Teacher Turnover in High-Poverty Schools: What We Know and Can Do
by Nicole S. Simon & Susan Moore Johnson - 2015
Background/Context: Over the past three decades, teacher turnover has increased substantially in U.S. public schools, especially in those serving large portions of low-income students of color. Teachers who choose to leave high-poverty schools serving large numbers of students of color usually transfer to schools serving wealthier, Whiter student populations. Some researchers have interpreted this trend to mean that “teachers systematically favor higher-achieving, non-minority, non-low-income students.” These ideas have influenced policy analysis concerning high-poverty schools but offered little guidance for those who would address this problem.
Purpose/Objective/Research Question/Focus of Study: This article presents an alternative explanation for turnover—one grounded in organizational theory and substantiated by an emerging line of research. In doing so, it reframes the debate over what fuels high rates of teacher turnover in high-poverty schools and provides advice for policy makers and practitioners, as well as recommendations for productive possibilities for future research.
Research Design: This article reviews six studies analyzing turnover as a function of school context rather than as a function of student demographics. Based on the patterns regarding what factors influence teacher departures across these studies, we pursue these predictors by summarizing what is known about them and how each supports teachers’ work.
Findings/Results: The six overarching studies reviewed here collectively suggest that teachers who leave high-poverty schools are not fleeing their students. Rather, they are fleeing the poor working conditions that make it difficult for them to teach and for their students to learn. The working conditions that teachers prize most—and those that best predict their satisfaction and retention—are social in nature. They include school leadership, collegial relationships, and elements of school culture.
Conclusions/Recommendations: The poor working conditions common in America’s neediest schools explain away most, if not all, of the relationship between student characteristics and teacher attrition. This is important because, unlike demographic characteristics of students, working conditions can be addressed. Policy makers and practitioners have many options for improving aspects of the school environment, and, although more research can inform this work, much is already known about what matters to teachers as they are deciding whether to continue teaching in their schools.
Over the past three decades, teacher turnover has increased substantially in U.S. public schools (Ingersoll & Merrill, 2012). In historically underserved communities, the problems caused by turnover are especially pronounced (Allensworth, Ponisciak, & Mazzeo, 2009; Hanushek, Kain, & Rivkin, 2004; Hemphill & Nauer, 2009; Ingersoll, 2001; Johnson, Berg, & Donaldson, 2005; Marinell & Coca, 2013; Ronfeldt, Loeb, & Wyckoff, 2013). High rates of turnover make it difficult for schools to attract and develop effective teachers and, as a result, low-income and minority students who attend so-called hard-to-staff schools are routinely taught by the least experienced, least effective teachers (Borman & Dowling, 2008; Carroll, Reichardt, Guarino, & Mejia, 2000; Clotfelter, Ladd, Vigdor, & Wheeler, 2007; Hanushek et al., 2004; Ingersoll, 2001; Sanders & Rivers, 1996). Efforts to solve these staffing problems have focused primarily on recruiting promising teachers into high-poverty schools, often with little attention to systematically supporting and retaining them once they are there (Ingersoll & May, 2011; TNTP, 2012). Consequently, problematic teacher turnover persists in public schools that serve low-income communities, making sustained improvement an extraordinary challenge (Allensworth et al., 2009; Ingersoll, 2001, 2004).
Curbing the constant churn of teachers through high-poverty schools is necessary if students are to receive the education they deserve (Ronfeldt et al., 2013). Attaining stability in staffing is especially important for low-income students, who, research suggests, are especially dependent on their teachers (Downey, Von Hippel, & Hughes, 2008). Further, because of high turnover, these are the students most likely to be taught by inexperienced teachers who, on average, are less effective than their more experienced colleagues (Clotfelter, Ladd, & Vigdor, 2005; Grissom, 2011; Ost, 2014; Rivkin, Hanushek, & Kain, 2005). To date, most studies of teacher turnover in high-poverty schools have focused on the characteristics of students and their teachers rather than on the school organizations where they learn and teach. Although it is clear that factors such as salary, professional status, and geographic location (Auguste, Kihn, & Miller, 2010; Reininger, 2012) influence teachers career decisions, past accounts of teacher turnover have paid far less attention to the school environmentespecially in high-poverty schools where many teachers have chosen to begin careers despite well-publicized drawbacks. Here, we reframe the issue of teacher turnover so that policy makers and practitioners can better understand the role of the work environment in teachers career decisions and ultimately address the problem of turnover more effectively.
A NEW PERSPECTIVE ON TEACHER TURNOVER
Traditionally, researchers have used demographic characteristics of students and individual teachers as the predictors of teachers career choices (Boyd, Lankford, Loeb, & Wyckoff, 2005; Carroll et al., 2000; Clotfelter, Ladd, & Vigdor, 2011; Hanushek et al., 2004; Scafidi, Sjoquist, & Stinebrickner, 2007). These studies have repeatedly found that teachers in schools serving high concentrations of low-income, low-achieving students of color are more likely to leave than their counterparts in other schools. When they leave, these teachers usually either exit the profession or transfer to schools that have better academic records and serve Whiter, wealthier students. In 2004, Hanushek and his colleagues interpreted this trend to mean that teachers systematically favor higher-achieving, non-minority, non-low-income students (p. 337). Other researcherssome building directly on Hanushek et al.s workreached similar conclusions (Boyd et al., 2005; Carroll et al., 2000; Scafidi et al., 2007).
However, an emerging but substantial body of literature has reframed the study of turnover by exploring whether the notoriously poor working conditions that prevail in low-income schools (Allensworth et al., 2009; Johnson, Kraft, & Papay, 2012) might be a more powerful driver of teacher turnover than student demographics. Together, this second set of studies1 (Allensworth et al., 2009; Boyd et al., 2011; Loeb, Darling-Hammond, & Luczak, 2005; Johnson et al., 2012; Ladd, 2011; Marinell & Coca, 2013) suggests that, on average, when teachers leave schools serving low-income, minority students, they are not fleeing their students. On the contrary, teachers often enter such schools precisely because of their humanistic commitment to teaching in long underserved communities (Achinstein, Ogawa, Sexton, & Freitas, 2010, p. 71; Cochran-Smith et al., 2012; Kraft et al., 2013). When these teachers leave, it is frequently because the working conditions in their schools impede their chance to teach and their students chance to learn (Johnson, 1990, 2006; Johnson & Birkeland, 2003). Therefore, these findings suggest that policy makers and practitioners who wish to retain talented, effective teachers in high-poverty, hard-to-staff schools must pursue retention strategies that are designed to improve the teaching environment.
We begin by describing the prevalence of teacher turnover in American public schools and explaining why turnover is problematic. We next discuss the early research on turnover in high-poverty schools, which focuses on the demographic characteristics of students and teachers. Based on that summary of early research, we introduce a set of recent studies analyzing turnover as a function of school context rather than as a function of student demographics. Repeatedly in this recent research, the working conditions found to be most important to teachers and the most salient predictors of their satisfaction and predicted retention are social in natureschool leadership, collegial relationships, and elements of school culture. We pursue each of these predictors by summarizing what is known about these three components of working conditions. Finally, we conclude the article by explaining the implications of this work for research, policy, and practice.
THE PREVALENCE OF THE PROBLEM: TEACHER TURNOVER IN HIGH-POVERTY SCHOOLS
Nationwide, rates of turnover among teachers have been rising since the mid-1980s (Ingersoll & Merrill, 2012). Of the 3.5 million public school teachers in the United States, roughly half a million leave their schools each year. Approximately 60% of this turnover results from teachers transferring between schools, while about 40% results from teachers leaving the profession (Alliance for Excellent Education, 2008). However, regardless of why teachers leave or where they go, vacancies remain in their wake, and ultimately, schools must cope with their departure.
Turnover rates in teaching are much higher than in most high-status, high-paying professions (Ingersoll & Merrill, 2012). Instability is especially worrisome in high-poverty schools, wherewhen last documented over a decade agothe turnover rate has been roughly 50% higher than in wealthier schools (Ingersoll, 2001). Although some high-poverty schools do not experience significant turnover (Chenoweth, 2009; R. Ferguson, 2009; Johnson & Birkeland, 2003), many do. On average each year, high-poverty public schoolsespecially those in urban areaslose 20% of their faculty (Ingersoll, 2004). In fact, many schools serving Americas neediest children lose over half of their teaching staff every five years (Allensworth et al., 2009; Hemphill & Nauer, 2009), and in New York City middle schools, 66% of teachers exit within five years of entry (Marinell & Coca, 2013). Consequently, students at high-poverty schools are more likely than their peers in wealthier schools to experience inconsistent staffing from one year to the next and to be taught by teachers who are new to their school and, often, new to the profession (Hanushek et al., 2004; Hemphill & Nauer, 2009; Johnson et al., 2005).
Not surprisingly, schools that have trouble retaining teachers also struggle to fill vacancies as they arise, contributing to a cycle of chronic turnover as principals who have trouble finding strong candidates are forced to settle for teachers who are not a good fit for their school (Neild, Useem, Travers, & Lesnick, 2003). Shallow applicant pools coupled with poor hiring practices lead to mismatches, and subsequently, to more dissatisfaction and turnover (Liu, Rosenstein, Swan, & Khalil, 2008, p. 299). Ingersoll (2001) dubbed this phenomenonwhere large numbers of quality teachers depart their jobs for reasons other than retirementteachings revolving door (p. 501).
WHY TURNOVER MATTERS
Modest rates of turnover might positively affect schools if the departing teachers were ineffective instructors or uncooperative colleagues. It is normal for some teachers to leave their schools each year, either by choice or because they are dismissed, and certainly some departures may be beneficial. But a pattern of chronic turnover exacts instructional, financial, and organizational costs that destabilize learning communities and directly affect student learning (Achinstein et al., 2010; Allensworth et al., 2009; Balu, Beteille, & Loeb, 20092010; Guin, 2004; Ingersoll, 2001; Johnson et al., 2005; Ronfeldt et al., 2013). These costs may differ depending on how teachers interact within a school (Hanselman, Grigg, Bruch, & Gamoran, 2013) or on the demographic makeup of a school. In one of the first empirical studies about how turnover compromises the instructional effectiveness of schools, Ronfeldt and colleagues (2013) found that the negative effects of turnover on academic achievement are greater for low-performing and Black students than for their higher performing, non-Black peers.
Repeated waves of new teachers create problems for several reasons. First, schools with high turnover commonly employ a disproportionately large number of novice teachers, who, on average, are less effective than those with more teaching experience (Clotfelter et al., 2005; Grissom, 2011; Ost, 2014; Rivkin et al., 2005). When students are repeatedly taught by new teachers, they pay a substantial price year after year in the quality of instruction they receive. Second, schools with high turnover often must reconfigure their teaching assignments each year in response to staffing changes caused by transfers and new arrivals. Disruptions in instructional continuity result in less comprehensive and unified instructional programs for students (Guin, 2004, p. 19), and this directly affects their learning (Allensworth et al., 2009; Balu et al., 20092010; Boyd, Grossman, Lankford, Loeb, & Wyckoff, 2006; Ingersoll, 2001; Ronfeldt et al., 2013). This may be because, as Ost (2014) recently demonstrated, when a teacher switches grade assignments, the consequences for students are similar to teacher attrition because [the teachers] grade-specific human capital is wasted (p. 24).
Third, repeated turnover thwarts the kind of continuity needed to build sustained, trustful relationships among teachers, students, and families. Such relationships develop over time and are critical for forming a sense of community unified by a common mission and an agreed-on strategy for achieving it (Hanselman et al., 2013). Sustained and stable relationships also allow schools to establish norms for instructional quality, professional conduct, student behavior, and parental involvementall of which are linked to student achievementespecially for financially impoverished students (Bryk & Schneider, 2002; Bryk, Sebring, Allensworth, Easton, & Luppescu, 2010; Ronfeldt et al., 2013).
Finally, when schools lack the social capital that strong collegial relationships create, teachers may be reluctant to take on leadership roles or to form professional learning communities. Likewise, in a school with few experienced teachers, the human capital necessary to effectively mentor new teachers may simply not exist (Loeb et al., 2005). These factors, in turn, reduce a schools capacity to develop programs and implement curricula, and they impede teachers ability to improve instruction together, over time (Allensworth et al., 2009). Ronfeldt and colleagues (2013) suggested that an unstable teaching culture may account for the disruptional effects (p. 17) that turnover has on all students at high-turnover schoolsincluding the students of teachers who remain. In response to their finding that turnover has a broader, harmful influence on student achievement since it can reach beyond just those students of teachers who left or of those that replaced them, the authors surmised that one possibility is that turnover negatively affects collegiality or relational trust among faculty; or perhaps turnover results in loss of institutional knowledge among faculty that is critical for supporting all student learning (p.18).
Of course, turnover also compromises instruction by diverting resources from classrooms, thereby widening the gaps between low-income and wealthier schools (Grissom, 2011). Schools and districts that struggle to retain teachers incur large costs when they must repeatedly recruit, hire, induct, and develop replacement teachers. The National Commission on Teaching & America's Future (Barnes, Crowe, & Schaefer, 2007) estimated that nationally, $7.34 billion are spent each year replacing teachers. They found that on average in urban districts, individual schools spend $70,000 annually on costs associated with turnover. In contrast, nonurban schools spend $33,000 on average. However, the true cost of turnover is likely much higher. These calculations do not account for the additional expenses associated with the loss of human capital and related productivity, given that students are repeatedly taught by beginning teachers who are, on average, less effective than the somewhat more experienced teachers they replace (Milanowski & Odden, 2007; Papay & Johnson, 2011).
WHAT LEADS TEACHERS TO LEAVE? EARLY EXPLANATIONS
In an effort to explain teachers decisions to leave their schools, researchers initially examined the relationship between turnover and either the individual characteristics of teachers or the demographics of students. Although it is important to understand which teachers leave and which schools they exit, these studies do not explain why patterns of teacher turnover have developed and persist in high-poverty schools.
TEACHERS PERSONAL CHARACTERISTICS
Teachers influence students learning more than any other school-based factor
(McCaffrey, Koretz, Lockwood, & Hamilton, 2004; Rivkin et al., 2005; Rockoff, 2004; Sanders & Rivers, 1996; Wright, Horn, & Sanders, 1997). Those who are most effective are likely to leave the schools that need them most (Allensworth et al., 2009; Boyd et al., 2005; Goldhaber & Hansen, 2009). On average, teachers with strong academic backgrounds are prone to move to higher income schools with better student achievementif they dont leave the profession altogether (Alliance for Excellent Education, 2008; Boyd et al., 2005; Deangelis & Presley, 2007; Goldhaber, Choi, & Cramer, 2007). Teachers who are certified as accomplished by the National Board for Professional Teaching Standards are more apt to leave high-minority schools than are teachers who applied but failed to meet the boards standards (Goldhaber & Hansen, 2009). Further, those who teach hard-to-staff subjects rarely stay in hard-to-staff schools (Boe, 2006; Ingersoll & May, 2011; Murnane, Singer, Willett, Kemple, & Olsen, 1991). In general, since 2000, the least experienced teachers have been especially likely to leave. Typically, these novices are replaced by even newer teachers (Allensworth et al., 2009; Boyd et al., 2005; Hanushek et al., 2004; Ingersoll, 2004; Leukens, Lyter, Fox, & Chandler, 2004; Marinell & Coca, 2013). Consequently, the odds that low-income children will be taught by inexperienced teachers are now higher than ever before (Ingersoll & Merrill, 2012; Rice, 2013).
There is an exception to the rule that promising teachers leave high-poverty schools to work in wealthier, Whiter communities. Compared with White teachers, teachers of colorparticularly Black and Latino teachersare more likely to stay in teaching and at schools that serve students with similar racial and socioeconomic backgrounds (Achinstein et al., 2010; Hanushek et al., 2004; Ingersoll, 2001; Ingersoll & May, 2011; Kirby, Berends, & Naftel, 1999; Marinell & Coca, 2013; Murnane et al., 1991). This is notable because data from Tennessees Project STAR experiment reflect rather large educational benefits to both math and reading achievement for students who are assigned to teachers of their own race (Dee, 2004, p. 209; Dee, 2005). These patterns suggest that teachers personal characteristics matter in their decisions about where to teach and when to leave.
STUDENT BODY DEMOGRAPHICS
Many researchers have investigated the factors influencing teacher mobility by modeling departures as a function of the schools student demographics (Boyd et al., 2005; Carroll et al., 2000; Clotfelter et al., 2011; Hanushek et al., 2004; Scafidi et al., 2007). Repeatedly, they have found that on average, when teachers transfer, they move to schools serving fewer low-income, low-achieving minority students.
Hanushek, Kain, and Rivkins 2004 article Why Public Schools Lose Teachers is perhaps the most frequently cited study employing this design, and other researchers have explicitly modeled their work on it (Boyd et al., 2005; Scafidi et al., 2007). Hanushek and his colleagues used matched student/teacher panel data to investigate how salary and student body demographics influence turnover in Texas public elementary schools serving low-income and minority students. As the authors explained, because Texas turnover patterns mirror national patterns, these findings are potentially relevant more broadly.
Hanushek et al. (2004) found that on average, when teachers transfer, they seek out schools with fewer academically and economically disadvantaged students (p. 340). The authors noted that this pattern aligns with the hypothesis of other research: When new teachers are placed in the most difficult teaching situations within urban districts (p. 340), they leave as they gain experience (Raymond, Fletcher, & Luque, 2001, and Raymond & Fletcher, 2002, as cited in Hanushek et al., 2004). Hanushek et al. also found that moves by teachers were much more strongly related to student race and achievement than to salary differentials. They suggested that for some teachers, the salary premia required to offset the turnover effects of student characteristics would be extraordinarily largeespecially in many large urban schools, [which] display a combination of achievement deficits and concentrations of minority students (p. 350). For example, they estimated that the average salary differential required to counter this trend among inexperienced females would be 25%40% above current pay rates.
The authors further suggested that teachers career decisions are driven by their own preferences for student race or ethnicity (p. 352) and that if this is the case, districts possess few policy options for remedying the problems of turnover. That is, if teachers prefer to work with a different type of student, there is little that policy makers can do to change their minds. Hanushek et al. (2004) did mention that working conditions might play a role in teachers decisions but noted that their data set did not permit them to disentang[le] the various potential aspects of working conditions such as disciplinary problems, rigid bureaucracies, poor leadership, high student turnover, and general safety concerns (p. 352).
Scafidi and colleagues (2007), who conducted a similar study in Georgia, reported, The message from our work is remarkably similar to that of Hanushek et al. (2004) (p. 158). They found that the proportion of students that are black has strong relationships to virtually all of the exit reasons (p. 158), and they confirmed that teachers are more likely to leave high-poverty, high-minority schools. The researchers suggested that this may be because teachers potentially receive less enjoyment in minority schools (p. 158).
REFRAMING THE DISCOURSE: EXAMINING TURNOVER THROUGH AN ORGANIZATIONAL LENS
Others have reframed the analysis. Instead of modeling departures as a function of student or teacher demographics, these researchers employed an analysis grounded in organization theory and the sociology of organizations. They suggested that the problem rests with the schools, not with the students. Johnson et al. (2012) wrote that teachers who leave high-poverty, high-minority schools reject the dysfunctional contexts in which they work, rather than the students they teach (p. 4). Using somewhat different survey instruments, sample constructions, and definitions of working conditions, these researchers focused on problems within the school organization that can potentially be manipulated through policy amendments and practice-based interventions (Ingersoll, 2004).
The importance of considering school context in teacher movement patterns was first reported by Richard Ingersoll (2001). Using the National Center for Education Statistics nationally representative School and Staffing Survey (SASS) data set and its supplement, the Teacher Followup Survey (TFS), Ingersoll estimated the effects of teacher characteristics, school characteristics, and organizational conditions on teacher turnover. He found that organizational factors pertaining to administrative support, teacher input in decision-making, salary, and aspects of school culture (especially student discipline) were associated with higher rates of turnovereven when he controlled for school location, school level, and demographic characteristics of teachers and students.
Complementing Ingersolls work is a longitudinal, qualitative study conducted by researchers at the Project on the Next Generation of Teachers (Johnson & Birkeland, 2003; Johnson & Project on the Next Generation of Teachers, 2004). Based on interviews with 50 new teachers over 4 years of teaching, Johnson and Birkeland (2003) reported that although teachers personal circumstances influenced their career decisions, it was their sense of success (p. 581) with their own students that most influenced their decision about whether to stay in their school, move to a different school, or leave teaching altogether. Overall, the teachers said that it was the environment of their school that made success likelyor unlikely. New teachers in schools that were organized to support them through collegial interaction, opportunities for growth, appropriate assignments, adequate resources, and schoolwide structures supporting student learning were more likely to stay in those schools and in teaching than were the new teachers working at schools that lacked such supports. For example, novice teachers who stayed through Year 3 reported having benefitted from working conditions, such as reduced teaching or administrative assignments, regular feedback on their teaching, high-quality professional development, and graduated expectations for improving their instruction. These schools featured the frequent, collegial exchange of information and ideas among experienced and novice teachers in what the authors called an integrated professional culture (p. 605; see also Kardos, Johnson, Peske, Kauffman, & Liu, 2001).
Neither Ingersoll nor Johnson and her colleagues specifically sought to explain turnover rates in high-poverty schools. However, their organizational perspective provided a foundation for subsequent research about the revolving door phenomenon in this subset of schools. In an effort to explain why teachers leave high-poverty schools, some researchers have used this organizational perspective to disentangle the effects of student demographics from working conditions.
EXPLAINING AWAY THE LINK BETWEEN STUDENT DEMOGRAPHICS AND TEACHER TURNOVER
During the last decade, a number of researchers have refined methods for using an organizational perspective to frame analyses of turnover in high-poverty schools. All the studies discussed here use large quantitative data sets, and many build on the research designs employed in prior work. In every study, the researchers concluded that organizational conditionsa concept defined and measured somewhat differently in each caseare strong predictors of turnover. In fact, researchers who included both working conditions and race in their model found that working conditions explain away all or most of the relationship between student demographics and teacher turnover. Further, most of these researchers attempt to identify and disentangle the components of working conditions that affect turnover most profoundly. Repeatedly, they found that teachers perceptions of their principal are among the most important in teachers career decisions. In studies that included collegial relationships and aspects of school culture as predictors in their models of turnover, these factors proved to be especially important as well.
How Teaching Conditions Predict Teacher Turnover in California Schools
Loeb, Darling-Hammond, and Luczak (2005) conducted phone surveys with a representative sample of California teachers1,071 teachers randomly sampled from nearly as many schoolsabout how they experienced working at their school. The researchers used responses to develop a school conditions construct based on teachers ratings of professional development, working conditions for teachers, the teachers own job satisfaction, experiences with required standardized tests, parental involvement, instructional materials, physical facilities, and the availability of technology (p. 55). Then, using regression models that included district data on salary, teacher background characteristics, student demographics, and the working conditions constructs, Loeb and her colleagues examined the relationship between these predictors and two school-level dependent variables: whether teachers reported that turnover was a serious problem and whether teachers reported that vacancies were hard to fill.
Although the racial and economic composition of a school was a strong predictor of both turnover and hiring problems, Loeb et al. found that when the school conditions factor was added to the model, the predictive power of student characteristics on both outcomes was reduced. Notably, the relationship between the proportion of Latino students and teacher turnover was no longer statistically significant. Similarly, when school conditions were added to the model predicting difficulty filling vacancies (p. 62), no student characteristics remained statistically significant predictors of this outcome.
The Schools Teachers Leave
Researchers at the Consortium on Chicago School Research (Allensworth et al., 2009) conducted a similar study in the Chicago Public Schools (CPS), where roughly 20% of schools lose over 30% of their staff annually. At these schools, almost all students are low-income and Black or Latino. In addition, most of the schools risk disciplinary action from district and state officials if student achievement remains low. However, the authors reported that turnover inhibits these schools from building the capacity they need to improve. To examine the mechanisms that drive teachers away from these schools, Allensworth and her colleagues linked individual CPS teachers personnel records with data from the teachers schools. School-level data included students racial and economic composition, school size, student test scores, student mobility rates, the schools probation history, and the principals experience. The researchers also surveyed teachers and students at 400 schools about their schools professional capacity, learning climate, instructional leadership, and parent involvement.
Allensworth et al. (2009) found that factors affecting school climate and organization (p. 25) explained over 75% of the difference in teacher stability rates among elementary schools and nearly all the variation among high schools. Among school organizational factors, the presence of positive, trusting, working relationships (p. 25) was most influential. In schools where teachers reported a strong sense of collective responsibilitywhere there is a shared commitment among the faculty to improve the school so that all students can learn (p. 25)1-year stability rates were 45 percentage points higher than in other schools with comparable demographics. In addition, schools that offered novices support retained more teachers.
Factors related to the quality of the principal also proved to be important. Stability rates were higher in schools where teachers reported having high levels of influence over school decisions, trust in their principals, a strong instructional leader as principal, and coherent instructional programming (Allensworth et al., 2009, p. 26). The authors suggested that these are the schools where the principal and teachers work together to coordinate instruction and programs in a coherent and sustained way (p. 26). Further, they found that although some of the relationship between school leadership and teacher stability was explained by other school-level working conditions, principal leadership remain[ed] a strong, significant predictor of teacher stability on its own (p. 26).
Allensworth et al. (2009) also provided new findings about school culture and teacher retention. First, teacherparent relationships predicted a statistically significant portion of turnover in both elementary and high schools. Second, teachers were more likely to remain at schools that provided safe and supportive environments for students. This was especially true for high schools, where students reports of appropriate academic behavior among peers as well as teachers perceptions of school safety predicted stability rates among teachers that were 6% higher than in comparison schools.
Teachers Perceptions of Their Working Conditions: How Predictive of Planned and Actual Teacher Movement?
Analyzing data from North Carolinas Teacher Working Conditions Survey, which included responses from over 70% of the states teachers, Helen Ladd (2011) found that teachers reports about their working conditions were highly predictive of their intentions to leave their schools, even when controlling for student race and socioeconomic status. Her study relied on a definition of working conditions that included the physical features of the workplace, the organizational structure, and the sociological, political and psychological and educational features of the work environment (p. 237). Because certain conditions are difficult to measure and may overlap, Ladd employed exploratory factor analysis to sort teachers responses to the wide range of questions about their working conditions.
Consistent with prior studies that included both student demographics and working conditions, Ladd (2011) found that by far the dominant factor predicting both intended and actual school-level turnover was teachers perceptions of school leadership. Teachers who viewed their schools leader positively were less likely to plan to depart than those who did not. Moreover, schools where teachers widely viewed their principals positively experienced less actual turnover one year later than schools where teachers did not view their principals positively. In addition, when elementary and middle school teachers had time in their schedules for collaboration and planning, they were less likely to say that they intended to leave the school. Similarly, in high schools that provided expanded roles for teachers, survey respondents were less likely to report that they planned to leave. Notably, Ladd determined that working conditions predict not only individual teachers plans to stay or leave, as reported in the survey, but also the overall turnover rates at their school one year later.
Ladds (2011) analysis revealed that working conditions contribute more explanatory power to models of turnover than other factorsincluding student demographics. She was clear, however, that these findings do not negate the importance of a schools student demographics. Rather, she urged that working conditions be considered in tandem with demographic information because of the strong correlation between the two. She cautioned that models that try to capture the quality of a teachers work environment solely by easy-to-measure variables such as the proportion of students who are Black are likely to overstate the importance of student race as a repelling factor for teachers (p. 255).
The Influence of School Administrators on Teacher Retention Decisions
Boyd et al. (2011) introduced yet another approach to understanding the factors that influence teachers decisions. They realized that using survey responses by teachers to assess overall working conditions at a school might be problematic because an individuals career plans might influence his outlook on his school. Therefore, they used one group of teachers perceptions about working conditions to predict the career decisions of other teachers at that school. Combining their survey data with administrative data for all New York City public schools and teachers, Boyd et al. examined the links between teachers assessments of working conditions and the career trajectories of both the respondents and all the other teachers in their school. Like other researchers examining this issue, Boyd et al. found that when school context factors were included in their regression model, the predictive power of student race dropped substantially. They reported that their results offer further support for the causal relationship between school working conditions and teacher retention decisions (p. 311).
Basing their survey on the SASS, suggestions from district staff, and prior research, Boyd et al. (2011) included six primary factors to measure school working conditions: teacher influence, administration, staff relations, student composition, quality of facilities, and school safety. With rare exception, teachers in schools with fewer low-income and/or fewer Black and Latino students reported better working conditions. In addition, among these contextual factors, the researchers found that novice teachers perceptions of their schools administration were by far the strongest predictor of decisions to transfer or to leave teaching among both novice teachers and their more experienced colleagues. Although each factor (except perceptions of school safety) was a statistically significant predictor of teacher turnover when included in the model separately, in the full model, perceptions of the schools administration remained the only statistically significant predictor of attrition.
How Context Matters in High-Need Schools: The Effects of Teachers Working Conditions on Their Professional Satisfaction and Their Students Achievement
Johnson, Kraft, and Papay (2012) came to similar conclusions in their investigation of why teachers leave high-poverty schools. Analyzing data from the statewide Massachusetts Teaching, Learning and Leading (MassTeLLS) survey, they found that measures of school context explained a large portion of the apparent relationship between student demographic characteristics and both teacher satisfaction and career intentions. They also found that favorable working conditions predicted higher rates of academic growth, as measured by standardized tests, even in schools serving similar low-income, high-minority student populations. They suggested that students would be well-served to attend a school that is known to be a good place to teach, since that school is likely to attract and retain other like-minded teachers (p. 28).
Moreover, Johnson et al. sought to gauge the relative weight in predicting turnover of nine key theory-based factors: colleagues, community support, facilities, governance, principals, professional expertise, resources, school culture, and time. Based on their analysis of the survey data, they reported that the working conditions that mattered most to teachers were the principals leadership, collegial relationships, and the schools organizational culture, defined as the extent to which the school environment is characterized by mutual trust, respect, openness and commitment to student achievement (p. 14). These factors, all involving social interactions, most strongly predicted teachers satisfaction with their jobs and their future career plans. However, because these factors are highly correlated, it was difficult to disentangle their effects using a quantitative analysis.
Who Stays and Who Leaves? Findings From a Three-Part Study of Teacher Turnover in NYC Middle Schools
In the most comprehensive study of middle school teacher turnover to date, the Research Alliance for New York City Schools (Marinell & Coca, 2013) drew on 10 years of administrative data from the nations largest district. They found that of all middle school teachers who had entered teaching between 2002 and 2009, 66% left their schools within 5 years. This pattern was similar among high school teachers (65%) and elementary school teachers (59%). To explore what motivates this mass exodus, the authors surveyed over 4,000 middle school teachers about how long they intended to stay at their school or in teaching, whether they were considering leaving, and how they experienced various aspects of their work environment, including principal leadership, school disorder, teacher collegiality, and professional controldefined as the extent to which teachers feel they have control over various aspects of their work (p. 25). The researchers complemented their findings with actual turnover data from all middle schools in the district and with interviews at four case study schools.
Using regression analyses to determine whether working conditions predicted turnover, Marinell and Coca (2013) found that teachers were more likely to remain at schools where they considered the principal trusting and supportive of the teaching staff, a knowledgeable instructional leader, an efficient manager, and adept at forming partnerships with external organizations. Higher levels of order at a school, indicated by fewer incidents of violence, theft, disrespect towards teachers, and student absenteeism (p. viii), were also associated with lower levels of turnover. Although less influential than the presence of a strong leader and a sense of order in the school, collegiality (defined as support, rapport, trust and respect among teachers) was also associated with lower rates of turnover (p. 27).
Together, these studies make an important case for reframing the discourse about teacher turnover to focus on the school as a workplace. Using a range of methods and an array of samples, they demonstrate that the conditions in which teachers work affect their decisions about whether to stay at their school. When working conditions are considered, the effect of student demographics on turnover is considerably diminished or eliminated altogether. In addition, the positive correlation between working conditions and student achievement suggests that working conditions may directly affect student learning.
ADMINISTRATIVE SUPPORT, COLLEGIAL RELATIONSHIPS, AND SCHOOL CULTURE: HOW DO THEY MATTER?
In virtually every study discussed here, some combination of social working conditionsthe quality of school leadership, the caliber of collegial relationships, and specific aspects of school cultureis found to influence teachers satisfaction and their anticipated or actual career decisions. However, these studies do not closely analyze what it is about school leadership that matters, why teachers care who their colleagues are, or whether some elements of school culture (variously defined) drive teachers decisions more strongly than others. In part, this is because it is challenging for quantitative studies to separate out the effects of elements that are highly correlated. Yet, if policy makers and practitioners are to identify actions and initiatives that would improve teachers working conditions and therefore students learning conditions, we must understand these factors much better.
The literature is rich with examples of how strong principals broker strong workplace conditions (Johnson, 2006, p. 15). As leaders, principals can articulate a clear vision for their learning community and set the stage for seeing that vision through. They can invest in human capital by hiring skilled teachers and support staff who are committed to their schools mission (Beteille, Kalogrides, & Loeb, 2011; Liu et al., 2008). They can assign teachers to appropriate subject areas and grades (Balu et al., 20092010; Donaldson & Johnson, 2011), and they can create formal structures that support instruction (Borman & Dowling, 2008), such as peer mentoring, common planning time, and well-defined disciplinary systems (Supovitz & Poglinco, 2001). Principals can set a positive, professional tone by building trust (Bryk & Schneider, 2002), rewarding collaboration, and developing opportunities for differentiated roles for teachers (Rosenholtz, 1989). Principals can navigate tricky political situations and take steps to shelter teachers from the brunt of external demands and mandates that may distract them from their teaching (Achinstein et al., 2010; Waters, Marzano, & McNulty, 2003). Principals can listen to students views and promote student leadership (Mitra, 2008), build relationships with parents (Achinstein et al., 2010; Mapp, 2003), and partner with community organizations to maximize students opportunities to learn (Spillane, Hallett, & Diamond, 2003; Warren, 2005).
Grissom (2011) suggested that principals may be even more critical in high-poverty schools than they are in wealthier schools, which is especially problematic because on average, high-poverty schools are led by inexperienced, weaker-than-average principals (Loeb, Kalogrides, & Horng, 2010). Analyzing SASS/TFS survey data, Grissom (2011) found that teachers working in a high-poverty school led by an effective principal are generally more satisfied than teachers in a non disadvantaged school (p. 2552) working under an equally effective principal. Grissom concluded that ultimately, an effective principal completely offsets teacher turnover in disadvantaged schools (p. 2576).
It is difficult to disentangle the many ways in which principals affect teachers work and their decisions about whether to stay or go. However, teachers repeatedly cite a small number of factorsthe principals effectiveness as a school manager, instructional leadership, and inclusiveness in decision-making.
Principals as General Managers
The basic responsibility of a principal is to ensure that the school work[s] properly (Bryk et al., 2010, p. 62)that teachers have supplies, meetings operate on schedule, facilities are cared for, budgets are sensibly managed, and communication channels work. Grissom and Loeb (2011) identified principals organizational management skills as a predictor of student achievement and suggest that without attention to organizational management, principals rarely produce overall school improvement. They surmise that this is partly because teachers who can find positions elsewhere often leave when frustrated by poor management. They therefore recommend that policy makers consider encouraging or assigning principals with strong organizational management skills to work in high-poverty schools as a strategy for addressing income-achievement gaps. Johnson and Birkeland (2003) found similar evidence regarding the importance of principals organizational management skills as they interviewed new teachers. For example, one teacher who had left the classroom altogether did so in part because of a principal whose lack of leadership skills caused the schoola community-based charter serving African American studentsto unravel (p. 596). After encouraging teachers to develop a school improvement plan, the principal abandoned it suddenly. Because the school had not really [set] up structures [to help teachers] to do the best job, given the population, and the things that [teachers] have to deal with (p. 596), the teacher explained that ultimately, an estimated 70% of the staff left. As Bryk and colleagues (2010) explained, poor management undermines teachers classroom work by eating away at the amount of effective instructional time (p. 61).
Teachers identify fairness as an especially important component of their principals management skills, and they report leaving schools to avoid principals who are arbitrary, abusive, or neglectful (Johnson & Birkeland, 2003, p. 594). When they transfer, teachers seek out fair and encouraging leaders (p. 599). Using SASS data, Grissom and Keiser (2011) explored ideas of fairness in their study of factors influencing minority teacher retention. They found that White principals tend to allocate special benefitsincluding more pay and more specialized positionsto same-race teachers. Consequently, Black teachers earned less than their observationally equivalent White colleagues at the same school (Grissom & Keiser, 2011). The authors speculated that this trend may explain why, as teachers gain experience, they sort toward schools with principals of the same race. These findings raise further concerns because schools that serve large numbers of minority students attract the greatest proportion of minority teachers but are rarely led by principals of color (Grissom, 2011).
Principals as Instructional Leaders
Teachers are more likely to stay in schools where their principals recognize the many things that they, as school leaders, can do to influence instruction and invest in it by engaging in a deliberate orchestration of people, programs, and extant resources (Bryk et al., 2010, p. 63). Teachers want to be in schools where principals strategically hire the right people and actively retain them (Balu et al., 20092010; Brown & Wynn, 2009; Johnson & Birkeland, 2003; Johnson, Kardos, Kauffman, Liu, & Donaldson, 2004; Liu et al., 2008). They want to work for principals who regularly conduct fair evaluations of their teaching practice and in the process provide useful suggestions for improving pedagogy (Reinhorn, 2013). They also want to work for principals who give them an appropriate teaching load (Borman & Dowling, 2008; Johnson et al., 2005; Leukens et al., 2004). Too often, high-poverty schools lose teachers when they are assigned large classes, classes that are outside their field, or assignments that span multiple subjects or grade levels. It is especially common for novice teachers in such schools to cope with several of aspects of misassignment simultaneously, which bears heavily on their sense of efficacy (Donaldson & Johnson, 2010; Ingersoll, 2002; Johnson et al., 2005) and likely affects their students. As Ost (2014) found, even when teachers grade-level assignments are reconfigured, the negative effects on student achievement are especially pronounced among relatively new teachers.
In schools that retain teachers, principals think about their teachers as learners and commit to helping them improve continually (Borman & Dowling, 2008; Cochran-Smith et al., 2012; Johnson & Birkeland, 2003). They can enable teachers to work collaboratively on instruction. As Rosenholtz (1989) explained in her study of elementary schools in Tennessee, norms of collaboration do not spring spontaneously out of teachers mutual respect and concern for each other (p. 44). Rather, they are deliberately constructed in large part by principals. Strong principals intentionally develop and maintain an integrated professional culture in their schools (Kardos et al., 2001, p. 250), where novices benefit from ongoing guidance and support from more experienced teachers while veterans enjoy the challenges of mentoring their new colleagues.
Principals as Inclusive Decision-Makers
Gaining teachers support for programs and other initiatives is an important strategy for retaining teachers. In high-poverty schools, teachers readily grant their principals the authority they need to lead school improvementbut only if they perceive that the principal will listen to their ideas and engage them as partners in the change process (Achinstein & Ogawa, 2011; Bryk et al., 2010; Johnson et al., 2013). Similarly, teachers who experience autonomy and discretion in a range of decisions, from selecting texts to setting expectations for discipline within their classes, are less likely to report feeling discouraged and more likely to exert their best effort and consider teaching for the long term (Rosenholtz, 1989; Weiss, 1999). On the other hand, when principals interrupt, abandon, criticize and maintain control over teachers, teachers report low motivation, feelings of being unsupported, fear and confusion, avoidance [of work] and feelings of being manipulated or abused (Johnson et al., 2005, p. 72). Such problematic power relations (Achinstein & Ogawa, 2011, p. 72), which are common in high-poverty schools (Bryk et al., 2010; Darling-Hammond, 2001), often drive teachers to leave.
Inclusive governance is important to teachers in all schools, but it may be especially important for minority teachers. In their study exploring the minority teacher shortage, Ingersoll and May (2011) documented that on average, more minority teachers leave the profession each year than join it. When asked why they depart, the most common explanations involved feeling dissatisfied about the level of collective decision-making and influence granted to teachers and, similarly, the small degree of autonomy they could exercise within their classroom.
Although Lorties (1975) classic depiction of the school as an egg-crate, where teachers work in isolation and prize their classroom privacy, remains apt in many schools today, subsequent survey research shows that teachers preferences about collaboration have changed in the last half-century. In every survey conducted by the National Education Association since 1956, teachers have identified cooperative/competent colleagues/mentors (Wolman, 2010, p. 328) as one of the top six factors that help them teach, and since the 1990s, it has consistently been ranked most important (see also Drury & Baer, 2011). Therefore, it is not surprising that research reports a clear relationship between teachers experiences with their colleagues and their continued commitment to teaching at their school (Allensworth et al., 2009; Guarino, Santibanez, & Daley, 2006; Johnson et al., 2005; Rosenholtz, 1989). As noted earlier, principals can enable and facilitate collegial relationships, but productive collaboration requires teachers investment as well. The research suggests that three factors support teachers work with their colleagues: an inclusive environment characterized by respect and trust among colleagues, formal structures that promote collaboration, and the presence of a shared mission among teachers.
Inclusive Environment of Respect and Trust
Working in an inclusive environment where teachers respect one another and the students they serve is important to teachers (Achinstein & Ogawa, 2011; Bryk & Schneider, 2002). In interviews with Johnson (1990), teachers said that they relied on their colleagues for both professional and personal needs, including social interaction, reassurance and psychological support (p. 156). In schools where students needs are greatas they often are in high-poverty schoolsit is plausible that teachers depend on one another even more than they do in other schools.
In their work on school-based relationships, Bryk and Schneider (2002) found trust to be a prerequisite for building communities where teachers count on one another. In schools where teachers doubt whether their colleagues are doing the right thing . . . [for] the right reasons (p. 21), they are less willing to work together. Although scholars have not systematically studied how components of trust affect turnover, teachers who do not feel that they can trust their coworkers are likely to seek employment elsewhere.
Formal Structures for Collaboration and Support
Teachers want to work in schools where educators support one another by teach[ing] each other the practice of teaching (Little, 1982, p. 331). As Little explained, schools do this by systematically engaging faculty in frequent, continuous and increasingly concrete and precise talk about teaching practice (as distinct from teacher characteristics and failings, the social lives of teachers, the foibles and failures of students and their families and the unfortunate demands of society on the school) (p. 331).
In their study of professional culture, Kardos and colleagues (2001) considered how norms of collaboration might influence teacher satisfaction and retention. In schools with an integrated professional culture (p. 250), teachers work responsibilities were deliberately arranged to intersect (p. 277) through exchanges that drew on both the new ideas of novices and the wisdom of veterans. Consequently, teachers viewed themselves as being part of a collective with joint responsibility for each other, their students, and the school community. Teachers benefiting from an integrated professional culture reported greater satisfaction with their schools than did teachers in schools where the professional culture was exclusively oriented toward the concerns of novices or veterans.
The norms of collaboration that emerge in an integrated professional culture are deliberately developed, often through formal structures, such as well-designed mentoring and induction programs. These structures help novices acclimate to their new role (Kardos et al., 2001) and have been linked to teacher retention (Cochran-Smith, et al., 2012; Ingersoll, 2003; Kapadia, Coca, & Easton, 2007). However, many mentoring and induction programs are poorly structured and thus fail to deliver the positive results many expect (Glazerman et al., 2010). This problem is especially prevalent in low-income schools, where novices are less likely to have mentors (Donaldson & Johnson, 2010; Johnson et al., 2004; Kardos & Johnson, 2010). When they do have mentors, they are less likely to report having meaningful interactions with them about their teaching, in part because they often do not teach the same subject or grade level or even at the same school. Without the personalized support they need, teachers frequently flounder (Cochran-Smith, et al., 2012; Donaldson & Johnson, 2010; Johnson et al., 2004; Kardos & Johnson, 2010).
Teachers instructional teams are another common structure designed to support collaboration and school improvement. Many schools rely on teams as the key to school improvement, yet instituting teams does not necessarily lead to better schools (Rosenholtz, 1989; Troen & Boles, 2011). Teams vary widely in quality, and without proper training and support from veteran teachers and an effective principal, they often fail. Recently, Charner-Laird and colleagues (2013) interviewed teachers in six high-poverty urban schools and concluded that teachers found their work on teams valuable when it was congruent with their individual needs as classroom teachers and with a larger, meaningful agenda for improvement within their program or school. However, when teams failed to provide both, teachers felt frustrated about their formal collaborative work and tended to withdraw.
A Shared Set of Professional Goals and Purposes
In schools where teachers lack a common set of goals, they are far less likely to collaborate (Rosenholtz, 1989). In Chicago, Allensworth and her team (2009) found that teachers prefer to work in schools where their colleagues are committed to innovation, share a can-do attitude (p. 25), and take responsibility for bettering the school at-large (p. 30). When teachers see their coworkers as uncooperative and resistant to change (p. 30), they leave. Other scholars have documented similar findings (Cochran-Smith, et al., 2012; Johnson & Birkeland, 2003; Kardos & Johnson, 2007).
In high-poverty schools, teachers of all racial and economic backgrounds often say they have chosen to teach at their school precisely because they are motivated to help low-income, minority students succeed (Achinstein & Ogawa, 2011; Cochran-Smith et al., 2012; Johnson et al., 2014; Kraft et al., 2013; Love & Krueger, 2005; Quartz, 2003; Tamir, 2009; Yarrow, 2009). Teachers also care whether their colleagues share a perspective on social justice (Achinstein & Ogawa, 2011; Cochran-Smith, et al., 2012). Achinstein and Ogawa (2011) reported that the 21 teachers of color they interviewed wanted to work with colleagues who shared their humanistic commitments (p. 19) to teaching in low-income and minority communities. When they left their schools, they frequently cited a lack of multicultural capital (low expectations for, or negative attitudes about, students of color and lack of support for culturally responsive or socially just teaching) (p. 74) as a chief reason for doing so.
School culture is the ephemeral, taken for granted aspect of school (Deal & Peterson, 2009, p. 7), including prevailing norms and values that are expressed through individuals practices and behaviors. A strong, positive school culture reinforces the sense of community and social trust necessary for school improvement (Bryk & Schneider, 2002; Deal & Peterson, 2009). Not surprisingly, strong school culture is also linked to increased teacher retention.
Components of school culture are difficult to disentangle from other elements of school context. For example, principals influence teachers professional relationships and in doing so shape the schools norms for how work is done. The quality of teachers experiences as they work with one another contributes to the overall sense of what matters in the school. Because school culture overlaps with so many other factors, researchers have not yet agreed on a specific definition of school cultureas evidenced in the studies we cite here. For example, Johnson et al. (2012) studied school culture as a predictor of teacher satisfaction and teacher turnover. They defined it as the extent to which the school environment is characterized by mutual trust, respect, openness, and commitment to student achievement (p. 14). Others have investigated specific components of school culture (for example, order and discipline, parental engagement, or student satisfaction) as independent predictors of teacher turnover. We already have addressed some aspects of school culture, such as the principals responsiveness to teachers ideas and teachers readiness to support new colleagues. Here, we focus on factors that involve teachers interactions with students and parentsbecause repeatedly, norms of student discipline and parental involvement surface as powerful drivers of teacher turnover.
Teachers leave schools where a lack of student discipline impedes their ability to teach (Allensworth et al., 2009; Johnson et al., 2005; Ladd, 2011; Marinell & Coca, 2013). This effect has been found in high-poverty schools (Allensworth et al., 2009; Ingersoll, 2004), among novice teachers (Kapadia et al., 2007; Kukla-Acevedo, 2009), and among White teachers (Ingersoll & May, 2011). Disruptive classroom behavior frustrates teachers, especially in high schools, and teacher turnover is strongly predicted by students perceptions of their peers behavior (Allensworth et al., 2009). School-level decorum also affects teachers instruction and influences their decisions to stay or leave. Teachers want to be in schools that have strong schoolwide norms for behavior and consistent discipline policies (Johnson & Birkeland, 2003; Kraft et al., 2013; Marinell & Coca, 2013). It is also important to teachers that they teach in environments where adults and students feel safe (Allensworth et al., 2009; Public Agenda, 2004). However, problematic behavior is not limited to violence or disruption. Teachers also may leave in response to apathy or disengagement that undermines their teaching and leaves them feeling unsuccessful (Metz, 1993; Public Agenda, 2004; Steinberg, 1996).
At high-poverty schoolsespecially those serving large portions of minority studentsreports of student disciplinary infractions are far more common than at wealthier schools (Gregory, Skiba, & Noguera, 2010). The reasons for this are complicated. Scholars recognize that experiencing poverty detrimentally affects student behavior in school (Anyon, 1995; Bowen & Bowen, 1999; Kraft et al., 2013) and that high-poverty schools often lack the strong leadership and other resources for coping with disruptive student behaviors. However, they also report that low-income, minority students are far more likely to be punished than their non-poor, White counterparts who commit similar infractions (A. A. Ferguson, 2001; Gregory & Weinstein, 2008; Skiba, Michael, Nardo, & Peterson, 2002). This may result in part from sociocultural differences between teachers and students. Teachers report that their work is more difficult when they do not share social expectations or similar characteristics, including race, ethnicity, and language, with their students (Johnson & Birkeland, 2003; Marinell & Coca, 2013).
In high-poverty schoolswhere the overwhelming majority of teachers are White and middle class (Ingersoll & May, 2011)teachers often express frustration that their training did not prepare them to work in communities so different from their own (McLaughlin & Talbert, 2001). Notably, however, teachers of all backgrounds report struggling with student discipline and needing administrative support for handling disciplinary situations (Marinell & Coca, 2013). Across New York City middle school teachers, 75% of those considering departure indicated student discipline problems and/or lack of student motivation (p. 28) as frustrations that might compel them to leave their schools (p. 30). In addition, Marinell and Coca (2013) explained that student behavior both influences and is influenced by teacher turnover. At one school, teachers felt that students neither respected nor trusted incoming teachers. They had been demoralized by teachers repeated departures. One teacher explained, The kids see that the stronger teachers leave, and it does something to the overall tone of the building as far as the students behavior, and that affects teachers that stay (p. 33).
A schools ongoing relationships with parents also contribute to teachers satisfaction with working conditions (Allensworth et al., 2009; Loeb et al., 2005). Parents influence teachers commitment to their school (Henkin & Holliman, 2009) and predict turnover at all school levels (Allensworth et al., 2009). At a minimum, teachers count on parents to ensure that their children attend school ready to learn each day (Bryk & Schneider, 2002; Bryk et al., 2010; Kraft et al., 2013). Teachers also expect parents to engage in joint problem solving about student behavior (Bryk et al., 2010, p. 58). Importantly, this type of parentteacher interaction is far more predictive of teacher retention than are other forms of parent engagement, such as helping with their childs schoolwork (Allensworth et al., 2009).
Parentteacher interactions are shaped by school structures that foster communication between teachers and parents and enable parents to participate in the school community (Achinstein & Ogawa, 2011; Allensworth et al., 2009; Bryk & Schneider, 2002). Kraft et al. (2013) explained that teachers in high-poverty schools understand that challenges related to poverty may inhibit parents from active involvement in their childrens education. Some teachers view these challenges as insurmountable and enormously frustrating, but others feel supported by schoolwide parent engagement efforts. For example, at some schools in Kraft and colleagues (2013) study, teachers traveled to meet with parents, or family coordinators offered special training seminars on childrens socioemotional needs. Bryk and Schneider (2002) explained that engaging parents with the school is essential for developing trust between teachers and parents in low-income schools where power asymmetry (p. 28) can impede productive homeschool relationships.
NEXT STEPS: IMPROVING WORKING CONDITIONS AND RETENTION IN HIGH-POVERTY SCHOOLS
DIRECTIONS FOR FUTURE RESEARCH
The six major studies discussed earlier (Allensworth et al., 2009; Boyd et al., 2011; Johnson et al., 2012; Ladd, 2011; Loeb et al., 2005; Marinell & Coca, 2013) clearly establish the importance of the school organization in teachers decisions about whether to remain in high-poverty schools. They lay the groundwork for future research by identifying themes in teachers preferences and highlighting patterns in teachers career decisions. This body of work is complemented by an informative but limited group of studies exploring how the particularly influential predictors of turnover in high-poverty schoolsadministrators, colleagues, and aspects of school cultureinfluence teachers decisions about whether to stay at their schools. However, far more researchboth qualitative and quantitativeis needed to explain how these components of the school environment affect retention.
Much of the research on teacher retention draws on large quantitative survey-based data sets that illuminate the elements of school context that are most likely to drive teachers out of high-poverty schools. Although the survey instruments vary, they all primarily rely on a multiple-choice format that asks teachers to interpret complex ideas and respond within a limited set of choices, thus restricting researchers ability to probe teachers responses more deeply. For instance, several surveys ask teachers to consider the importance of colleagues, but none specifies a particular aspect of teachers collegial relationships or even defines who is included in this category. More qualitative research might complement and inform the existing survey research. For example, in a school where novices report that they stay because of the support they receive from their colleagues, researchers could gain insights into precisely how those colleagues influence one anothers decisions by interviewing novice teachers, observing grade team meetings, and tracking the everyday activities of new teachers mentors. Such an iterative use of quantitative and qualitative methods would also enable researchers to build more precise research instrumentssuch as better surveys or more focused interview protocolsfor future studies.
Second, existing research unveils important information about how subgroups of teachers respond to different aspects of the work environment. For example, the concerns of teachers who share background characteristics with their students appear to differ somewhat from the concerns of teachers who have had limited encounters with a particular ethnic, racial, or socioeconomic group. Because Black teachers are more likely to stay at schools serving Black students, it is crucial that researchers determine what aspects of the work environment are especially important to Black teachers. Similarly, because the overwhelming majority of teachers in public schools are White women, exploring the types of supports that keep White women teaching in high-poverty schools would likely benefit their students.
Third, although we discussed three social working conditions (school leadership, collegiality, and school culture) separately in an effort to disentangle the effects of each component, these three working conditions do not function independently. Future research should consider how these components are related and whether some aspects of school context serve as protective factors against others. For example, in schools where collegial relationships are especially strong and student discipline is effective and consistent, does the importance of the principal diminish? In schools where the principal is regarded as especially fair, does trust among teachers become less important to them? Are there schools where parent involvement is strong but the quality of the principal is weakor does parent involvement always depend on a strong principal? Intensive case studies could illuminate these relationships and set the stage for further quantitative analysis.
Finally, very few of the existing studies track teachers or schools over time. Longitudinal research that examines the same environment over time or that follows teachers responses to interventions over several years would illuminate which strategies show promise for reducing or alleviating unnecessary turnover. For example, extended case studies of high-poverty schools where student achievement is high and turnover rates are low could provide insights into the types of initiatives that keep teachers in their school. Detailed state-level databases, which currently are expanding rapidly, make it possible to track changes in school leadership and their influence on teacher turnover over time. For example, they can suggest what transpires for teachers when a new principal, with a particular history of preparation and work experience, replaces another.
IMPLICATIONS FOR POLICY AND PRACTICE
Policy makers and practitioners, who have recently focused more on selecting, assessing, and retaining individual teachers than on improving their workplace, must recognize the important role that teachers working environments play in their retention and in their students learning. The research suggests that building and sustaining strong work environments should be central to every districts school improvement strategy. But the type of working conditions that retain teachers in their high-poverty schools are very unlikely to develop in response to state or district mandates or inducements targeted toward individuals alone. Creating school environments where teachers can offer the transformative pedagogy that will prepare historically underserved students for college and careers is extraordinarily difficult workand it is highly dependent on the collective capacities of school-based practitioners. Therefore, to ensure that policies translate into desired practices, policy makers must focus on capacity-building (McDonnell & Elmore, 1987) in schools, themselves, and in the district and state offices intended to support the schools work. Although we recognize that far more research is needed to fully understand teacher turnover and how to address it, the findings from numerous robust studies can inform immediate action. Here, we will discuss a few broad approaches and highlight some considerations for implementation.
Although factors such as salary and work hours matter to teachers, working conditions that are social in nature likely supersede marginal improvements to pay or teaching schedules in importance. Teachers who leave their schools routinely report dissatisfaction with their administration as a chief reason. Therefore, improving the caliber of principals in high-poverty schools would be a high-leverage approach for districts intent on retaining teachers. When appointing and assigning principals, districts might consider the qualities that teachers in high-poverty schools say they seek in a leader: effective management, fair and encouraging leadership, instructional support, and inclusive decision-making. Rather than assigning new principals with little demonstrated success to head these schools, district officials would be wise to encourage their most effective principals to lead them. Principal preparation and professional development programs should focus on the managerial, social, instructional, and political skills that school leaders will need to succeed in high-poverty schools. Experienced principals who are struggling should be given ongoing support but be dismissed if they fail to meet the difficult demands of school leadership today.
It is important to recognize that turnover patterns among principals mirror those of teachers. Like teachers, novice principals are often placed in schools serving poor, minority, and low-achieving students, and as principals gain experience, they tend to either leave the profession or transfer to schools with more favorable working conditionsand, not surprisingly, fewer disadvantaged students (Loeb et al., 2010). As with teachers, many aspects of their environment influence principals retention. Research on principal turnover is thin, but policy makers should pay special attention to the current conception of the principals roleespecially in the many high-poverty schools that are under mandate to improve quickly. Too often, principals in such schools carry the enormous burden of being singularly responsible for the success of their school, making their position both undesirable and unsustainable. Policy makers might increase support and distribute responsibilities among district and school-based personnel so that principals are not left fully accountable for improving their struggling schools (Bryk et al., 2010).
In high-poverty schools, principals and school-level teams might be given the final say in hiring decisions so that they can cultivate teams of teachers who have shared goals and purposes as well as the collective skill set needed to get the work done. Districts might support principals efforts to attract a strong staff by redefining the role of their human resources departments, reworking district hiring timelines, and creating teacher-leader positions. Positions for teacher leaders would allow principals to strategically distribute some of their responsibilities and decision-making authority to outstanding teachers. Similarly, if principals of high-poverty schools were granted control of their budget, they could hire support staff, such as social workers, college advisors, and parent coordinators, to complement the expertise of their teaching staff and buffer students from the effects of poverty. This arrangement could be complemented by the services of a district-level partnership office that helps schools develop relationships with community-based organizations, such as healthcare agencies and extracurricular programs. Districts might grant high-poverty schools additional funds to hire operations managers so that principals could focus on instruction, students, and parents. If high-poverty schools offered strong educators unmatched opportunities to develop their skills as leaders, perhaps promising principals would choose to work in them, and in turn, effective and dedicated teachers would follow.
To increase teachers opportunities for collegial interaction, districts might rethink the portion of their time that teachers spend teaching or engaging in administrative duties. In high-performing countries, where schools do not struggle with turnover, teachers spend considerably less time delivering instruction and more time planning lessonsoften collaboratively with colleagues and curriculum coaches. They also spend more time conferring with parents, students, and other support staff about the needs of individual students (Darling-Hammond, 2010; Tucker, 2011). To address the challenges posed by student behavior in high-poverty schools, districts might invest in initiatives that help schools prevent and resolve discipline problems more effectively (Wadhwa, 2015). As part of this work, policy makers might also allocate additional funding to high-poverty schools to ensure that trained professionals focus on students complex socioemotional needs.
As noted earlier, most high-poverty schools routinely struggle with problems caused by high rates of turnover. Administrators and teachers in these schools require space and time to adapt promising strategiessuch as school-site hiring, teacher teams, and differentiated roles for teacher leadersto suit the unique needs of their school. The capacity-building processes of developing trust, norms, and a positive school culture are often unpredictable, and they look different in different school communities. Principals and teachers need assurance that their serious, though failed, efforts will not result in sudden school closings or job terminations triggered by impatient district or state officials. Principals and teachers need to know that their improvement efforts will be supported by seasoned experts who share responsibility for getting things right. If districts and schools are going to be held accountable for improving working conditions and, ultimately, for reducing teacher turnover, their leaders must be assured of what Elmore (2000) called reciprocity of accountability and capacity (p. 21): For every increment of performance I demand from you, I have an equal responsibility to provide you with the capacity to meet that expectation (Elmore, 2000, p. 5). In other words, before policy makers hold practitioners accountable for stabilizing the teacher workforce, principals and teachers in high-poverty schools need the tools and time necessary to do that work.
Nearly four decades ago, Lortie (1975) wrote that most teachers enter the profession because they want to make a difference in the lives of their students. This remains true today, especially among teachers working with historically underserved students in high-poverty schools. However, too often, when teachers find that their school environments thwart their efforts to make education the great equalizer (Mann, 1848), they leave. Using a range of research methods and definitions of working conditions, researchers have repeatedly found that when teacher turnover is analyzed from an organizational perspective, the poor working conditions common in Americas neediest schools explain away most, if not all, of the relationship between student characteristics and teacher attrition. This is important because unlike demographic characteristics of students, working conditions can be changed. Policy makers and practitioners possess many options for improving aspects of the school environment, and although more research can inform this work, much is already known about what matters to teachers as they decide whether to continue teaching in their schools.
1. To select the studies in this review, we conducted an extensive electronic search of scholarship on teacher turnover through Harvard Universitys library portal. We also carefully scanned the reference sections of identified studies to ensure a comprehensive search. We read over 60 studies and, along with members of our research team, analyzed each studys research methods. We selected studies that employed a rigorous research design and included measures of both student demographics and elements of working conditions as predictors of turnover.
Achinstein, B., & Ogawa, R. (2011). Change(d) agents: New teachers of color in urban schools. New York, NY: Teachers College Press.
Achinstein, B., Ogawa, R. T., Sexton, D., & Freitas, C. (2010). Retaining teachers of color: A pressing problem and a potential strategy for hard-to-staff schools. Review of Educational Research, 80(1), 71107. doi:10.3102/0034654309355994
Allensworth, E., Ponisciak, S., & Mazzeo, C. (2009). The schools teachers leave: Teacher mobility in Chicago Public Schools. Chicago, IL: Consortium on Chicago School ResearchUniversity of Chicago.
Alliance for Excellent Education. (2008). What keeps good teachers in the classroom? Understanding and reducing teacher turnover (pp. 19). Retrieved from http://www.all4ed.org/files/TeachTurn.pdf
Anyon, J. (1995). Race, social class, and educational reform in an inner-city school. Teachers College Record, 97, 6994.
Auguste, B. G., Kihn, P., & Miller, M. (2010). Closing the talent gap: Attracting and retaining top-third graduates to careers in teaching: An international and market research-based perspective. McKinsey. Retrieved from http://mckinseyonsociety.com/closing-the-talent-gap/
Balu, R., Beteille, T., & Loeb, S. (2009-2010). Examining teacher turnover: The role of school leadership. Politique Americaine, 15, 5579.
Barnes, G., Crowe, E., & Schaefer, B. (2007). The cost of teacher turnover in five school districts. Washington, DC: National Commission on Teaching and Americas Future.
Beteille, T., Kalogrides, D., & Loeb, S. (2011). Effective schools: managing the recruitment, development, and retention of high-quality teachers (Working Paper No. 17177). Cambridge, MA: National Bureau of Economic Research.
Boe, E. E. (2006). Long-term trends in the national demand, supply, and shortage of special education teachers. Journal of Special Education, 40(3), 138150. doi:10.1177/00224669060400030201
Borman, G. D., & Dowling, N. M. (2008). Teacher attrition and retention: A meta-analytic and narrative review of the research. Review of Educational Research, 78(3), 367409. doi:10.3102/0034654308321455
Bowen, N. K., & Bowen, G. L. (1999). Effects of crime and violence in neighborhoods and schools on the school behavior and performance of adolescents. Journal of Adolescent Research, 14(3), 319342. doi:10.1177/0743558499143003
Boyd, D., Grossman, P., Ing, M., Lankford, H., Loeb, S., & Wyckoff, J. (2011). The influence of school administrators on teacher retention decisions. American Educational Research Journal, 48(2), 303333.
Boyd, D., Grossman, P., Lankford, H., Loeb, S., & Wyckoff, J. (2006). How changes in entry requirements alter the teacher workforce and affect student achievement. Education Finance and Policy, 1(2), 176216.
Boyd, D., Lankford, H., Loeb, S., & Wyckoff, J. (2005). Explaining the short careers of high-achieving teachers in schools with low-performing students. American Economic Review, 95(2), 166171.
Brown, K., & Wynn, S. (2009). Finding, supporting, and keeping: The role of the principal in teacher retention issues. Leadership and Policy in Schools, 8(1), 3763. doi:10.1080/15700760701817371
Bryk, A. S., & Schneider, B. L. (2002). Trust in schools: A core resource for school reform. New York, NY: Russell Sage Foundation Publications.
Bryk, A. S., Sebring, P. B., Allensworth, E., Easton, J. Q., & Luppescu, S. (2010). Organizing schools for improvement: Lessons from Chicago. Chicago, IL: University of Chicago Press.
Carroll, S., Reichardt, R., Guarino, C., & Mejia, A. (2000). The distribution of teachers among Californias school districts and schools. Santa Monica: RAND. Retrieved from http://www.rand.org/pubs/monograph_reports/MR1298z0
Charner-Laird, M., Ng, M., Johnson, S. M., Kraft, M., Papay, J. P., & Reinhorn, S. K. (2013). Teachers instructional teams in highpoverty, urban schools: Boon or burden? Unpublished manuscript.
Chenoweth, K. (2009). How its being done: Urgent lessons from unexpected schools. Cambridge, MA: Harvard Education Press.
Clotfelter, C. T., Ladd, H. F., & Vigdor, J. (2005). Who teaches whom? Race and the distribution of novice teachers. Economics of Education Review, 24(4), 377392.
Clotfelter, C. T., Ladd, H. F., & Vigdor, J. L. (2011). Teacher mobility, school segregation, and pay-based policies to level the playing field. Education Finance and Policy, 6(3), 399438.
Clotfelter, C. T., Ladd, H. F., Vigdor, J. L., & Wheeler, J. (2007). High-poverty schools and the distribution of teachers and principals. North Carolina Law Review, 85(1348).
Cochran-Smith, M., McQuillan, P., Mitchell, K., Terrell, D. G., Barnatt, J., DSouza, . . . Gleeson, A. (2012). A longitudinal study of teaching practice and early career decisions: A cautionary tale. American Educational Research Journal, 49(5), 844880.
Darling-Hammond, L. (2001). The right to learn: A blueprint for creating schools that work. San Francisco, CA: Jossey-Bass.
Darling-Hammond, L. (2010). The flat world and education: How Americas commitment to equity will determine our future. New York, NY: Teachers College Press.
Deal, T., & Peterson, K. (2009). Shaping school culture: Pitfalls, paradoxes, and promises (2nd ed.). San Francisco: Jossey-Bass.
Deangelis, K. J., & Presley, J. B. (2007). Leaving schools or leaving the profession: Setting Illinois record straight on new teacher attrition. Edwardsville: Illinois Education Research Council.
Dee, T. S. (2004). Teachers, race and student achievement in a randomized experiment. Review of Economics and Statistics, 86(1), 195210.
Dee, T. (2005). A teacher like me: Does race, ethnicity, or gender matter ? American Economic Review, 95(2), 158–165.
Donaldson, M. L., & Johnson, S. M. (2010). The price of misassignment: The role of teaching assignments in Teach For America teachers’ exit from low-income schools and the teaching profession. Educational Evaluation and Policy Analysis, 32(2), 25.
Donaldson, M. L., & Johnson, S. M. (2011). Teach For America teachers: How long do they teach? Why do they leave? Phi Delta Kappan, 93(2), 4751.
Downey, D. B., Von Hippel, P. T., & Hughes, M. (2008). Are failing schools really failing? Using seasonal comparison to evaluate school effectiveness. Sociology of Education, 81(3), 242270. doi:10.1177/003804070808100302
Drury, D., & Baer, J. (2011). The American public school teacher: Past, present, and future. Cambridge, MA: Harvard Education Press.
Elmore, R. F. (2000b). Bridging the gap between standards and achievement in education. Washington, DC: Albert Shanker Institute.
Ferguson, A. A. (2001). Bad boys: Public schools in the making of Black masculinity. Ann Arbor: University of Michigan Press.
Ferguson, R. (2009). How high schools become exemplary. Cambridge, MA: Achievement Gap Initiative.
Glazerman, S., Isenberg, E., Dolfin, S., Bleeker, M., Johnson, A., Grider, M., & Jacobus, M. (2010). Impacts of comprehensive teacher induction. Final results from a randomized controlled study. Washington, DC: Institute of Educational Sciences.
Goldhaber, D., Choi, H.-J., & Cramer, L. (2007). A descriptive analysis of the distribution of NBPTS-certified teachers in North Carolina. Economics of Education Review, 26(2), 160172. doi:10.1016/j.econedurev.2005.09.003
Goldhaber, D., & Hansen, M. (2009). National Board Certification and teachers career paths: Does NBPTS certification influence how long teachers remain in the profession and where they teach? Education Finance and Policy, 4(3), 229262.
Gregory, A., Skiba, R. J., & Noguera, P. (2010). The achievement gap and the discipline gap: Two sides of the same coin? Educational Researcher, 39(1), 5968. doi:10.3102/0013189X09357621
Gregory, A., & Weinstein, R. S. (2008). The discipline gap and African Americans: Defiance or cooperation in the high school classroom. Journal of School Psychology, 46(4), 455475.
Grissom, J. A. (2011). Can good principals keep teachers in disadvantaged schools? Linking principal effectiveness to teacher satisfaction and turnover in hard-to-staff environments. Teachers College Record, 113(11), 25522585.
Grissom, J. A., & Keiser, L. R. (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), 557580.
Grissom, J. A., & Loeb, S. (2011). Triangulating principal effectiveness. American Educational Research Journal, 48(5), 10911123.
Guarino, C. M., Santibanez, L., & Daley, G. A. (2006). Teacher recruitment and retention: A review of the recent empirical literature. Review of Educational Research, 76(2), 173208.
Guin, K. (2004). Chronic teacher turnover in urban elementary schools. Education Policy Analysis Archives, 12(42), 130.
Hanselman, P., Grigg, J., Bruch, S., & Gamoran, A. (2013). The consequences of principal and teacher turnover for school social resources. Unpublished working paper.
Hanushek, E. A., Kain, J. F., & Rivkin, S. G. (2004). Why public schools lose teachers. Journal of Human Resources, 39(2), 326354.
Hemphill, C., & Nauer, K. (2009). The new marketplace: How small-school reforms and school choice have reshaped New York Citys high schools. New York, NY: Center for New York City Affairs, The New School for Management and Urban Policy.
Henkin, A. B., & Holliman, S. L. (2009). Urban teacher commitment. Urban Education, 44(2), 160180.
Ingersoll, R. M. (2001). Teacher turnover and teacher shortages: An organizational analysis. American Educational Research Journal, 38(3), 499534.
Ingersoll, R. M. (2002). Out-of-field teaching, educational inequality, and the organization of schools: An exploratory analysis. Center for the Study of Teaching and Policy, University of Washington.
Ingersoll, R. M. (2003). Is there really a teacher shortage? A research report. Center for the Study of Teaching and Policy. Retrieved from http://www.ctpweb.org
Ingersoll, R. M. (2004). Why do high-poverty schools have difficulty staffing their classrooms with qualified teachers? Washington DC: Center for American Progress and Institute for Americans Future.
Ingersoll, R. M., & May, H. (2011). Recruitment, retention and the minority teacher shortage. Philadelphia, PA: Consortium for Policy Research in Education.
Ingersoll, R. M., & Merrill, L. (2012). Seven trends: The transformation of the teaching force. Philadelphia, PA: Consortium for Policy Research in Education.
Johnson, S. M. (1990). Teachers at work : Achieving success in our schools. New York, NY: Basic Books.
Johnson, S. M. (2006). The workplace matters: Teacher quality, retention, and effectiveness (Working paper). Washington, DC: National Education Association Research Department.
Johnson, S. M., Berg, J. H., & Donaldson, M. L. (2005). Who stays in teaching and why? A review of the literature on teacher retention. Washington, DC: National Retired Teachers Association.
Johnson, S. M., & Birkeland, S. E. (2003). Pursuing a sense of success: New teachers explain their career decisions. American Educational Research Journal, 40(3), 581617.
Johnson, S. M., Kardos, S. M., Kauffman, D., Liu, E., & Donaldson, M. L. (2004). The support gap: New teachers experiences in high-income and low-income schools. Education Policy Analysis Archives, 12, 25.
Johnson, S. M., Kraft, M., & Papay, J. P. (2012). How context matters in high-need schools: The effects of teachers working conditions on their professional satisfaction and their students achievement. Teachers College Record, 114(10), 139.
Johnson, S. M., & The Project on the Next Generation of Teachers. (2004). Finders and Keepers: Helping new teachers survive and thrive in our schools. San Francisco, CA: Jossey-Bass.
Johnson, S. M., Reinhorn, S., Charner-Laird, M., Kraft, M., Ng, M., & Papay, J. P. (2014). Ready to lead, but how? Teachers experiences in high-poverty urban schools. Teachers College Record, 116(10), 150.
Kapadia, K., Coca, V., & Easton, J. Q. (2007). Keeping new teachers: A first look at the influences of induction in the Chicago Public Schools. Chicago, IL: Consortium on Chicago School Research, University of Chicago.
Kardos, S. M., & Johnson, S. M. (2007). On their own and presumed expert: New teachers' experience with their colleagues. Teachers College Record, 109(9), 24.
Kardos, S. M., & Johnson, S. M. (2010). New teachers experiences of mentoring: The good, the bad, and the inequity. Journal of Educational Change, 11(1), 2344.
Kardos, S. M., Johnson, S. M., Peske, H. G., Kauffman, D., & Liu, E. (2001). Counting on colleagues: New teachers encounter the professional cultures of their schools. Educational Administration Quarterly, 37(2), 250290.
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), 4766. doi:10.3102/01623737021001047
Kraft, M., Papay, J. P., Charner-Laird, M., Johnson, S. M., Ng, M., & Reinhorn, S. K. (2013). Committed to their students but in need of support: How school context influences teacher turnover in high-poverty, urban schools. Unpublished manuscript.
Kukla-Acevedo, S. (2009). Leavers, movers, and stayers: The role of workplace conditions in teacher mobility decisions. Journal of Educational Research, 102(6), 443452.
Ladd, H. F. (2011). Teachers perceptions of their working conditions. Educational Evaluation and Policy Analysis, 33(2), 235261.
Leukens, M. T., Lyter, D. M., Fox, E. E., & Chandler, K. (2004). Teacher attrition and mobility: Results from the Teacher Follow-Up Survey, 2000-1. Washington, DC: National Center for Education Statistics.
Little, J. W. (1982). Norms of collegiality and experimentation : Workplace conditions of school success, 19(3), 325–340.
Liu, E., Rosenstein, J. G., Swan, A. E., & Khalil, D. (2008). When districts encounter teacher shortages: The challenges of recruiting and retaining mathematics teachers in urban districts. Leadership & Policy in Schools, 7(3), 296323. doi:10.1080/15700760701822140
Loeb, S., Darling-Hammond, L., & Luczak, J. (2005). How teaching conditions predict teacher turnover in California schools. Peabody Journal of Education, 80(3), 4470. doi:10.1207/s15327930pje8003_4
Loeb, S., Kalogrides, D., & Horng, E. L. (2010). Principal preferences and the uneven distribution of principals across schools. Educational Evaluation and Policy Analysis, 32(2), 205229.
Lortie, D. (1975). Schoolteacher: A sociological study. Chicago, IL: University of Chicago Press.
Love, A., & Krueger, A. C. (2005). Teacher beliefs and student achievement in urban schools serving African American students. Journal of Educational Research, 99(2), 8798.
Mann, H. (1868). Life and works of Horace Mann (Vol. 3). Boston, MA: Walker, Fuller and Company.
Mapp, K. (2003). Having their say: Parents describe why and how they are engaged in their childrens learning. School Community Journal, 13(1), 3564.
Marinell, W. H., & Coca, V. M. (2013). Who stays and who leaves? Findings from a three part study of teacher turnover in NYC middle schools. New York, NY: Research Alliance for NYC Schools.
McCaffrey, D., Koretz, D., Lockwood, J. R., & Hamilton, L. (2004). Evaluating value-added models for teacher accountability. Santa Monica, CA: RAND.
McDonnell, L. M., & Elmore, R. F. (1987). Getting the job done: Alternative policy instruments. Educational Evaluation and Policy Analysis, 9(2), 133152.
McLaughlin, M. W., & Talbert, J. E. (2001). Professional communities and the work of high school teaching. Chicago, IL: University of Chicago Press.
Metz, M. H. (1993). Teachers ultimate dependence on their students. In J. W. Little & M. W. McLaughlin (Eds.), Teachers work: Individuals, colleagues, and contexts (pp. 104136). New York, NY: Teachers College Press.
Milanowski, A. T., & Odden, A. R. (2007). A new approach to the cost of teacher turnover. Washington, DC: School Finance Redesign Project, Center on Reinventing Public Education.
Mitra, D. L. (2008). Student voice in school reform: Building youth-adult partnerships that strengthen schools and empower youth. Albany: SUNY Press.
Murnane, R. J., Singer, J. D., Willett, J. B., Kemple, J. J., & Olsen, R. J. (1991). Who will teach? Policies that matter. Cambridge, MA: Harvard University Press.
Neild, R. C., Useem, E., Travers, E., & Lesnick, J. (2003). Once and for all: placing a highly qualitified teacher in every Philadelphia classroom. Philadelphia, PA: Research for Action.
Ost, B. (2014). How do teachers improve? The relative importance of general and specific human capital. American Economic Journal: Applied Economics, 6(2), 127151.
Papay, J. P., & Johnson, S. M. (2011). Is PAR a good investment? Understanding the costs and benefits of teacher Peer Assistance and Review programs. Educational Policy, 26(5), 696729. doi:10.1177/0895904811417584
Public Agenda. (2004). Teaching interrupted: Do discipline policies in todays public schools foster the common good ? Retrieved from http://www.publicagenda.org/files/pdf/teaching_interrupted.pdf
Quartz, K. H. (2003). “Too angry to leave”: Supporting new teachers’ commitment to transform urban schools. Journal of Teacher Education, 54(2), 99–111.
Raymond, M., & Fletcher, S. (2002). The Teach For America evaluation. Education Next, 2(1), 62–68.
Raymond, M. E., Fletcher, S. H., & Luque, J. A. (2001). Teach For America : An evaluation of teacher differences and student outcomes in Houston, Texas. Palo Alto, CA: CREDO.
Reinhorn, S. (2013). Teachers experiences of teacher evaluation in six high-poverty urban schools. Unpublished manuscript.
Reininger, M. (2012). Hometown disadvantage? It depends on where youre from. Teachers location preferences and the implications for staffing schools. Educational Evaluation and Policy Analysis, 34(2), 127145.
Rice, J. K. (2013). Learning from experience? Evidence on the impact and distribution of teacher experience and the implications for teacher policy. Education Finance and Policy, 332348. doi:10.1162/EDFP
Rivkin, S. G., Hanushek, E. A., & Kain, J. F. (2005). Teachers, schools, and academic achievement. Econometrica, 73(2), 417458.
Rockoff, J. (2004). The impact of individual teachers on student achievement: Evidence from panel data. American Economic Review, 94(2), 247252.
Ronfeldt, M., Loeb, S., & Wyckoff, J. (2013). How teacher turnover harms student achievement. American Educational Research Journal, 50(1), 436. doi:10.3102/0002831212463813
Rosenholtz, S. (1989). Teachers workplace: The organizational context of schooling. New York, NY: Teachers College Press.
Sanders, W. L., & Rivers, J. C. (1996). Cumulative and residual effects of teachers on future student academic achievement. Retrieved from http://news.heartland.org/sites/all/modules/custom/heartland_migration/files/pdfs/3048.pdf
Scafidi, B., Sjoquist, D. L., & Stinebrickner, T. R. (2007). Race, poverty, and teacher mobility. Economics of Education Review, 26(2), 145159.
Shen, J. (1997). Teacher retention and attrition in public schools: Evidence from SASS91. Journal of Educational Research, 91(2), 8188.
Skiba, R. J., Michael, R. S., Nardo, A. C., & Peterson, R. L. (2002). The color of discipline: Sources of racial and gender disproportionality in school punishment. The Urban Review, 34(4), 317342.
Spillane, J. P., Hallett, T., & Diamond, J. B. (2003). Forms of capital and the construction of leadership: Instructional leadership in urban elementary schools. Sociology of Education, 76(1), 117.
Steinberg, L. (1996). Beyond the classroom: Why school reform has failed and what parents need to do. New York, NY: Simon and Schuster.
Supovitz, J. A., & Poglinco, S. M. (2001). Instructional leadership in a standards-based reform. Philadelphia, PA: Consortium for Policy Research in Education.
Talbert, J. E., & McLaughlin, M. W. (1996). Teacher professionalism in local school contexts. In I. F. Goodson & A. Hargreaves (Eds.), Teachers professional lives (pp. 127154). London, England: Falmer Press.
Tamir, E. (2009). Choosing to teach in urban schools among graduates of elite colleges. Urban Education, 44(5), 522–544.
TNTP. (2012). Understanding the real retention crisis in America’s urban schools. Retrieved from http://tntp.org/publications/view/the-irreplaceables-understanding-the-real-retention-crisis
Troen, V., & Boles, K. (2011). The power of teacher teams. Thousand Oaks, CA: Corwin Press.
Tucker, M. (Ed.). (2011). Surpassing Shanghai. Cambridge, MA: Harvard Education Press.
Wadhwa, A. (2015). Restorative Justice in Urban Schools: Disrupting the School-to-Prison Pipeline. Routledge.
Warren, M. R. (2005). Communities and schools: A new view of urban education reform. Harvard Educational Review, 75(2), 133174.
Waters, T., Marzano, R. J., & McNulty, B. (2003). Balanced leadership: What 30 years of research tells us about the effect of leadership on student achievement. Aurora, CO: Mid-continent Research for Education and Learning.
Weiss, E. M. (1999). Perceived workplace conditions and first-year teachers morale, career choice commitment, and planned retention: a secondary analysis. Teaching and Teacher Education, 15(8), 861879.
Wolman, P. (Ed.). (2010). Status of the American public school teacher 20052006. (2010). Washington, DC: National Education Association.
Wright, S. P., Horn, S. P., & Sanders, W. L. (1997). Teacher and classroom context effects on student achievement: Implications for teacher evaluation. Journal of Personnel Evaluation in Education, 11(1), 5767.
Yarrow, A. L. (2009, October 19). State of mind. Education Week. Retrieved from http://www.edweek.org/ew/articles/2009/10/21/08publicagenda_ep.h29.html?r=1554997124