 (0.71) 2.28 (0.7) 2.35 (0.73) School Leadership Support (scale 1-4) 3.32 (0.65) 3.32 (0.65) 3.33 (0.67) School Resources (scale 1-4) 3.14 (0.89) 3.17 (0.87) 3.0 (0.95) Faculty Influence (scale 1-4) 2.21 (0.61) 2.21 (0.6) 2.24 (0.68) Teacher Autonomy (scale 1-4) 3.38 (0.52) 3.40 (0.51) 3.3 (0.54) Discipline-Focused Prof. Dev. (scale 1-4) 1.77 (1.04) 1.73 (1.01) 1.97 (1.14) Content-Focused Prof. Dev. (scale 1-4) 2.64 (1.03) 2.60 (1.02) 2.79 (1.04) Note: Means and deviations are at the teacher level and associated with teachers in the sample. demographic characteristics, we also include measures of whether the school is urban, rural, or suburban, the proportion of each school’s student population at or below the poverty level (i.e., eligible for free or reduced-price lunch), the proportion of each school’s student population that is minority, and the proportion of the school faculty that is minority. Because these demographic factors are often highly intercorrelated and confounded, we estimate their effects both separately and simultaneously in conjunction to discern the extent to which they are independent or redundant. Recruitment, Employment, Retention and the Minority Teacher Shortage 11 intercorrelated and confounded, we estimate their effects both separately and simultaneously in conjunction to discern the extent to which they are independent or redundant. Finally, after controlling for the above teacher and school factors, we focus on the relationship to turnover of eight key aspects of the organizational character and conditions in schools: teacher salary, student discipline problems, school leadership and support, school resources, faculty schoolwide decision- making influence, teacher classroom autonomy, teacher professional development activities focused on student discipline and classroom management, and professional development activities focused on the teacher’s subject-area content. This study does not attempt to provide a comprehensive analysis of all the many aspects of schools that may impact the turnover of minority teachers. We focus on this set of eight particular characteristics of schools because they have long been considered among the important aspects of effective school organization (see, e.g., Bryk al., 1990; Chubb & Moe, 1990; Coleman et & Hoffer, 1987; Goodlad, 1984; Grant, 1988), are ostensibly policy amenable, and are available from our data source. The second stage of the analysis examines whether the likelihood that individual teachers will move from or leave their teaching jobs is related to the above measures of school organizational characteristics and conditions, while controlling for individual-level characteristics of teachers and school-level characteristics. Because different school organizational conditions are often interrelated, and their relationship to turnover is possibly confounded, we estimate the coefficients for each measure of school organizational conditions both in separate models and simultaneously. Unlike most empirical analyses, which use either individual teacher’s salaries or the school’s mean teacher salary, we use the normal yearly base salary for teachers at the highest step on the district or school salary schedule because it better assesses differences in the organizational-level compensation structure.3 Our measures of organizational conditions, other than salaries, are based 3 Especially with an aging teaching workforce, it can be unclear if differences in average salary levels are due to real differences in the compensation offered to comparable teachers at different schools or are due to differences in the experience and education levels of the teachers employed. That is, a school with older teachers may appear to offer better salaries, when, in fact, it does not. A more accurate method of comparison across schools is to compare the normal salaries paid by schools to teachers at common points in their careers. Teacher salary levels are often standardized by school districts according to a uniform salary schedule, based on the teachers’ education levels and years of experience. In this analysis, we tested two salary schedule measures—each based on a different point on school salary schedules: 1) the normal yearly base salary for a teacher with 10 years of experience and a master’s degree; and 2) the normal yearly base salary for a teacher at the highest possible step on salary schedule. The latter measure had a slightly stronger association with turnover than the former, and had relatively fewer missing data; hence, it is used in this study. This measure represents the organizational financial rewards teachers can look forward to at an advanced point in their careers if they stay at their schools, which we expect could affect their decisions to depart or stay. This measure also may have limitations. Some might argue that school salary schedules do not accurately capture the effect of salary on rates of teacher turnover because candidates can obtain this information in deciding whether to accept a particular teaching job. From this viewpoint, since public school teachers are compensated according to published salary schedules that change only infrequently, new entrants can predict with almost complete certainty how much they will be paid in each year in the future. Hence, if a teacher did accept a job, it suggests that they are satisfied with their school’s salary levels and, consequently, most likely low salaries would not be a factor in future turnover. On the other hand, sometimes teachers may, of course, accept jobs with salaries below what they would prefer and then move in a few years when a better paying job opens up. Goodlad (1984) and others have argued that, while money is not a major factor in teachers’ choice of a job, it is a major factor in their decision to move or leave teaching. In this view, beginning teachers are primarily motivated by nonpecuniary and intrinsic values, but if these kinds of expectations are frustrated, salaries can become a source of considerable dissatisfaction. Hence, from this viewpoint, salary schedules would be related to turnover Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 12 on teachers’ self-reports. Teachers’ responses in any individual school, of course, may vary because teachers in the same building may perceive various conditions differently. In background analyses, we partitioned the variance of each measure of organizational conditions into within-school and between-school components. The intraclass correlation, or the portion of the variation that lies between schools, ranged from 13% for subject-area professional development to 43% for student discipline, indicating that part of each measure is unique to each teacher-respondent and that part is common to all teachers within a school. Elsewhere, we have explicitly compared the relative association with turnover of these two levels of measures of organizational conditions (Ingersoll & May 2012). Our focus here is on whether particular schoolwide organizational conditions on average are related to minority and nonminority turnover. As defined in Table 1, other than salary, for our measures of organizational conditions, we calculate averages across the entire sample of teachers in each school. Our analysis used PROC GENMOD in SAS (version 9.2) because it adjusts for the non- random clustering of teachers within schools resulting from the multilevel structure of the sample and uses within-school and between-school predictor variables to estimate separate effects across multiple levels. This procedure also supports logistic regression and allows for inclusion of sampling design weights. Weights are necessary because the SASS and TFS over- or under-sample certain segments of the teaching population. Though the TFS data are longitudinal in that the turnover outcomes transpired a year after the collection of the SASS measures of school characteristics and organizational conditions, any relationships found between these variables and turnover represent statistical associations between measures and do not imply causality. Stage 3 In the third stage of our analysis, to address our final research question, we undertook several analyses of all seven cycles of the TFS data to document the magnitude of minority teacher attrition and to illustrate its role in the minority teacher staffing problems of the school system. To assess the latter, we undertook simulation analyses to estimate the growth in the minority teaching force that could have occurred over the past two and a half decades under alternative hypothetical scenarios where the rates of minority teacher attrition had been lower. Results Has the Number of Minority Teachers Changed? The data show that minority teachers continue to represent a small portion of the teaching force and that a gap persists between the percentage of minority students and the percentage of minority teachers in the U.S. school system. For instance, in the 2011–12 school year, 44% of all elementary and secondary students were minority, and only 17.3% of all elementary and secondary teachers were minority (see Table 3). This student-teacher gap also exists for each of the major minority subgroups, as illustrated in Table 4. For example, in 2011–12, while 21% of elementary and precisely because they allow teachers to predict how much they will be paid in the future. This analysis does not presume the validity of either view but simply tests whether differences in advanced salaries among schools are related to turnover. Recruitment, Employment, Retention and the Minority Teacher Shortage 13 Table 3 Trends in the Nation’s Population, K-12 Student Enrollment, and Teaching Force, by Race/Ethnicity (1987-2012) 1987-88 School Year 1990-91 School Year 1993-94 School Year 1999-00 School Year 2003-04 School Year 2007-08 School Year 2011-12 School Year % Increase 1987-88 to 2011-12 Population of U.S. 244,499,000 252,153,000 260,327,000 281,422,000 292,805,000 304,060,000 313,914,000 28 Number of Minorities 56,479,000 61,273,000 66,644,000 79,080,000 93,991,000 104,597,000 116,148,000 106 % Minority population 23.1 24.3 25.6 28.1 32.1 34.4 37.0 Bachelor’s Degree or Higher Data Unavailable Data Unavailable 36,544,000 44,845,000 51,748,000 57,787,000 60,046,000 Number Minority degree holders 5,590,000 8,139,000 10,754,000 13,107,000 15,037,000 % Minority degree holders 15.3 18.1 20.8 22.7 24.7 Total K-12 Student Enrollment 45,220,953 44,777,577 46,592,207 50,629,075 52,375,110 53,644,872 53,988,330 19 Number Non-minority Students 31,641,098 31,213,142 31,895,394 32,700,441 32,419,640 31,864,127 30,164,827 -5 Number Minority Students 12,335,372 13,564,435 14,696,813 17,928,634 19,955,470 21,780,745 23,825,612 93 % Minority Students 27.3 30.3 31.5 35.4 38.1 40.6 44.1 Total K-12 Teaching Force 2,630,335 2,915,774 2,939,659 3,451,316 3,717,998 3,894,065 3,850,058 46 Number Non-minority Teachers 2,303,094 2,542,720 2,564,416 2,933,591 3,113,249 3,252,234 3,183,837 38 Number Minority Teachers 327,241 373,054 375,243 517,725 604,749 641,830 666,221 104 % Minority Teachers 12.4 12.8 12.8 15.0 16.3 16.5 17.3 Recruitment, Employment, Retention and the Minority Teacher Shortage 14 Table 4 Percentage Students and Teachers, by Race-ethnicity (2011-2012) Non- minority Total Minority Total Black Hispanic Asian Nat. Amer. Multiple Races Students 55.9 44.1 14.4 21.2 5.1 1.2 2.3 Teachers 82.7 17.3 6.4 7.5 1.9 .4 1 secondary students in the US were Hispanic, only 7.5% of teachers were Hispanic. To provide context, in the top half of Table 3, we also include data on the racial/ethnic composition of the national population and of the portion of the nation’s population (age 25 or older) with a bachelor’s. degree or higher (U.S. Census Bureau, 2012). These data indicate that in 2011–12, 37% of the nation’s population were minorities, and 25% of the college-educated were minorities But the data also show that this student-teacher parity gap is not due to a failure to recruit minority teachers. The gap has persisted in recent years largely because the number of nonminority students has decreased, while the number of minority students has increased – leading to an increase in the proportion of all students that are minority. After a period of decline during the 1970s, elementary and secondary student enrollments began to grow steadily in the US, beginning in the mid-1980s and continuing. As Table 3 shows, over the two and a half decades between the 1987-88 and 2011-12 school years, the elementary and secondary student population as a whole increased by 19%. But this varied by the race/ethnicity of students. While the number of nonminority students decreased by 5% during those decades, the number of minority students increased by 93%. The teaching force, as a whole, also increased over this same two and a half decade period— strikingly, by 46%—a rate over two times that of the overall growth rate for students of 19%. Elsewhere, we present a closer examination of the reasons behind this relatively dramatic growth in the teaching force (Ingersoll, Merrill, Stuckey & Collins, 2018); our focus here is on the increase of teachers by their race/ethnicity. From the late 1980s to 2012, the number of minority teachers more than doubled from about 325,000 to 666,000. While the number of nonminority teachers increased by 38%, the number of minority teachers increased by 104% (see Figure 1)—at about the same rate as the growth in the nation’s minority population (Table 3). Even as the size of the teaching force grew, the proportion of the teaching force that is minority increased steadily—from 12 to 17% (bottom row of Table 3). Hence, the data show that, while there is still not parity between the proportions of minority students and minority teachers in schools, the U.S. teaching force has grown more diverse by race/ethnicity since the late 1980s. There have also been some interesting differences in teacher race/ethnicity by teacher gender. Teaching has long been a predominantly female occupation and, in recent decades, it has become increasingly so (Ingersoll, Merrill, Stuckey & Collins, 2018). But this varies by race/ethnicity. Over the two-and-a-half-decade period, from 1987 to 2012, the number of nonminority male teachers increased by only 12%, but the number of minority male teachers increased by 109%. In 2011–12, males represented about 24% of all nonminority teachers and about 25% of all minority teachers. The overall growth from 1987 to 2012 in the number of minority teachers also greatly varied across different minority subgroups and across different time periods. This is shown in Figures 2 and 3, which disaggregate the data in Figure 1, by both racial/ethnic group and by time period. During the 20-year period, from 1987 to 2008, both the overall number of teachers and students Recruitment, Employment, Retention and the Minority Teacher Shortage 15 increased. Moreover, with one exception, growth in minority teachers outpaced growth in minority students (see Figure 2). While the number of nonminority teachers increased by 41%, the number of Hispanic teachers increased by 245% and Asian teachers by 148%. Black teachers also grew in number, but at a far slower rate. The one exception to this growth was Native American teachers, who declined in number by 30%. Native Americans comprise only 1% of students and less than 0.5% of the teaching force. 19 -5 93 46 38 104 -50 -30 -10 10 30 50 70 90 110 All White, non-Hispanic Minority Percent Students Teachers Figure 1. Percentage Change in Students and Teachers, by Race/ethnicity, 1987-88 to 2011-12. Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 16 -30 148 245 31 97 41 48 62 113 159 22 77 -1 19 -100 -50 0 50 100 150 200 250 300 Nat. Amer. Asian Hispanic Black MINORITY WHITE, NON-HISPANIC ALL Percent Students Teachers Figure 2. Percentage Change in Students and Teachers, by Race/ethnicity, from 1987-88 to 2007-08 Figure 3. Percentage Change in Students and Teachers, by Race/ethnicity, from 2007-08 to 2011-12 Recruitment, Employment, Retention and the Minority Teacher Shortage 17 This overall pattern subdivided after 2008, when the economic downturn and recession began. Figure 3 shows trends for the period from 2008 to 2012. During that period, there was a decline in the numbers of non-minorities, Blacks, and Native Americans for both teachers and students, while the number of Hispanic and Asian teachers and students both continued to increase. Where Are Minority Teachers Employed? While there has been a dramatic increase in minority teachers, this growth has not been equally distributed across different types of schools. As shown in Table 5, in 2011–12, 92% of minority teachers were employed in public schools. Moreover, of those employed in public schools, minority teachers were overwhelmingly working in high-poverty, high-minority, urban communities.4 For example, almost two-thirds of minority teachers worked in schools serving predominantly minority students. A similar proportion was employed in high-poverty schools. Table 5 Of Minority and Nonminority Teachers, Percentage Employed in Different Types of Schools (2011-2012) School Type Non- minority Minority Total Black Hispanic Asian Nat. Amer. Multiple Races Public 87.1 91.9 93.2 91.5 87.5 97.6 91.5 Urban 25 45 50 44 49 19 33 Suburban 33 29 27 32 28 20 38 High Poverty 31 62 68 63 52 59 42 Low Poverty 23 11 8 11 18 7 21 High Minority 21 64 67 67 59 43 41 Low Minority 21 3 1 3 5 5 8 Private 12.9 8.1 6.8 8.5 12.5 2.4 8.5 Note: High-poverty schools are those in which 60% or more of the students are eligible for the federal free or reduced- price lunch program for students from families below poverty level. Low-poverty schools are those in which less than 20% of the students are eligible for the federal free or reduced-price lunch program. High-minority schools are those in which 75% or more of the students are minority. Low-minority schools are those in which less than 10% of the students are minority. Minority teachers were two to three times more likely than nonminority teachers to work in such schools. In contrast, only 3% of minority teachers were in low-minority schools (those in which less than a 10th of the students were minority). Elsewhere, we have examined trends over 4 To illustrate the public sector distribution for 2011–12, we subdivided the public teaching force into quartiles according to the poverty and minority student enrollments of their schools. In Table 5 and Figure 4, high-poverty schools refer to those in which 60% or more of the students are eligible for the federal free or reduced-price lunch program for students from families below poverty level. Low-poverty schools refer to those in which less than 20 % of the students are eligible for the federal free or reduced-price lunch program. High-minority schools refer to those in which 75% or more of the students are minority. Low-minority schools refer to those in which less than 10% of the students are minority. Note: These categories represent quartiles of the total SASS sample of public school teachers; these categories are not of equal size in number of schools or students. Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 18 recent decades in the employment of minority teachers across different types of schools and have documented the persistence of this uneven distribution of teachers, by race ethnicity. For instance, during the two-and-half-decade period between 1987 and 2012, the number of minority teachers in higher-poverty schools increased by 288%; the increase in the number of minority teachers in lower- poverty schools was only 1% for the same period (Ingersoll & Merrill, 2017). It is also important to recognize that since minority teachers represented only 17.3% of the teaching force in 2011–12, in the same types of schools where minority teachers were disproportionately employed, the teaching staff overall was nevertheless predominantly nonminority. Figure 4 illustrates this continuing lack of demographic parity. For instance, in high-minority public schools (i.e., those with 75% or more minority students), only 40% of teachers were minority. Likewise, in high-poverty public schools, only 31% of teachers were minority. Figure 4. Of Different Types of Public Schools, Race/ethnicity of Teaching Staffs (2011-2012. Recruitment, Employment, Retention and the Minority Teacher Shortage 19 How High is Minority Teacher Turnover? In the two and a half decades from the late 1980s to 2013, despite some fluctuations, the annual rate of teacher turnover increased overall. Moreover, during this period, the data also indicate that minority teachers tended to have higher rates of turnover than nonminority teachers. Table 6 presents turnover, attrition, and migration data for teachers, by race/ethnicity.5 As illustrated, for five of the seven cycles of the TFS data, total turnover rates for minorities were higher than those for nonminority teachers, at a statistically significant level. In none of the cycles were minority turnover rates lower than those of nonminority teachers at a statistically significant level. Moreover, this gap appears to have widened in the last decade. In the 2004–2005, 2008–09, and 2012–13 school years, minority turnover was, respectively, 18%, 24%, and 25% higher than nonminority teacher turnover. This gap also appears to hold for each of the major minority subgroups, but given smaller sample sizes, such data must be interpreted with caution. For instance, the 2008–09 TFS data suggest that Blacks, Hispanics, Asians, and Native American teachers each had higher rates of turnover than did nonminority teachers. Table 6 Percentage Annual Public and Private Teacher Migration and Attrition, by Race/ethnicity of Teachers, and by Year Minority Teachers Non-Minority Teachers Year Moves Leaves Total Moves Leaves Total 1988-89 9.2 5.9 15.1 7.9 6.5 14.4 1991-92 7.0 6.1 13.1 7.2 6.0 13.2 1994-95 9.2 7.6 16.8 6.7 7.2 13.9 2000-01 8.4 7.5 15.9 7.7 8.2 15.9 2004-05 9.0 10.4 19.4 7.6 8.8 16.4 2008-09 10.1 9.2 19.3 6.7 8.9 15.6 2012-13 (public only) 10.6 8.3 18.9 7.5 7.6 15.1 What Are the Sources of Minority Teacher Turnover? Self-report data. These data raise an important question: What are the reasons for, and sources of, these levels and patterns of nonminority and minority teacher turnover? One way to answer this question is to examine self-report data from those who departed. Figures 5 and 6 present data on the percentage of teachers in the TFS who reported that particular reasons were “very” or “extremely” important in their decisions to move or leave, on a five-point scale from “not important” to “extremely important.” We grouped the individual reasons into categories as shown. Note that the percentages in the tables add up to more than 100% because respondents could indicate more than one reason for their departures. We focus here on public schools. 5 Unlike in earlier data cycles, the 2012–13 TFS did not include turnover data for teachers in private schools. In Table 6, the apparent decrease in minority and nonminority turnover rates between the 2008–09 and 2012–13 TFS cycles is due to the omission of private school teachers in the latter. Turnover rates are, on average, higher in private schools. Our examination of these trends in turnover for only public schools shows that minority teacher turnover in public schools increased slightly during this period. Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 20 Figure 5. Percentage Minority Public School Teachers Reporting General Types of Reasons for their Turnover, 2012-13. Retirement is not an especially prominent factor (see Figure 5). The latter was reported by only 17% of those who departed. At 25%, school staffing cutbacks due to layoffs, terminations, school closings, and reorganizations account for a larger proportion of turnover than does retirement. These staffing actions result in migration to other teaching jobs more often than leaving the teaching occupation altogether. Of those who depart because of job dissatisfaction, most link their turnover to the way their school is administered, to how student assessments and school accountability affected teaching, to student discipline problems, and to a lack of input into decisions and lack of classroom autonomy over their teaching (see Figure 6). The data (not shown here) also show that nonminority teachers report similar reasons behind their turnover, and, in general, similar kinds of dissatisfactions underlie both teacher migration and teacher attrition. Recruitment, Employment, Retention and the Minority Teacher Shortage 21 Figure 6. Of Those Minority Public School Teachers Reporting Dissatisfaction, Percentage Reporting Particular Reasons for Their Turnover, 2012-13. In sum, the data indicate that minority teachers depart their jobs for a variety of reasons. Retirement accounts for a relatively small number of total departures. Some departures are due to school staffing actions; a large proportion of departures is for personal reasons; and another large proportion is for job dissatisfaction or to seek better jobs or other career opportunities. These findings are important because of their policy implications. Unlike explanations that focus on external demographic trends, these findings suggest there is a role for the internal organization and management of schools in minority teacher staffing problems. But, as discussed in the Data, Measures, and Methods section, there are limitations to these self-report data. We follow up below with a multivariate analysis examining the relationship between turnover and a specific set of school organizational characteristics and conditions, based on data from the full set of respondents in SASS, while controlling for other factors, such as teacher age, gender, school grade level, school size, and the demographic characteristics of schools. Individual, school, and organizational predictors of minority teacher turnover. We estimated a series of regression models using the SASS/TFS data to examine whether our predictor variables (in Tables 1 and 2) were associated with teacher turnover. The predictor variables and associated regression estimates from each model are shown in Tables 7a and 7b. To evaluate whether relationships between the predictors and turnover differed by the teachers’ race/ethnicity, we separately estimated our models for minority teachers and for nonminority teachers; these are displayed side by side in the tables. Again, we focus here on public schools. Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 22 In Table 7a, we sequentially entered the sets of measures for teacher characteristics and school characteristics, then added the school demographic measures separately and, finally, included all the measures in a full model. In Table 7b, we sequentially added each of the organizational condition variables to a basic model that included the set of teacher characteristics, the set of school characteristics, and the set of school demographic measures. Tables 7a and 7b present the maximum likelihood estimates for logistic regression, along with significance tests for individual parameters. To make the results more understandable to the reader, in the text we transformed these estimates into odds ratios. The odds ratios are measures indicating the odds of teachers departing (i.e., the ratio of the probability of staying or departing) for particular types of teachers (e.g., male teachers) or for particular types of schools (e.g., small schools). As shown in Model 1 of Table 7a, our analyses found that individual demographic characteristics of teachers were related to their likelihood of staying or departing at a statistically significant level, after controlling for other factors. But this differed by the race/ethnicity of the teachers. The age of teachers was a salient predictor of the likelihood of turnover, but only for nonminority teachers. Both younger (less than 30) and older (greater than 50) nonminority teachers were far more likely to depart than were middle-aged nonminority teachers. For instance, the relative odds of young nonminority teachers departing were more than two times higher than for middle-aged nonminority teachers. In contrast, younger and older minority teachers did not depart at higher rates than other minority teachers (Model 1). Gender was also a factor, but only for minority teachers. The odds of male minority teachers departing were over 50% higher than for female minority teachers; for non-minorities, there was little or no gender difference. Some school characteristics were also related to turnover, but again this differed by the race/ethnicity of the teachers. Minority teachers in smaller schools departed at higher rates; an enrollment difference of 100 students was associated with a 3% difference in the odds of minority teachers departing. For nonminority teachers, the relationship with school size was very small and not statistically significant. As shown in Models 2, 3, 4, and 5, schools in urban areas, schools with higher percentages of low-income students, schools with higher percentages of minority students, and schools with higher percentages of minority teachers each had higher nonminority turnover. For instance, a 10- percentage-point increase in the proportion of poverty-level students was associated with a 6% increase in the odds of nonminority teachers departing. In contrast, there was no consistent or statistically significant relationship between the likelihood of minority teachers departing and these demographic characteristics of schools (with the exception of lower minority turnover in rural compared to urban schools). In other words, after controlling for the background characteristics of teachers and schools, minority teachers, on average, did not depart at significantly different rates from schools with different poverty levels, with different minority student levels, or with different proportions of minority faculty. Recruitment, Employment, Retention and the Minority Teacher Shortage 23 Table 7a: Logistic Regression Analysis of the Likelihood of Minority and Non-minority Teacher Turnover Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Minority Non- minority Minority Non- minority Minority Non- minority Minority Non- minority Minority Non- minority Minority Non- minority Teacher N 6,766 36,378 6,766 36,378 6,181 34,014 6,753 36,346 6,181 34,014 6,181 34,014 School N 3,304 8,223 3,304 8,223 2,977 7,549 3,294 8,213 2,977 7,549 2,977 7,549 Intercept -1.70*** -2.10*** -1.68*** -1.94*** -1.69*** -2.11*** -1.72*** -2.10*** -1.80*** -2.08*** -1.77*** -2.09*** Teacher Characteristics Younger 0.08 0.84*** 0.07 0.83*** 0.05 0.84*** 0.08 0.82*** 0.05 0.84*** 0.04 0.82*** Older -0.08 0.34*** -0.08 0.33*** -0.06 0.34*** -0.08 0.33*** -0.06 0.33*** -0.07 0.33*** Male 0.41** 0.10 0.41** 0.10~ 0.48** 0.10 0.41** 0.10~ 0.49** 0.10 0.48** 0.10 School Characteristics Secondary Level 0.06 -0.10 0.09 -0.07 0.02 -0.02 0.06 -0.04 0.07 -0.05 0.05 0.00 School Size (in 100s) -0.03** 0.00 -0.04** 0.00 -0.04** 0.00 -0.03** -0.01 -0.04** -0.01 -0.04** -0.01 School Demographics Rural -0.33~ -0.26*** -0.20 -0.01 Suburban 0.05 -0.20** 0.11 0.03 Poverty Enrollment (in 10s) -0.02 0.06*** -0.03 0.01 Minority Enrollment (in 10s) 0.01 0.05*** -0.02 0.05** Minority Faculty (in 10s) 0.02 0.11*** 0.05 0.04~ Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 24 Table 7b: Logistic Regression Analysis of the Likelihood of Minority and Non-minority Teacher Turnover Model 7 Model 8 Model 9 Model 10 Model 11 Model 12 Minority Non- minority Minority Non- minority Minority Non- minority Minority Non- minority Minority Non- minority Minority Non- minority Teacher N 5,418 28,957 6,181 34,014 6,181 34,014 6,181 34,014 6,181 34,014 6,181 34,014 School N 2,558 6,408 2,977 7,549 2,977 7,549 2,977 7,549 2,977 7,549 2,977 7,549 Intercept -1.99*** -2.07*** -1.76*** -2.07*** -1.77*** -2.08*** -1.77*** -2.09*** -1.79*** -2.09*** -1.88*** -2.15*** Teacher Characteristics Younger 0.10 0.88*** 0.03 0.82*** 0.03 0.82*** 0.04 0.82*** 0.03 0.82*** 0.04 0.82*** Older 0.05 0.36*** -0.08 0.33*** -0.07 0.33*** -0.07 0.33*** -0.10 0.32*** -0.08 0.32*** Male 0.50** 0.12~ 0.47** 0.08 0.48** 0.09 0.48** 0.1 0.49*** 0.10 0.49*** 0.11~ School Characteristics Secondary Level 0.02 -0.02 0.00 -0.07 0.04 -0.02 0.05 -0.01 0.02 0.00 0.11 0.06 School Size (in 100s) -0.04* -0.01 -0.04** -0.01* -0.04** -0.01 -0.04** -0.01 -0.04** -0.01 -0.04** -0.01~ School Demographics Rural -0.09 -0.07 -0.17 0.02 -0.18 -0.02 -0.21 0.00 -0.16 -0.01 -0.08 0.05 Suburban 0.29~ 0.01 0.13 0.05 0.12 0.03 0.10 0.03 0.14 0.03 0.15 0.04 Poverty Enrollment (in 10s) -0.02 0.00 -0.03 -0.01 -0.03 0.01 -0.03 0.01 -0.03 0.01 -0.04 0.01 Minority Enrollment (in 10s) 0.00 0.06** -0.02 0.05** -0.02 0.05** -0.01 0.05** -0.02 0.05** -0.02 0.05** Minority Faculty (in 10s) 0.04 0.04 0.05 0.04~ 0.05 0.04~ 0.05 0.04~ 0.05 0.04~ 0.05 0.04~ Organizational Conditions Highest Salary (in $10,000s) -0.06 -0.06~ Student Discipline Problems 0.13 0.27*** School Leadership Support -0.11 -0.25*** School Resources 0.06 -0.10 Faculty Influence -0.47** -0.19* Teacher Autonomy -0.51* -0.38*** Note. ~p<.10, *p<.05, **p<.01, ***p<.001 Recruitment, Employment, Retention and the Minority Teacher Shortage 25 Model 6 includes all of these predictors simultaneously; it examines whether the effects of the different measures of school demographics were independent or redundant. Interestingly, after controlling for the other school demographic characteristics, the student poverty enrollment of schools was no longer significantly related to nonminority teacher turnover; minority enrollment and minority faculty remain related, but with only borderline statistical significance. This suggests that the percentage of poverty-level students in schools is not independently related to nonminority teachers’ likelihood of departing, once the percentage of minority students in schools is held equal. On the other hand, this also suggests that the percentage of minority students in schools is significantly and independently related to nonminority teacher turnover, even after holding school poverty levels constant. In contrast, for minority teachers, as in the other models, none of the demographic characteristics of schools was significantly related to turnover. After controlling for these demographic characteristics of teachers and schools, were the organizational conditions of schools associated with turnover? In each of the models shown in Table 7b, the introduction of the organizational variable improved the model likelihood statistic by a statistically significant amount; moreover, after controlling for the background characteristics of teachers and schools, most of the measured conditions were significantly associated with turnover. But, again, this depended on the race/ethnicity of the teacher. The measure for top salaries (the highest annual salary in the school district’s teacher salary scale) had a statistically significant negative bivariate relationship with turnover before controlling for school characteristics; not surprisingly, higher salaries were associated with lower turnover. However, once other background factors were held constant, as shown in Model 7b, the coefficient for highest salaries was of only borderline statistical significance for nonminority teachers. The coefficient for minority teachers was the same magnitude (-.06) as for nonminority teachers but was not statistically significant. The SASS data indicate that in 2003–04, the average starting salary in public schools for a teacher with a bachelor’s degree and no experience was about $32,000, and the average maximum salary (the measure used here) was about $61,000. As shown in Model 8, in schools with higher levels of student discipline problems, turnover rates were distinctly higher for nonminority teachers; the relationship was in the same direction for minority teachers but not at a statistically significant level. The former is one of the stronger relationships we found. A one-unit increase in average reported student discipline problems between two schools (on a five-unit scale) was associated with a 31% increase in the odds of a non-minority teacher departing. As shown in Model 9, in schools that provide better principal leadership and administrative support, as reported by teachers, turnover rates for both minority and non-minority teachers were lower. However, again, the relationship with minority teacher turnover was not strong enough to reach a statistical significance. A one-unit difference between schools in average reported leadership support (on a four-unit scale) was associated with a 22% decrease in the odds of a non-minority teacher departing. In schools where teachers reported that necessary materials were available, such as textbooks and supplies, turnover appeared lower for non-minority teachers, but not at a level of statistical significance (Model 10). As shown in Model 11, schools with higher levels of schoolwide faculty decision-making influence had lower levels of turnover for both nonminority and minority teachers. This is one of the strongest relationships we found and especially so for minority teachers. A one-unit increase in reported faculty influence between schools (on a four-unit scale) was associated with a 37% decrease in the odds of a minority teacher departing. As shown in Model 12, schools with higher average levels of individual teachers’ classroom autonomy had lower levels of turnover—and, again, this was especially true for minority teachers. This is the strongest relationship we found. A one-unit difference in reported teacher classroom Recruitment, Employment, Retention and the Minority Teacher Shortage 26 autonomy (on a four-unit scale) was associated with a 40% difference in the odds of a minority teacher departing. We also examined the relationship to turnover of whether teachers participated in, and found useful, two types of professional development: 1) professional development focused on student discipline and classroom management, and 2) professional development focused on the content of the subjects taught. For both types of development, and for both nonminority and minority teachers, the association with turnover was small and not statistically significant and, hence, we did not display them here. Moreover, we estimated our sequential sets of models in Tables 7a and 7b on several permutations of our teacher data file. First, we tested our models (table not displayed here) using a comprehensive data file combining both nonminority and minority teachers, while adding a predictor for minority teachers. Consistent with our descriptive statistics in Figure 5, minority teachers had a statistically significantly higher likelihood of turnover than did nonminority teachers. The odds of minority teachers departing were almost 50% higher than for nonminority teachers, even after controlling for the characteristics of teachers and schools. However, once organizational conditions were controlled, the coefficient for minority teachers became statistically insignificant and small (the odds of minority teachers departing were less than 5% higher than for nonminority teachers). In other words, we found that less positive organizational conditions in schools accounted for the higher rates of minority teacher turnover. Second, we also estimated our same set of models separately for movers and leavers to explore differences in the predictors of each component of total turnover (table not displayed here). This analysis necessarily used the smaller TFS sample. Given the smaller sample, as expected, some of the coefficients for organizational conditions failed to achieve statistical significance. However, in almost all cases, the direction and magnitude of the coefficients for organizational conditions were similar to those found in the models analyzing the full sample in Tables 7a and 7b. In other words, organizational conditions associated with differences in rates of teacher migration were similarly associated with differences in rates of teacher attrition. There were, however, some interesting differences in the relationship of school demographic characteristics to these types of departures. For nonminority teachers, school demographic characteristics, especially minority student enrollment, had a stronger relationship to moving than to leaving. After controlling for our other variables, schools with higher percentages of low-income students, higher percentages of minority students, and higher percentages of minority teachers had higher nonminority teacher migration to other schools; for non-minorities leaving teaching altogether, the differences were small and not statistically significant. In contrast, for minority teachers, there continued to be little relationship between school demographic characteristics and the likelihood that minority teachers would either move between schools or leave teaching, with one interesting exception: There was a small but statistically significant relationship between poverty enrollment and minority teacher migration. Minority teachers were slightly less likely to move from schools with higher poverty-level enrollments. Third, and finally, we estimated our same set of models on a subset of turnover that excluded those who departed because of retirement, layoffs, terminations, or school closings in order to test our findings focusing on only voluntary departures. When looking at departures that are, ostensibly, a matter of choice, we would expect organizational conditions to have a clearer and stronger relationship to turnover. On the other hand, to do this analysis, it was necessary to use the smaller TFS sample; given the smaller sample, we would expect some variables to have a weaker relationship. We found that the magnitude of the association of many of our predictors increased and that the findings were highly consistent with those in the models in Tables 7a and 7b. Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 27 The separate models in Table 7b estimate the independent relationships to turnover of each organizational condition. However, as discussed in the Data, Measures, and Methods section, the above organizational conditions do not exist in isolation; schools with higher levels of one were also likely to have higher levels of others. To get a sense of the joint association with turnover of multiple organizational conditions, we first estimated an overall model that included all eight organizational conditions, along with the background controls. We then utilized the coefficients from this model to predict turnover rates for a range of values of the set of organizational variables. Holding the control variables constant at the sample mean, we set the eight organizational condition variables to values corresponding to the 10th percentile, the 25th percentile, the mean, the 75th percentile, and the 90th percentile for the sample. This allowed us to predict the turnover rates of minority and nonminority teachers for a range of hypothetical schools, beginning with those that have the worst organizational conditions (i.e., at the 10th percentile on each of the eight organizational measures) and concluding with those that have the best organizational conditions (i.e., at the 90th percentile on each of the eight organizational measures). Results from this analysis are depicted in Figure 7 and reveal a clear collective relationship between organizational conditions and turnover. This relationship is remarkably strong for minority teachers, whose predicted annual turnover rates are only 12% in the schools with the best organizational conditions versus nearly 21% in schools with the worst organizational conditions. For nonminority teachers, the relationship is not as strong, ranging from 12% in the best schools to 15% in the worst schools. Figure 7. Predicted Turnover Rates, by Public School Organizational Conditions Percentile. Recruitment, Employment, Retention and the Minority Teacher Shortage 28 What is the Role of Minority Teacher Attrition in the Staffing Problems of Schools and in the Minority Teacher Shortage? It is important to recognize that teacher turnover is not necessarily detrimental. In general, theory and research from the fields of organizational theory, economics, and sociology have long held that some degree of employee turnover is normal and inevitable, and can be efficacious for individuals, for organizations, and for the economic system as a whole (e.g., Abelson & Baysinger, 1984; Hom & Griffeth, 1995 ; Jovanovic, 1979a, 1979b; Mobley, 1982; Price, 1977, 1989; Siebert & Zubanov, 2009). Across a range of occupations and industries, job and career changing are normal and common, perhaps increasingly so, and some hold that high levels of employee turnover are a sign of economic opportunity and a dynamic, well-functioning economy (e.g., Kimmitt, 2007). Moreover, researchers have concluded that effective organizations usually promote some degree of employee turnover and benefit from it by the departure of low-caliber performers and the recruitment of “new blood” to facilitate innovation. However, though there can be benefits to employee turnover, theory and research in these fields have also long held that employee turnover is not cost-free. There is a general consensus that a variety of costs and consequences are associated with employee turnover, including the loss of human capital and of investments in employee development, the cost of replacement hiring and training, and disruption of production processes, and that such costs vary by industry and occupation. In the education sector, from the viewpoint of those managing schools and those seeking to employ more minority teachers in school classrooms, all of these types of departures have the same effect: They reduce the number of minority teachers in the organization. One consequence of attrition, in particular, our analysis reveals, is that it undermines efforts to increase the number of minorities in the teaching force as a whole. As shown in Table 3, between the 1987-88 and 2011-12 school years the minority teaching force grew from about 327,000 to 666,000, a gain of 104%. In 1987–88, minorities represented 12.4% of the teaching force; in 2011–12 minorities represented 17.3%. But, notably, this increase in the minority teaching force occurred in spite of the high attrition rate among minority teachers, as shown in Table 6. For instance, the SASS/TFS data indicate that at the beginning of the 2003–04 school year, about 47,600 minority teachers entered teaching; however, by the following school year, 20% more—about 56,000—had left teaching altogether. Of these, about 18,500 retired, 20,000 indicated that they left to pursue another job or career, and 20,000 indicated that they left because of job dissatisfaction. This raises the question: If minority teacher attrition rates could have been lower in recent decades, what would have been the gain in the total number of minority teachers employed? To answer this question, we undertook simulation analyses designed to predict the growth in the minority teaching force over the past two and a half decades under two alternative hypothetical scenarios, wherein rates of minority teacher attrition were lower.6 We drew from the results in our 6 We simulated racial/ethnic representation in the teaching force by modeling entry to, and exit from, teaching by minority and nonminority teachers for each year from 1987–88 through 2011–12. We projected the number of minority teachers in each year by subtracting from the previous year’s total the number of minority teachers who left teaching and adding the number of minority teachers hired under each of our two alternative attrition rate scenarios. To determine the number of minority attriters, we applied our hypothetical attrition rates to the simulated total number of minority teachers from the previous year. In our first simulation scenario, we applied nonminority attrition rates for each year to minority teachers. We estimated these rates from SASS/TFS for the seven on-cycle years and linearly interpolated for years for which no survey was administered. Under the second simulation scenario, we applied to minority teachers the attrition Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 29 earlier analyses, in Tables 6 and 7b, to choose two examples of lower attrition rates. Figure 8 displays the actual growth of the minority teaching force as estimated in SASS and the hypothetical growth under our two alternative scenarios. In the first scenario, we estimated the growth in the number of minority teachers if the attrition rates for minority teachers had been the same as those for nonminority teachers from 1987- 88 to 2011-12. Recall in Table 6 for most years of the survey, minority attrition rates were lower than those of nonminority teachers. In this scenario, our simulation indicates that, by 2012, the minority teaching force would have grown to 721,000—a gain of 55,000 teachers over the actual levels. Under this scenario, by 2012, minorities would have represented 18.7% of the teaching force, rather than 17.3%. In our second scenario, we estimated the growth in the number of minority teachers if their attrition rates had been equal to those in schools with high levels of teacher classroom autonomy. We chose this factor because, as shown earlier in our regression analyses displayed in Table 7b, the association between teachers’ classroom autonomy and turnover was a relatively strong relationship. In this second scenario, our simulation indicates that, by 2012, the minority teaching force would have grown to 897,000—a gain of 213,000 teachers over the actual levels. Under this scenario, by 2012, minorities would have represented 23% of the teaching force—still far less than the percentage of students that were minority (44% in Table 3), but close to the percentage of the college-educated population that were minority (25% in Table 3). Our second simulation analysis suggests that had the schools in which minority teachers have been working afforded them the classroom autonomy held by teachers employed in schools that were in the top 10th percentile of teacher classroom autonomy, it is conceivable that the US would have had nearly a quarter million more minority teachers by 2012. rate for teachers employed in schools scoring in the top decile of teachers’ classroom autonomy in the 2011– 12 SASS/TFS. In order to project the number of minority teachers hired in each year, we multiplied the actual contemporaneous proportion of new teachers who were minorities by the simulated total number of teachers hired in that year. We estimated the proportions of new teachers who were minorities from SASS/TFS for on-cycle years and linearly interpolated for years for which no survey was administered. To project the number of teachers hired in any year, we assumed that the total size of the labor force would not change between attrition scenarios. That is, the total numbers of teachers in each year of each of our simulations match the historical estimates from SASS. From these annual totals, we calculated the growth (or decline) in the teacher labor force each year. We added to this growth the total number of teachers who would have left the occupation under each attrition scenario each year, as teachers would need to be hired to replace those exiting teaching. To this total number of teachers hired, we applied the proportion of new teachers who were minorities to arrive at the number of minority teachers hired each year. Recruitment, Employment, Retention and the Minority Teacher Shortage 30 - 100,000 200,000 300,000 400,000 500,000 600,000 700,000 800,000 900,000 1,000,000 1987-88 1990-91 1993-94 1996-97 1999-00 2002-03 2005-06 2008-09 2011-12 N u m b e r o f M in o ri ty T e a c h e rs With Attrition Rate in High-Autonomy Schools With Attrition Rate of Non-Minority Teachers Actual Number of Minority Teachers Figure 8. Trends in the Number of Minority Teachers, by Actual and Alternative Attrition Conditions (1987-2012). Summary and Implications It is widely believed that the nation’s schools suffer from dire shortages of minority teachers. Numerous scholars and commentators have held that there is a growing mismatch between the degree of racial/ethnic diversity in the nation’s student population and the degree of diversity in the nation’s elementary and secondary teaching force, and this is detrimental to the growth and learning of students. In response, in recent decades, numerous government and non-government organizations have implemented a variety of minority teacher recruitment programs and initiatives. Has the number of minority teachers grown? The national data show a gap persists between the percentage of minority students and the percentage of minority teachers in the U.S. school system. For instance, in 2012, 37% of the nation’s population was minority, and 44% of all elementary and secondary students were minority, but only 17.3% of all elementary and secondary teachers were minority. But the data also show this gap is not Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 31 due to a failure of teacher recruitment. Indeed, since the late 1980s, the number of minority elementary and secondary teachers has increased by over 100%, outpacing growth in the number of nonminority teachers and outpacing growth in minority students. The result is that the teaching force has rapidly grown more diverse. Moreover, minority teachers are overwhelmingly employed in public schools serving high- poverty, high-minority, and urban communities. Minority teachers are two to three times more likely than nonminority teachers to work in such hard-to-staff schools. Hence, the data suggest that, in spite of any possible barriers to entry, and competition from other occupations for minority college graduates, there has been a large increase in the number of minority teachers, especially in schools serving disadvantaged and minority student populations. While our data analysis does not test this, nor allow us to attribute these gains to particular reforms, this success is likely connected to the ongoing minority recruitment initiatives. However, overall, the data also show that, over the past two and a half decades, minority teachers were more likely to depart from their schools than nonminority teachers. This was especially true for male minority teachers. The result has been that, numerically, there has been a large degree of job transition among minority teachers each year. Some turnover of teachers is, of course, normal, inevitable, and beneficial. For individuals, departures leading to better jobs, either in teaching or not, can be a source of upward mobility. For schools, departures of low-performing employees can enhance organizational outcomes. For the educational system, some teacher outflows, such as cross-school migration, temporary attrition, or those leaving classroom teaching for other education-related jobs, do not represent a net or permanent loss of human capital to the education system as a whole. However, from an organizational level of analysis, and from the viewpoint of those managing schools, none of these types of departures are cost-free, whether permanent, to other schools, or to other education jobs. All have the same effect; they typically result in a decrease in minority classroom instructional staff in that organization. One consequence of attrition, in particular, our analysis reveals, is that it undermines efforts to address the minority teacher shortage. Why do minority teachers depart schools at higher rates? Strikingly, while the demographic characteristics of schools appear to be highly important to minority teachers’ initial employment decisions, this does not appear to be the case for their later decisions about whether to depart. A school’s enrollment of poverty-level students, a school’s minority student enrollment, the school’s proportion of minority teachers, or whether the school lies in an urban or suburban community were not strongly or consistently related to the likelihood that minority teachers would decide to stay or depart. Contrary to the argument that minority teachers have a cultural synchronicity with, and commitment to, minority students (e.g., Irvine, 1988), when it comes to the turnover of minority teachers, there almost seems to be a kind of cultural immunity to the demographic characteristics of the students. Among the most prominent reasons minority teachers gave for leaving or moving were a desire to obtain a better job or career, or dissatisfaction with some aspect of their teaching job. The data further specify that particular school working and organizational conditions were strongly related to minority teacher departures. Hard-to-staff schools that are more likely to employ minority teachers often also have less desirable organizational conditions. And less desirable conditions, our data suggest, account for the higher rates of minority teacher turnover. In other words, the data indicate that minority teachers departed at higher rates because the schools in which they were employed tended to have less positive organizational conditions. The strongest organizational factors for minority teachers were the levels of collective faculty decision-making influence in their school and the degree of individual instructional autonomy held by teachers in their classrooms. Schools that provided more teacher classroom discretion and autonomy, as well as schools with Recruitment, Employment, Retention and the Minority Teacher Shortage 32 higher levels of faculty input into school decision-making influence, had lower levels of minority teacher turnover. This finding is consistent with a long line of our research showing the importance of professional autonomy and teacher “voice” in schools (see, Ingersoll, 1996, 2003, 2012). However, teachers’ classroom autonomy appears to have shrunk in recent years with the implementation of accountability reforms, especially in urban school districts. Some studies have found a growing tension with teachers increasingly held accountable for issues, decisions and outcomes over which they may have little, or even diminishing, control – leading to higher teacher turnover (Guggino & Brint, 2010; Ingersoll & Collins, 2017; Ryan et al., 2017). Organizational accountability and employee authority are not necessarily contradictory imperatives. Leading thinkers in the applied field of organizational leadership have long advocated a balanced approach wherein organizational accountability and employee authority go hand in hand (e.g., Drucker, 1973, 1992). In this view, employees should not be held accountable for things over which they have no control; likewise employees should not be granted control or autonomy without commensurate accountability. The importance of balancing teachers’ responsibilities and authority is borne out in our own research showing that schools in which teachers are both held to high academic standards and allowed substantial input into decision-making have higher teacher retention (Ingersoll & Collins, 2017; Ingersoll, Merrill & May, 2016) and higher student achievement (Ingersoll, Sirinides & Dougherty, 2017). Our present study presents an overall portrait across all public schools and across all minority subgroups. Underlying our study is the assumption that common patterns across schools and across minority subgroups can be informative. But, local and state contexts vary and minority subgroups are, of course, not homogeneous, between or within. Hence, drawing conclusions about the nation as a whole and minority teachers as a whole runs the risk of overgeneralizing. Throughout our study, where sample sizes permit, we disaggregate minority by subgroup. In an earlier related study, Connor (2010) focused specifically on Black teachers, comparing them to nonminority teachers. His findings on turnover were similar to those reported here from our study. But, further research is warranted examining whether the overall patterns we discovered apply across contexts and groups. What are the implications of these results for the widespread policy and reform efforts to diversify the teaching force? In supply and demand theory, any imbalance between labor demand and supply can be referred to as a shortage, in the sense that there is an inadequate quantity of individuals able and willing to offer their services under given wages and conditions. From this perspective, the problems many schools encounter retaining minority teachers can technically be referred to as a shortage. However, in the context of minority teachers and schools, the term shortage is typically given a narrower connotation—an insufficient production and recruitment of new minority teaching candidates in the face of increasing minority student enrollments. These terminological and diagnostic differences have crucial implications for prescription and policy. As noted in the beginning of this report, increased production and recruitment of minority candidates has long been the dominant strategy to diversify the teaching force and address the minority teacher shortage. Numerous high-profile groups have called for dramatic increases in the recruitment of new minority teachers across the nation (e.g., Education Commission of the States, American Association of Colleges of Teacher Education, National Educational Association). Beginning in the late 1980s, such efforts received substantial support and funding—the Ford Foundation and the DeWitt Wallace Readers’ Digest Fund alone committed over $60 million. Nothing in our research suggests that bringing new qualified minority candidates into teaching is not a worthwhile step. But the data indicate that new teacher recruitment strategies alone do not directly address a major source of minority teacher staffing problems: attrition. This is Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 33 especially true for minority teacher recruitment efforts aimed at male teachers, because male minority teachers have especially high departure rates. Indeed, the increase in the number of minority teachers is all the more remarkable because it has occurred in spite of the high attrition rate among minority teachers. Improving the retention of minority teachers brought into teaching by recruitment initiatives could prevent the loss of the investment and help to lessen the ongoing need for more recruitment initiatives. However, nothing in our research suggests that improving minority teacher retention alone will close the parity gap. Our perspective suggests the efficacy of developing teacher recruitment and retention initiatives together in order to solve the minority teacher shortage. A recent report by Albert Shanker Institute (2015) highlighted a number of promising examples of schools that have emphasized both recruitment and retention of minority teachers. Our analyses support the view that school organization, management, and leadership matter, and they shift attention to discovering which policy-amenable aspects of schools as organizations— their practices, policies, characteristics, and conditions—are related to their ability to retain minority teachers. The data suggest that poor, high-minority, urban schools with improved organizational conditions will be far more able to do so. To be sure, the data do not suggest that altering any of the organizational conditions we examined would be easy. However, unlike reforms such as teacher salary increases, professional development, and class-size reduction, changing some conditions, such as, would appear to be less costly financially—an important consideration, especially in low-income settings and in periods of budgetary constraint. The analysis especially draws attention to the importance of teachers’ classroom autonomy and faculty’s schoolwide influence on teacher retention. Promising examples of schools that balance accountability with high levels of teacher autonomy and decision-making influence have sprung up in recent years in the U.S. For example, there is a growing network of schools that are operated and run by teachers (Kolderie, 2008, 2014). These schools are often referred to as “partnership schools” because they are modeled after law partnerships, where lawyers both manage, and ultimately are accountable for, the organization and its success (Hawkins, 2009). In this approach, the focus of reform would shift from solely attracting or developing “better people for the job” to also securing “a better job for the people” (Kolderie, 2008, 2014). Rather than simply forcing the existing arrangement to work better, this alternative perspective suggests the importance of also viewing the roots of shortages as an organizational and occupational design issue, implying the need for a different arrangement, better built for those who do the work of teaching. References Abelson, M. A., & Baysinger, B. D. (1984). Optimal and dysfunctional turnover: Toward an organizational level model. The Academy of Management Review, 9(2), 331–341. https://doi.org/10.5465/amr.1984.4277675 Achinstein, B., & Aguirre, J. (2008). Cultural match or cultural suspect: How new teachers of color negotiate socio-cultural challenges in the classroom. Teachers College Record, 110(8), 1505–1540. Achinstein, B., Ogawa, R. T., Sexton, D., & Freitas, C. (2010). Retaining teachers of color: A pressing problem and a strategy for “hard-to-staff” schools. Review of Educational Research, 80(1), 71–107. https://doi.org/10.3102/0034654309355994 Albert Shanker Institute. 2015. The state of teacher diversity in American education. Albert Shanker Institute. Washington, DC. https://doi.org/10.5465/amr.1984.4277675 https://doi.org/10.3102/0034654309355994 Recruitment, Employment, Retention and the Minority Teacher Shortage 34 Banks, J. (1995). Multicultural education: Historical development, dimensions, and practice. In J. Banks & C. Banks (Eds.), Handbook of research on multicultural education (pp. 3–24). New York, NY: Macmillan. Bristol, T. J. (2018). To be alone or in a group: An exploration into how the school-based experiences differ for black male teachers across one urban school district. Urban Education, 53(3), 334-354. https://doi.org/10.1177/0042085917697200 Bryk, A., Lee, V., & Holland, P. (1993). Catholic schools and the common good. Cambridge, MA: Harvard University Press. Carnegie Forum on Education and the Economy, Task Force on Teaching as a Profession. (1986). A nation prepared: Teachers for the 21st century. Washington, DC: Author. Carver-Thomas, D., & Darling-Hammond, L. (2019). The trouble with teacher turnover: How teacher attrition affects students and schools. Education Policy Analysis Archives, 27(36). http://dx.doi.org/10.14507/epaa.27.3699 Chandler, K., Luekens, M., Lyter, D., & Fox, E. (2004). Teacher attrition and mobility: Results from the Teacher Follow-up Survey (TFS), 2000–01. Washington, DC: National Center for Education Statistics. Chubb, J. E., & Moe, T. (1990). Politics, markets and America’s schools. Washington, DC: Brookings Institute. Clewell, B. C., & Villegas, A. M. (2001). Evaluation of the DeWitt Wallace–Reader’s Digest Fund’s Pathways to Teaching Careers program. Washington, DC: Urban Institute. Cochran-Smith, M. (2004). Walking the road: Race, diversity, and social justice in teacher education. New York, NY: Teachers College Press. Coleman, J., & Hoffer, T. (1987). Public and private schools: The impact of communities. New York, NY: Basic. Connor, R. (2010). Examining African American teacher turnover, past & present. (Unpublished PhD Dissertation). University of Pennsylvania. Dilworth, M. (1992). Diversity in teacher education. San Francisco, CA: Jossey-Bass. Dreeben, R., & Gamoran, A. (1986). Race, instruction and learning. American Sociological Review, 51, 660–669. https://doi.org/10.1177/0042085917697200 Drucker, P. F. (1973). Management: tasks, responsibilities, practices. New York, NY: Harper & Row. Drucker, P. F. (1992). Managing for the future: The 1990s and beyond. New York, NY: Truman Talley. Feistritzer, E. (1997). Alternative teacher certification: A state-by-state analysis (1997). Washington, DC: National Center for Education Information. Foster, M. (1994). Effective Black teachers: A literature review. In E. R. Hollins, J. E. King, & W. C. Hayman (Eds.), Teaching diverse populations: Formulating a knowledge base (pp. 225–241). Albany, NY: State University of New York Press. Foster, M. (1997). Black teachers on teaching. New York, NY: New Press. Fultz, M. 2004. The displacement of Black educators post-Brown: An overview and analysis. History of Education Quarterly, 30(1),11-45. https://doi.org/10.1111/j.1748- 5959.2004.tb00144.x Gandara, P. & Maxwell-Jolley, J. (2000). Preparing teachers for diversity: A dilemma of quality and quantity. Santa Cruz, CA: The Center for the Future of Teaching & Learning. Goodlad, J. (1984). A place called school: Prospects for the future. St. Louis, MO: McGraw-Hill. Grant, G. (1988). The world we created at Hamilton High. Cambridge, MA: Harvard University Press. Grissom, J., & Keiser, L. (2011). A supervisor like me: Race, representation, and the satisfaction and turnover decisions of public sector employees. Journal of Policy Analysis and Management, 30(3), 557-580. https://doi.org/10.1177/0042085917697200 http://dx.doi.org/10.14507/epaa.27.3699 https://doi.org/10.1177/0042085917697200 https://doi.org/10.1111/j.1748-5959.2004.tb00144.x https://doi.org/10.1111/j.1748-5959.2004.tb00144.x Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 35 Guggino, P., & Brint, S. (2010). Does the No Child Left Behind Act help or hinder K-12 education? Policy Matters, 3(3), 1–7. Haberman, M. (1996). Selecting and preparing culturally competent teachers for urban schools. In J. Sikula (Ed.), Handbook of research on teacher education (pp. 747–760). New York, NY: Simon & Schuster Macmillan. Hawkins, B. (2009). Teacher cooperatives: What happens when teachers run the school? Education Next, 9, 37–41. Haycock, K. (2001). Closing the achievement gap. Educational Leadership, 58(6), 6–11. Hirsch, E., Koppich, J., & Knapp, M. (2001). Revisiting what states are doing to improve the quality of teaching: An update on patterns and trends. Seattle, WA: Center for the Study of Teaching and Policy, University of Washington. Hom, P. & Griffeth, R. (1995). Employee turnover. Cincinnati, OH: South-Western. Ingersoll, R. (1996). Teachers’ decision-making power and school conflict. Sociology of Education, 69, 159-176. https://doi.org/10.1177/0042085917697200 Ingersoll, R. (2001). Teacher turnover and teacher shortages: An organizational analysis. American Educational Research Journal, 38(3), 499–534. https://doi.org/10.1177/0042085917697200 Ingersoll, R. M. (2003). Who controls teachers’ work? Power and accountability in America’s schools. Cambridge, MA: Harvard University Press. Ingersoll, R. (2012). Power, accountability and the teacher quality problem. In S. Kelly (Ed.), Assessing teacher quality: Understanding teacher effects on instruction and achievement (pp. 97-109). New York, NY: Teachers’ College Press. Ingersoll, R. M., & Collins, G. (2017). Accountability and control in American schools. Journal of Curriculum Studies, 49(1), 75–95. https://doi.org/10.1080/00220272.2016.1205142 Ingersoll, R., & May, H. (2012). The magnitude, destinations and determinants of mathematics and science teacher turnover. Educational Evaluation and Policy Analysis, 34(4), 435–464. https://doi.org/10.1177/0042085917697200 Ingersoll, R., May, H., & Collins, G. (2017) Minority teacher recruitment, employment and retention: 1987 to 2013. LPI Research Report. Learning Policy Institute, Palo Alto. Ingersoll, R., & Merrill, L. (2017). A quarter century of changes in the elementary and secondary teaching force: From 1987 to 2012. Statistical Analysis Report (NCES 2015-076). Washington, DC: National Center for Education Statistics. Ingersoll, R. M., Merrill, E., & May, H. (2016). Do accountability policies push teachers out? Educational Leadership, 73(8), 44-49. Ingersoll, R., Merrill, E., Stuckey, D. & Collins, G (2018). Seven trends: The transformation of the teaching force, Updated Oct., 2018. CPRE Report (#RR-2018-2). Philadelphia: Consortium for Policy Research in Education, University of Pennsylvania. https://repository.upenn.edu/cpre_researchreports/108 Ingersoll, R., Sirinides, P., & Dougherty, P. (2017). School leadership, teachers’ roles in school decisionmaking, and student achievement. CPRE Working Paper. Philadelphia: Consortium for Policy Research in Education Irvine, J. J. (1988). An analysis of the problem of the disappearing Black educator. Elementary School Journal, 88(5), 503–514. https://doi.org/10.1177/0042085917697200 Irvine, J. J. (1989). Beyond role models: An examination of cultural influences on the pedagogical perspectives of Black teachers. Peabody Journal of Education, 66(4), 51–63. https://doi.org/10.1177/0042085917697200 Jovanovic, B. (1979a). Job matching and the theory of turnover. Journal of Political Economy, 87(5), 972–990. https://doi.org/10.1177/0042085917697200 https://doi.org/10.1177/0042085917697200 https://doi.org/10.1177/0042085917697200 https://doi.org/10.1080/00220272.2016.1205142 https://doi.org/10.1177/0042085917697200 https://doi.org/10.1177/0042085917697200 https://doi.org/10.1177/0042085917697200 https://doi.org/10.1177/0042085917697200 Recruitment, Employment, Retention and the Minority Teacher Shortage 36 Jovanovic, B. (1979b). Firm-specific capital and turnover. Journal of Political Economy, 87(6), 1246– 1260. https://doi.org/10.1177/0042085917697200 Kimmitt. R. (2007, Jan. 23). Why job churn is good. The Washington Post (p. A17). Kirby, S. N., Berends, M., & Naftel, S. (1999). Supply and demand of minority teachers in Texas: Problems and prospects. Educational Evaluation and Policy Analysis, 21(1), 47–66 https://doi.org/10.1177/0042085917697200. Kolderie, T. (2008). The other half of the strategy: Following up on systemic reform by innovating with school and schooling. St Paul, MN: Education Evolving. Kolderie, T. (2014). The split screen strategy: How to turn education into a self-improving system. Edina, MN: Beaver’s Pond Press. Ladson-Billings, G. (1995). Toward a theory of culturally relevant pedagogy. American Educational Research Journal, 32(3), 465–491. https://doi.org/10.1177/0042085917697200 Lau, K. F., Dandy, E. B., & Hoffman, L. (2007). The Pathways Program: A model for increasing the numbers of teachers of color. Teacher Education Quarterly, 34(4), 27–40. Lewis, C. W. (2006). African American male teachers in public schools: An examination of three urban school districts. Teachers College Record, 108(2), 224–245. https://doi.org/10.1177/0042085917697200 Lewis, C. W., & Toldson, I. 2013. Black male teachers: Diversifying the United States’ teacher workforce. Bingley, UK: Emerald Group. https://doi.org/10.1177/0042085917697200 Liu, E., Rosenstein, J., Swann, A., & Khalil, D. (2008). When districts encounter teacher shortages: The challenges of recruiting and retaining math teachers in urban districts. Leadership and Policy in Schools, 7(3), 296–323. https://doi.org/10.1177/0042085917697200 Mobley, W. (1982). Employee turnover: Causes, consequences and control. Reading, MA: Addison-Wesley. Murnane, R., Singer, J., Willet, J., Kemple, J., & Olsen, R. (Eds.). (1991). Who will teach? Policies that matter. Cambridge, MA: Harvard University Press. National Center for Education Statistics. (2005). Schools and Staffing Survey (SASS) and Teacher Follow- up Survey (TFS). Data File. Washington, DC: U.S. Department of Education. Available from http://nces.ed.gov/surveys/SASS. National Collaborative on Diversity in the Teaching Force. (2004). Assessment of diversity in America’s teaching force: A call to action. Washington, DC: National Educational Association. Norton, R. (2005, April). Call me MISTER. Black Collegian, 27–28. Oakes, J. (1985). Keeping track: How schools structure inequality. New Haven, CT: Yale University Press. Oakes, J. (1990). Multiplying inequalities: The effects of race, social class, and tracking on opportunities to learn mathematics and science. Santa Monica, CA: The RAND Corporation. Price, J. (1977). The study of turnover. Ames, IA: Iowa State University Press. Price, J. 1989. The impact of turnover on the organization. Work and Occupations, 16, 461–473. https://doi.org/10.1177/0042085917697200 Quiocho, A. & Rios, F. (2000). The power of their presence: Minority group teachers and schooling. Review of Educational Research, 70(4), 485–528. https://doi.org/10.1177/0042085917697200 Rice, J., Roellke, C., Sparks, D., & Kolbe, T. (2008). Piecing together the teacher policy landscape: A policy-problem typology. Teachers College Record. Available from http://www.tcrecord.org/Content.asp?ContentId=15223. Rogers-Ard, R., Knaus, C. B., Epstein, K. K., & Mayfield, K. (2013). Racial diversity sounds nice; systems transformation? Not so much: Developing urban teachers of color. Urban Education, 48(3), 451–479. https://doi.org/10.1177/0042085912454441 Ronfeldt, M., & McQueen, K. (2017). Does new teacher induction really improve retention? Journal of Teacher Education, 68(4), 394-410. https://doi.org/10.1177/0022487117702583 Rosenbaum, J. (1976). Making inequality. New York, NY: John Wiley & Sons. https://doi.org/10.1177/0042085917697200 https://doi.org/10.1177/0042085917697200 https://doi.org/10.1177/0042085917697200 https://doi.org/10.1177/0042085917697200 https://doi.org/10.1177/0042085917697200 https://doi.org/10.1177/0042085917697200 http://nces.ed.gov/surveys/SASS/ https://doi.org/10.1177/0042085917697200 https://doi.org/10.1177/0042085917697200 http://www.tcrecord.org/Content.asp?ContentId=15223 https://doi.org/10.1177/0042085912454441 https://doi.org/10.1177/0022487117702583 Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 37 Ryan, S., von der Embse, N., Pendergast, L. L., Saeki, E., Segool, N., & Schwing, S. (2017). Leaving the teaching profession: The role of teacher stress and educational accountability policies on turnover intent. Teaching and Teacher Education, 66, 1-11. https://doi.org/10.1016/j.tate.2017.03.016 Shen, J. (1998). Alternative certification, minority teachers, and urban education. Education and Urban Society, 31(1), 30–41. https://doi.org/10.1177/0022487117702583 Siebert, S. W., & Zubanov, N. (2009). Searching for the optimal level of employee turnover: A study of a large U.K. retail organization. Academy of Management Journal, 52(2), 294–313. https://doi.org/10.1177/0022487117702583 Tillman, L. (2004). (Un)intended consequences? The impact of the Brown v. Board of Education decision on the employment status of Black educators. Education and Urban Society, 36(3), 280- 303. https://doi.org/10.1177/0013124504264360 Torres, J., Santos, J., Peck, N. L., & Cortes, L. (2004). Minority teacher recruitment, development, and retention. Providence, RI: Brown University, Educational Alliance. U.S. Bureau of the Census. (2012). The statistical abstract (141st ed.). Washington, DC: U.S. Department of Commerce. Valencia, R. R. (2002). The plight of Chicano students: An overview of schooling conditions and outcomes. In R. R. Valencia (Ed.), Chicano school failure and success, second edition (pp. 3–51). New York, NY: Routledge-Falmer. Villegas, A. M., & Lucas, T. (2004). Diversifying the teacher workforce: A retrospective and prospective analysis. In M. A. Smylie & D. Miretky (Eds.), Developing the teacher workforce (103rd Yearbook of the National Society for the Study of Education, Part 1, pp. 70–104). Chicago, IL: University of Chicago Press. https://doi.org/10.1177/0022487117702583 Villegas, A. M., & Irvine, J. J. (2010). Diversifying the teaching force: An examination of major arguments. Urban Education. 42:175–92. https://doi.org/10.1177/0022487117702583 Villegas, A. M., Strom, K., & Lucas, T. (2012). Closing the racial-ethnic gap Between students of color and their teachers: An elusive goal. Equity and Excellence in Education. 45(2), 283–301. https://doi.org/10.1177/0022487117702583 White, T. (2016). Teach For America’s paradoxical diversity initiative: Race, policy, and Black teacher displacement in urban schools. Education Policy Analysis Archives, 24(16). http://dx.doi.org/10.14507/epaa.24.2100 Wilson, W. J. (1996). When work disappears: The world of the new urban poor. New York, NY: Knopf. Zeichner, K. (1996) Educating teachers for cultural diversity. In K. Zeichner, S. Melnick, & M. L. Gomez (Eds.), Currents of reform in preservice teacher education (pp. 133–175). New York, NY: Teachers College Press. Zeichner, K., & Gore, J. (1990). Teacher socialization. In W. R. Houston, M. Haberman, & J. Sikula (Eds.), Handbook of research on teacher education (pp. 329–348). New York, NY: Macmillan. Zumwalt, K., & Craig, E. (2005). Teachers’ characteristics: Research on the demographic profile. In M. Cochran-Smith & K. M. Zeichner (Eds.), Studying teacher education: The report of the AERA panel on research and teacher education (pp. 111–156). Mahwah, NJ: Lawrence Erlbaum. https://doi.org/10.1016/j.tate.2017.03.016 https://doi.org/10.1177/0022487117702583 https://doi.org/10.1177/0022487117702583 https://doi.org/10.1177/0013124504264360 https://doi.org/10.1177/0022487117702583 https://doi.org/10.1177/0022487117702583 https://doi.org/10.1177/0022487117702583 Recruitment, Employment, Retention and the Minority Teacher Shortage 38 About the Authors Richard M. Ingersoll University of Pennsylvania rmi@upenn.edu Richard Ingersoll is Board of Overseers Professor of Education and Sociology at the University of Pennsylvania. His research is concerned with the character of elementary and secondary schools as workplaces, teachers as employees and teaching as a job. Henry May University of Delaware hmay@udel.edu Henry May is Director of the Center for Research in Education and Social Policy ( CRESP) and an associate professor in the School of Education at the University of Delaware . Dr. May specializes in the application of modern statistical methods and mixed -methods in randomized experiments and quasi-experiments studying the implementation and impacts of educational and social interventions and policies. Gregory J. Collins University of Pennsylvania gcollins@gse.upenn.edu Gregory J. Collins is an advanced PhD student at the University of Pennsylvania. His research applies quantitative methods to investigate schools and school districts as organizations and explore how changes in organization affect students, teachers, and other stakeholders. About the Guest Editors Linda Darling-Hammond Learning Policy Institute ldh@learningpolicyinstitute.org Dr. Linda Darling-Hammond is President of the Learning Policy Institute and Charles E. Ducommun Professor of Education Emeritus at Stanford University. She has conducted extensive research on issues of educator supply, demand, and quality. Among her award-winning publications in this area are What Matters Most: Teaching for America’s Future; Teaching as the Learning Profession; Powerful Teacher Education; and Preparing Teachers for a Changing World: What Teachers Should Learn and Be Able to Do. Anne Podolsky Learning Policy Institute apodolsky@learningpolicyinstitute.org Anne Podolsky is a Researcher and Policy Analyst at the Learning Policy Institute. Her research focuses on improving educational opportunities and outcomes, especially for students from underserved communities. mailto:rmi@upenn.edu mailto:hmay@udel.edu mailto:gcollins@gse.upenn.edu mailto:ldh@learningpolicyinstitute.org mailto:apodolsky@learningpolicyinstitute.org Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 39 education policy analysis archives Volume 27 Number 37 April 8, 2019 ISSN 1068-2341 Readers are free to copy, display, and distribute this article, as long as the work is attributed to the author(s) and Education Policy Analysis Archives, it is distributed for non- commercial purposes only, and no alteration or transformation is made in the work. More details of this Creative Commons license are available at http://creativecommons.org/licenses/by-nc-sa/3.0/. All other uses must be approved by the author(s) or EPAA. EPAA is published by the Mary Lou Fulton Institute and Graduate School of Education at Arizona State University Articles are indexed in CIRC (Clasificación Integrada de Revistas Científicas, Spain), DIALNET (Spain), Directory of Open Access Journals, EBSCO Education Research Complete, ERIC, Education Full Text (H.W. Wilson), QUALIS A1 (Brazil), SCImago Journal Rank; SCOPUS, SOCOLAR (China). Please send errata notes to Audrey Amrein-Beardsley at audrey.beardsley@asu.edu Join EPAA’s Facebook community at https://www.facebook.com/EPAAAAPE and Twitter feed @epaa_aape. http://www.doaj.org/ mailto:audrey.beardsley@asu.edu https://www.facebook.com/EPAAAAPE Recruitment, Employment, Retention and the Minority Teacher Shortage 40 education policy analysis archives editorial board Lead Editor: Audrey Amrein-Beardsley (Arizona State University) Editor Consultor: Gustavo E. Fischman (Arizona State University) Associate Editors: David Carlson, Lauren Harris, Eugene Judson, Mirka Koro-Ljungberg, Scott Marley, Molly Ott, Iveta Silova (Arizona State University) Cristina Alfaro San Diego State University Amy Garrett Dikkers University of North Carolina, Wilmington Susan L. Robertson Bristol University Gary Anderson New York University Gene V Glass Arizona State University Gloria M. Rodriguez University of California, Davis Michael W. Apple University of Wisconsin, Madison Ronald Glass University of California, Santa Cruz R. Anthony Rolle University of Houston Jeff Bale University of Toronto, Canada Jacob P. K. Gross University of Louisville A. G. Rud Washington State University Aaron Bevanot SUNY Albany Eric M. Haas WestEd Patricia Sánchez University of University of Texas, San Antonio David C. Berliner Arizona State University Julian Vasquez Heilig California State University, Sacramento Janelle Scott University of California, Berkeley Henry Braun Boston College Kimberly Kappler Hewitt University of North Carolina Greensboro Jack Schneider University of Massachusetts Lowell Casey Cobb University of Connecticut Aimee Howley Ohio University Noah Sobe Loyola University Arnold Danzig San Jose State University Steve Klees University of Maryland Jaekyung Lee SUNY Buffalo Nelly P. Stromquist University of Maryland Linda Darling-Hammond Stanford University Jessica Nina Lester Indiana University Benjamin Superfine University of Illinois, Chicago Elizabeth H. DeBray University of Georgia Amanda E. Lewis University of Illinois, Chicago Adai Tefera Virginia Commonwealth University David E. DeMatthews University of Texas at Austin Chad R. Lochmiller Indiana University A. Chris Torres Michigan State University Chad d'Entremont Rennie Center for Education Research & Policy Christopher Lubienski Indiana University Tina Trujillo University of California, Berkeley John Diamond University of Wisconsin, Madison Sarah Lubienski Indiana University Federico R. Waitoller University of Illinois, Chicago Matthew Di Carlo Albert Shanker Institute William J. Mathis University of Colorado, Boulder Larisa Warhol University of Connecticut Sherman Dorn Arizona State University Michele S. Moses University of Colorado, Boulder John Weathers University of Colorado, Colorado Springs Michael J. Dumas University of California, Berkeley Julianne Moss Deakin University, Australia Kevin Welner University of Colorado, Boulder Kathy Escamilla University ofColorado, Boulder Sharon Nichols University of Texas, San Antonio Terrence G. Wiley Center for Applied Linguistics Yariv Feniger Ben-Gurion University of the Negev Eric Parsons University of Missouri-Columbia John Willinsky Stanford University Melissa Lynn Freeman Adams State College Amanda U. Potterton University of Kentucky Jennifer R. Wolgemuth University of South Florida Rachael Gabriel University of Connecticut Kyo Yamashiro Claremont Graduate University Education Policy Analysis Archives Vol. 27 No. 37 SPECIAL ISSUE 41 archivos analíticos de políticas educativas consejo editorial Editor Consultor: Gustavo E. Fischman (Arizona State University) Editores Asociados: Armando Alcántara Santuario (Universidad Nacional Autónoma de México), Angelica Buendia, (Metropolitan Autonomous University), Alejandra Falabella (Universidad Alberto Hurtado, Chile), Antonio Luzon, (Universidad de Granada), José Luis Ramírez, (Universidad de Sonora), Paula Razquin (Universidad de San Andrés), Maria Alejandra Tejada-Gómez (Pontificia Universidad Javeriana, Colombia) Claudio Almonacid Universidad Metropolitana de Ciencias de la Educación, Chile Ana María García de Fanelli Centro de Estudios de Estado y Sociedad (CEDES) CONICET, Argentina Miriam Rodríguez Vargas Universidad Autónoma de Tamaulipas, México Miguel Ángel Arias Ortega Universidad Autónoma de la Ciudad de México Juan Carlos González Faraco Universidad de Huelva, España José Gregorio Rodríguez Universidad Nacional de Colombia, Colombia Xavier Besalú Costa Universitat de Girona, España María Clemente Linuesa Universidad de Salamanca, España Mario Rueda Beltrán Instituto de Investigaciones sobre la Universidad y la Educación, UNAM, México Xavier Bonal Sarro Universidad Autónoma de Barcelona, España Jaume Martínez Bonafé Universitat de València, España José Luis San Fabián Maroto Universidad de Oviedo, España Antonio Bolívar Boitia Universidad de Granada, España Alejandro Márquez Jiménez Instituto de Investigaciones sobre la Universidad y la Educación, UNAM, México Jurjo Torres Santomé, Universidad de la Coruña, España José Joaquín Brunner Universidad Diego Portales, Chile María Guadalupe Olivier Tellez, Universidad Pedagógica Nacional, México Yengny Marisol Silva Laya Universidad Iberoamericana, México Damián Canales Sánchez Instituto Nacional para la Evaluación de la Educación, México Miguel Pereyra Universidad de Granada, España Ernesto Treviño Ronzón Universidad Veracruzana, México Gabriela de la Cruz Flores Universidad Nacional Autónoma de México Mónica Pini Universidad Nacional de San Martín, Argentina Ernesto Treviño Villarreal Universidad Diego Portales Santiago, Chile Marco Antonio Delgado Fuentes Universidad Iberoamericana, México Omar Orlando Pulido Chaves Instituto para la Investigación Educativa y el Desarrollo Pedagógico (IDEP) Antoni Verger Planells Universidad Autónoma de Barcelona, España Inés Dussel, DIE-CINVESTAV, México Paula Razquin Universidad de San Andrés, Argentina Catalina Wainerman Universidad de San Andrés, Argentina Pedro Flores Crespo Universidad Iberoamericana, México José Ignacio Rivas Flores Universidad de Málaga, España Juan Carlos Yáñez Velazco Universidad de Colima, México javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/816') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/819') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/820') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/4276') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/1609') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/825') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/797') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/823') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/798') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/555') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/814') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/2703') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/801') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/826') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/802') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/816') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/3264') javascript:openRTWindow('http://epaa.asu.edu/ojs/about/editorialTeamBio/804') Recruitment, Employment, Retention and the Minority Teacher Shortage 42 arquivos analíticos de políticas educativas conselho editorial Editor Consultor: Gustavo E. Fischman (Arizona State University) Editoras Associadas: Kaizo Iwakami Beltrao, (Brazilian School of Public and Private Management - EBAPE/FGV, Brazil), Geovana Mendonça Lunardi Mendes (Universidade do Estado de Santa Catarina), Gilberto José Miranda, (Universidade Federal de Uberlândia, Brazil), Marcia Pletsch, Sandra Regina Sales (Universidade Federal Rural do Rio de Janeiro) Almerindo Afonso Universidade do Minho Portugal Alexandre Fernandez Vaz Universidade Federal de Santa Catarina, Brasil José Augusto Pacheco Universidade do Minho, Portugal Rosanna Maria Barros Sá Universidade do Algarve Portugal Regina Célia Linhares Hostins Universidade do Vale do Itajaí, Brasil Jane Paiva Universidade do Estado do Rio de Janeiro, Brasil Maria Helena Bonilla Universidade Federal da Bahia Brasil Alfredo Macedo Gomes Universidade Federal de Pernambuco Brasil Paulo Alberto Santos Vieira Universidade do Estado de Mato Grosso, Brasil Rosa Maria Bueno Fischer Universidade Federal do Rio Grande do Sul, Brasil Jefferson Mainardes Universidade Estadual de Ponta Grossa, Brasil Fabiany de Cássia Tavares Silva Universidade Federal do Mato Grosso do Sul, Brasil Alice Casimiro Lopes Universidade do Estado do Rio de Janeiro, Brasil Jader Janer Moreira Lopes Universidade Federal Fluminense e Universidade Federal de Juiz de Fora, Brasil António Teodoro Universidade Lusófona Portugal Suzana Feldens Schwertner Centro Universitário Univates Brasil Debora Nunes Universidade Federal do Rio Grande do Norte, Brasil Lílian do Valle Universidade do Estado do Rio de Janeiro, Brasil Flávia Miller Naethe Motta Universidade Federal Rural do Rio de Janeiro, Brasil Alda Junqueira Marin Pontifícia Universidade Católica de São Paulo, Brasil Alfredo Veiga-Neto Universidade Federal do Rio Grande do Sul, Brasil Dalila Andrade Oliveira Universidade Federal de Minas Gerais, Brasil