ABSTRACT Title of Dissertation: KNOWING WHAT TO DO: SCHOOL FOCUS, TEACHER MORALE, AND TEACHER TURNOVER William P. Bovender, Doctor of Philosophy, 2016 Dissertation directed by: Associate Professor Matthew J. Miller, Department of Counseling, Higher Education, and Special Education This study explored how schools’ focus – the collective perception among teachers of clarity and consistency regarding school goals, expectations, and priorities – related to schoolwide morale and school turnover rates. I examined the hypothesis that focus attenuates the deleterious effects of student misconduct on teacher morale and the contributory role of student misconduct leading to teacher turnover. In addition, I examined climate strength regarding perceptions school focus as an indicator of focus itself, as well a potential moderating effect of climate strength on the magnitude of school focus-school morale and school focus-turnover relationships. Data from a national sample of middle and high schools (N schools = 348, N teachers = 11,376) were analyzed using school-level multiple regression models. Schools with higher focus had significantly higher morale, independent of related perceptions of administrative leadership. No significant relationship was found between school focus and school turnover rates. The hypothesized moderating effect of focus on student misconduct and morale was not supported, though there was a significant indication that focus attenuated the positive relationship between student misconduct and turnover. Climate strength of school focus ratings significantly correlated with focus scores, but did not moderate relationships between focus and predicted outcomes. Findings suggest that school-level focus does represent a characteristic of schools that has a meaningful positive relationship with teacher morale but do not necessarily clarify how that relationship manifests in schools or if that relationship presents an avenue for intervention. KNOWING WHAT TO DO: SCHOOL FOCUS, TEACHER MORALE, AND TEACHER TURNOVER by William P. Bovender Dissertation submitted to the Faculty of the Graduate School of the University of Maryland, College Park, in partial fulfillment of the requirements for the degree of Doctor of Philosophy 2016 Advisory Committee: Associate Professor Matthew J. Miller, Chair Associate Professor Robert Croninger Clinical Assistant Professor Jill B. Jacobson Professor Dennis Kivlighan Associate Professor Emeritus William O. Strein © Copyright by William P. Bovender 2016 ii Table of Contents Table of Contents .......................................................................................................... ii List of Tables ............................................................................................................... iii Introduction ................................................................................................................... 1 Background ................................................................................................................... 5 When teachers leave ................................................................................................. 5 Keeping teachers and keeping them happy ............................................................... 6 Administrative support and leadership ................................................................... 10 Students as stressors ................................................................................................ 15 Roles, goals, and confusion .................................................................................... 17 From role clarity to school focus ............................................................................ 18 Present Study .............................................................................................................. 21 Hypotheses .................................................................................................................. 22 Methods....................................................................................................................... 26 Sample..................................................................................................................... 26 Measurement ........................................................................................................... 28 Data Analyses ......................................................................................................... 30 Results ......................................................................................................................... 36 Discussion ................................................................................................................... 40 Limitations .............................................................................................................. 45 Summary and conclusions ...................................................................................... 49 Appendices .................................................................................................................. 54 Appendix A: Sample characteristics and response rates ......................................... 54 Appendix B: Scale items and reliabilities ............................................................... 58 Appendix C: Descriptive statistics .......................................................................... 62 Bibliography ............................................................................................................... 65 iii List of Tables Table 1: Final Sample Sizes by Grade Level and Location Type ................................27 Table 2: Bivariate Correlations Between School Morale, Teacher Turnover Rates, and Predictor Variables ..................................................................................36 Table 3: Standardized Beta Coefficients and R-squared values in the School Morale as Outcome Models ...............................................................37 Table 4: Standardized Beta Coefficients and R-squared values in the School Turnover Rate as Outcome Models ..................................................37 Table 5: Standardized Beta Coefficients in the School Focus as Moderator Models............................................................................................................38 Table 6: Standardized Beta Coefficients in the Climate Strength as Moderator Models............................................................................................................39 Table 7: Proportion of Schools by Auspices................................................................54 Table 8: Proportion of Teachers by Age Range...........................................................54 Table 9: Proportion of Teachers by Ethnicity ..............................................................55 Table 10: Proportion of Teachers by Sex ....................................................................55 Table 11: Teacher Questionnaire School Participation Percentages by Grade Level and Location Type .............................................................55 Table 12: Principal Year 1 Questionnaire Response Percentages by Grade Level and Location Type .............................................................56 Table 13: Survey Participation Rates by School Auspices ..........................................57 Table 14: Teacher Organizational Focus Scale: Items and Reliability ........................58 Table 15: Teacher Classroom Disorder Scale: Items and Reliability ..........................59 Table 16: Administrator Leadership Scale: Items and Reliability ...............................60 Table 17: Teacher Morale Scale: Items and Reliability ...............................................61 Table 18: Descriptive Values for School Morale by Grade Level and Location Type .............................................................62 Table 19: Descriptive Values for School Focus by Grade Level and Location Type .............................................................62 Table 20: Descriptive Values for Classroom Disorder by Grade Level and Location Type .............................................................63 Table 21: Descriptive Values for Administrator Leadership by Grade Level and Location Type .............................................................63 Table 22: Descriptive Values for School Teacher Turnover Rate by Grade Level and Location Type .............................................................64 1 Introduction Teachers do not just teach. Depending on whom is asked, teachers’ mandated responsibilities, in addition to providing optimal instruction for each student, include maintaining consistent order and discipline; helping children master fundamental life skills, including self-care, self-advocacy, and social conventions; encouraging creativity and independent problem-solving; instilling an appreciation for lifelong learning; reducing the achievement gap between socio-economic classes and ethnic groups; preventing dropouts; promoting values of fairness and equity; and fostering an understanding of civic duty (Bierlein, 1993; Bovender, 2013; Theobald, 1990). Effectively meeting such a broad range of essential goals, which are prescribed by a multitude of stakeholders, and doing so with increasingly strained resources, is a challenging undertaking for schools and teachers. Maintaining schoolwide morale and preventing staff burnout in the face of that challenge may be just as difficult. Teachers’ workloads are also shifting to include the use of new technologies; more team-based lesson planning and program implementation; education of diverse populations with varying levels of English language mastery; and an increased emphasis on standardized data collection, documentation, and related paperwork (Valli & Buese, 2007). Furthermore, the ubiquitous pressures for school accountability and reform compound the scrutiny placed on schools and teachers (Conley & You, 2009; Ravitch, 2010). Policymakers emphasize all of these objectives without establishing clear plans to make their attainment possible, forcing schools into a precarious position that often results in superficial and nonsensical attempts to satisfy these imposed roles (Petersen, 1997). Reform and restructuring 2 initiatives are generally well-intentioned, but the pressure of added (and sometimes conflicting) policy demands is often confusing and discouraging for those teachers and school leaders who have to integrate shifting expectations into their already stressful occupations. Students and classroom compositions are changing as well. The U.S. recently passed the “majority minority” mark, meaning that there are now fewer White students in public K-12 schools nationwide than there are non-White students. According to government figures, between 2003 and 2013, the number of White students in public K-12 schools fell by more than 11%, the number of Black students fell by roughly 6%, the number of Hispanic students rose by nearly 40%, the number of Asian/Pacific Islander students rose by more than 20%, and the number of American Indian/Alaskan Native students fell by more than 15% (Hussar & Bailey, 2013). At the same time, students who were identified as English Language Learners (ELL) rose to more than 9% of the total student population (Kena et al., 2016). These demographic shifts vary across portions of the country; in 2013, ELL students made up less than 3% of all students in 14 US states, while in California, ELL students accounted for more than 22%. With these changes, many schools now serve a much larger proportion of students who are likely to enter schools less academically prepared and whose behavior may be less aligned with school expectations (Farkas, 2003). On average, students from ethnic minority groups have lower levels of family income and parental education (Hernandez, 2004), putting many of these students at a disadvantage for school readiness and adjustment to school behavioral standards, relative to peers. 3 Expectations for classroom behavior continue to reflect European-American norms, as do instructional methods (Boykin, Watkins-Lewis, & Kizzie, 2006). Consequently, in regions with large changes in student demographics, teachers often spend more time addressing student behaviors, trying to adapt their instruction, and worrying about student performance on high-stakes assessments that often assume a degree of mainstream cultural familiarity. As each school tries to address its priorities within the context of its strengths and challenges, its staff, teachers, and students may well face a jumbled and incongruous mix of tasks. The purpose of this study was to examine if schools with staff members who share a strong understanding of roles and expectations – that is, schools with “high focus” – were also schools with teachers who were more able to maintain morale in the presence of occupational stress and who were less likely to leave the school. High focus schools theoretically foster more consistent planning, fairer performance evaluations, better understanding of how to respond to specific challenges, and a clearer picture of how staff members’ professional identities fit into the overall identity of their respective schools. High focus schools should be schools where the staff members know what to do. Teachers are faced with numerous, sometimes contradictory priorities. It is therefore important to know how schools leaders communicate the priorities that are most valued within their school, and to know how the degree of clarity regarding teachers’ roles impacts their collective morale and the likelihood that they stay in their jobs. What follows are brief overviews of literature, primarily school-based research, pertaining to the outcomes of interest in this study – teacher turnover and 4 teacher morale/job satisfaction – and factors that predict them, primarily characteristics of schools as work environments. Extra attention is paid to the effects of school administrative support, student conduct, and role clarity, as these specifically relate to study hypotheses. Finally, the school focus construct is explicated further, after which the study itself is presented. 5 Background When teachers leave The National Commission on Teaching and America’s Future estimated that teacher turnover costs public schools at least $7.3 billion per year once federal, state, and district losses are counted (NCTAF, 2007). A look at teacher attrition data from the National Center for Education Statistics (Goldring, Taie, & Riddles, 2014) indicates that the turnover rate was more than 25% higher in 2012 than it was 20 years prior, currently estimated to be around 16%. However, whether that number by itself is cause for concern is a matter of perspective, because roughly half of the teachers who leave a school in a given year transfer to another school (Ingersoll, 2003). While school-to-school transfers do not diminish the national teacher pool, the financial, logistical, and school climate impacts for schools losing their teachers is largely the same, regardless of whether the teachers stayed in the profession. Turnover is not evenly distributed across schools, either; in communities with higher proportions of students from traditionally disadvantaged backgrounds and lower SES, the annual teacher attrition rate often rises higher than 20% (Keigher, 2010; Loeb & Darling-Hammond, 2005). These are schools that could most benefit from a stable and cohesive teaching staff, but high turnover creates a need for new teachers, who are themselves less likely to stay than are more experienced teachers, and so goes the cycle (Borman & Dowling, 2008; Hanushek, Kain, & Rivkin, 2004; Ladd, 2011). To achieve the goals of policymakers, administrators, and the public, teachers must stay in their jobs and provide stability for students and schools. The majority of teachers who leave the profession are pursuing different work rather than retiring, 6 suggesting that some attrition could conceivably be avoidable (Cox, 2007; Ingersoll & May, 2011; Goldring et al., 2014). In addition to the financial strain of having to recruit, hire, and train new teachers, especially in hard-to-staff schools, students bear the costs of high turnover in their lost opportunities. Within a context of nationwide budget cuts, the money devoted to immediate staffing issues diverts critical funds away from valuable student services, such as providing up-to-date textbooks, hands- on learning experiences, and devices for technological literacy training, as well as offering breakfast or afterschool care for children whose families are unable to regularly provide them (Conley & Woosley, 2000). Schools with greater staff stability can provide students with a more cohesive curriculum, are more able to build partnerships with parents and the surrounding community, and increase the chance that students will form supportive, ongoing relationships with school personnel (Shields, 2001). While much is known about the factors that might lead to teacher attrition, little progress has been made toward establishing policies and procedures that effectively address the issue (Ingersoll, 2001; Liu & Meyer, 2005; Miller, 2006). Keeping teachers and keeping them happy One well-established way to prevent teacher attrition is to maintain high levels of teacher job satisfaction and morale (Ingersoll, 2001; Liu & Meyer, 2005; Marvel, Lyter, Strizek, & Morton, 2007; Tickle, Chang, & Kim, 2011). Furthermore, effectively addressing the issues of teacher satisfaction and morale can produce numerous educational benefits on both schools (organizational commitment, organizational performance, and teacher quality) and students (positive behavior, academic achievement, and satisfaction with school) (Bogler, 2001; Caprara, 7 Barbaranelli, Steca, & Malone, 2006; Dinham, 2007; Johnson, Kraft, & Papay, 2012). Unfortunately, as teachers’ workloads have increased and their performance has come under increased scrutiny, teacher job satisfaction is on the decline in many countries (Scott, Stone, & Dinham, 2001). As with teacher retention, the existing knowledge of what factors support teacher satisfaction and morale has not been sufficiently translated into easily implementable and broadly effective organizational practices (Judge & Kammeyer-Mueller, 2012). One of the difficulties in applying what is known about teacher job satisfaction and morale in school interventions is the inconsistency with which these constructs are conceptualized and measured. In teacher job satisfaction research, as well as in organizational research elsewhere, there is no agreed-upon definition for job satisfaction, nor for morale, and the terms may be used interchangeably across studies. For purposes of this study, Evans’s (1997, p. 328) simple and comprehensive definition is sufficient: “a state of mind determined by the extent to which the individual [emphasis added] perceives her/his job-related needs to be being met.” However, the studies herein define and measure the construct differently. In some instances, teacher responses are dependent on the teacher’s own interpretation of what job satisfaction means, as the researchers and their measurement instruments do not specify. Researchers may measure job satisfaction as a global attitude or they may measure satisfaction with certain elements of the job. This distinction carries important implications for identifying how to approach areas of need; for example, one might argue that offering a wider variety of professional development opportunities will improve teachers’ job satisfaction, or one might as convincingly 8 argue that diversifying professional development opportunities will only improve teachers’ satisfaction with professional development. The former approach misses the specificity that might guide targeted intervention, while the latter approach neglects the likelihood that different teachers assign more or less value to specific elements of their work environment and that improvements in one domain may impact overall satisfaction (Evans, 1997; Shen, 1997; Skaalvik & Skaalvik, 2011). Defining “morale” across studies is even more problematic, as it may refer to the attitudes of an individual teacher, the collective teachers in a school, or the whole school, including principals, teachers, other staff, and students. As a result, researchers have sometimes taken wildly different approaches. For example, Jones (1997) measured morale using the Purdue Teacher Opinionaire (Bentley & Rempel, 1967), which contains ten subscales that encompass nearly all of the school working environment topics raised in this review: teacher rapport with principal, satisfaction with teaching, rapport among teachers, teacher salary, teacher load, curriculum issues, teacher status, community support of education, school facilities, and community pressures. On the opposite end of the spectrum, Weiss (1999) used a single item to measure morale, “I sometimes feel it is a waste of time to try to do my best as a teacher,” which is a profound statement that potentially taps into a number of attitudes, but does not allow for interpretation of what those attitudes are. Other articles in the review mention morale informally in their background and discussion, apparently as either aggregated satisfaction, as an ambiguously separate individual work attitude from job satisfaction, or as synonymous with job satisfaction. In the operational definition of this study, morale is measured by teachers’ perceptions of 9 the school-wide “esprit de corps” (Gottfredson et al., 2000), which subsumes their ratings of collective job satisfaction along with a sense of collegiality and trust, among other factors which will be outlined in the “measurement” section. It should not be assumed that this distinction between job satisfaction as an individual attitude and morale as a separate indicator of school-wide staff attitudes is made in each of the studies presented herein, though efforts have been made to present their findings within that schema. While addressing satisfaction and morale can be important for improving teacher retention, there are other equally vital outcomes to consider. After all, not all dissatisfied teachers leave the profession, and those who remain in their schools may well contribute to a decrease in overall morale through their interactions with coworkers, students, and parents (Baughman, 1996). Furthermore, dissatisfied teachers are unlikely to perform at their best, either through their own diminished motivation or because of the job stressors that precipitated their dissatisfaction (Conley & You, 2014; Griffith, 2006; Shen, Spybrook, & Ma, 2012). Low teacher morale also predicts greater teacher absenteeism, which, when common, presents a major threat to school functioning, regardless of attrition (Reyes & Imber, 1992; Woods & Mantagno, 1997). High teacher turnover is both a result of and a cause of low morale; as such, the recent emphasis on reducing turnover is akin to treating the symptoms, not the cause, if this effort is not pursued concurrently with a search for effective ways to improve teachers’ satisfaction with their occupation. Teacher morale is higher in schools with sufficient workplace amenities such as updated textbooks and classroom supplies (Ingersoll, 2001). The same is true 10 regarding the physical condition of school buildings themselves (Bolin, 2007). These school environmental characteristics are no doubt important determinants of the quality of teachers’ day-to-day work lives. However, both factors reflect school resources, which are themselves determined by school district resources and outside of the control of school administrators and staff. It is well-established that schools in lower-SES communities tend to have less satisfied teachers (Loeb & Darling- Hammond, 2005), but literature suggests that community demographics, in and of themselves, do not directly affect teacher satisfaction and retention (Baughman, 1996; Kelly, 2004; Shen, 1997). It is more likely that community demographic measurements serve as a proxy for unfavorable working conditions. If it is indeed the working conditions, rather than the surrounding community, which predict teacher morale, then there is reason to believe that schools in poorer areas can foster an equal level of teacher morale as schools in affluent areas, given a sufficient investment of resources. Unfortunately, unless individual schools themselves find ways to secure capital outside of their district-allotted budget, there is little that can be done by school colleagues to update their textbooks, renovate their building, and so forth. Administrative support and leadership “Administrative support” and “principal support” are well-supported correlates of teacher job satisfaction, but such broad terms are naturally open to interpretation; some studies are clearer than others about what they measured as “support.” Consider a single item from the Schools and Staffing Survey (NCES, 2012) which is especially open to interpretation: “The school administration’s behavior toward the staff is supportive and encouraging.” Moore (2012) included the 11 item as part of a larger “administration/colleague support” factor; Weiss (1999) included the item under “perceptions of school leadership/culture;” Grissom (2011) under “principal effectiveness;” and Stockard and Lehman (2004) under “social support.” The proposed study uses a similar item under the Teacher Administrator Leadership Scale (see Appendix B, Table 16). Across studies, “support” falls in one or more of the domains outlined by House (1981): emotional support, characterized by demonstrating empathy, warmth, and trust; instrumental support, which involves providing tangible assistance through resources and helping with tasks; informational support, the provision of knowledge and advice; and appraisal support, relating to appropriate praise and helpful feedback. Littrell, Billingsley, and Cross (1994) investigated these dimensions separately among a sample of 700 special education and general education teachers and found emotional support to be most positively related to job satisfaction, with appraisal support the second strongest predictor of the four support types. The behaviors associated with these two domains of administrative and principal support are reflected in the Administrator Leadership Scale, and are presented next. Principals provide emotional support and encouragement when they recognize and celebrate staff effort and achievement, which can go a long way toward establishing a positive, motivating, and satisfying work environment (Baughman, 1996; Grissom, 2011; Liu & Meyer, 2005; Mehta, Atkins, & Frazier, 2013; Moore, 2012; Perie & Baker, 1997; Perrachione, Rosser, & Petersen, 2008; Skaalvik & Skaalvik, 2011a; Stockard & Lehman, 2004; Taylor & Tashakkori, 1995; Tickle et al., 2011; Weiss, 1999). This is especially true for first-year teachers; principal praise 12 for hard work and teacher success has been identified as a key predictor of both first- year teachers’ job satisfaction (Stockard & Lehman, 2004) and their expressed commitment to the teaching profession and intentions to continue teaching (Weiss, 1999). Teachers spend much of their work days isolated in their classrooms, without immediate feedback from colleagues or administrators, and it clearly helps teacher morale when they receive notice and praise for their efforts. Principals are also managers, responsible for supporting the professional growth of teachers and staff in their schools (Gigante & Firestone, 2008). Principals can succeed in this role through their appraisal support, providing fair and balanced teacher evaluations with useful feedback (Johnson et al., 2012; Perrachione et al., 2008). First-year teachers, who are often trying to establish their professional skills and identity while attending to the multitude of job stressors that go with the job, place constructive feedback among the highest predictors of their job satisfaction and commitment (Ma & Macmillan, 1999; Weiss, 1999). In a recent Turkish teacher sample, Elma (2013) included fair evaluations as an element of “procedural justice”, which was just as predictive of teacher job satisfaction as “interactional justice,” which encompasses the interpersonal respect and support described above. As such, principals would do well to keep in mind that appraisal support, in addition to guiding teacher performance and professional development, is also imperative for teachers’ satisfaction with the teaching profession. For evaluations to be fair, teachers need to know what it is that they are supposed to do, and when teachers perceive the expectations of school administrators to be unclear or conflicting, they are less satisfied with the job and less likely to stay 13 in it (Reyes & Imber, 1992). While role clarity is important, teachers also value dialogue with school administration in the formation of roles and goals (Richardson, Alexander, & Castleberry, 2008). Clear school goals may not reflect teachers’ personal goals, and teachers are significantly more satisfied and committed when their schools’ goals and values are consonant with their own (Ma & MacMillan, 1999; Skaalvik & Skaalvik, 2011a). At the very least, schools with well-defined objectives provide their teachers the chance to make an informed determination of how they might wish to adapt their personal style to fulfill their job roles. The traditionally siloed setting of classroom teaching has historically allowed for teacher autonomy, and teachers who quit their jobs often cite a loss of autonomy among the motivations for that decision (Boyd et al, 2011; Goldring et al., 2014; Ingersoll, 2001). To resolve the apparent tension between clarity and autonomy, it is helpful to specify the job roles in question. For example, teachers feel especially supported when the procedures for addressing student behavior are explicit and consistent (Liu & Meyer, 2005; Perrachione et al., 2008), but are not likely to feel the need for equal regimentation regarding how they structure their lessons. Indeed, teachers’ job satisfaction has been persistently found to be positively related to their perceived autonomy and flexibility concerning pedagogy and curriculum (Moore, 2012; Perie & Baker, 1997; Rice & Schneider, 1994). Classroom autonomy is largely considered a product of principal trust in teachers’ abilities (Bogler, 2001; Elma, 2013; Johnson, Kraft, & Papay, 2012; Pearson & Moomaw, 2005; Shen, Spybrook, & Ma, 2012) and administration’s willingness to recognize teachers as the experts in their classrooms (Kloep & Tarifa, 1994), so long as 14 objectives are met. Teachers across cultures value a sense of occupational prestige (Bektas, Ocal, & Ibrahim, 2012; Bolin, 2007; Cerit, 2009), and they become teachers because they want to work with students (Farkas, Johnson, & Foleno, 2000). Each has his or her own preference in how to fulfill that role, according to personal style and professional judgment; therefore, it appears that teachers are most satisfied when they have a clear picture of what they are expected to do, not necessarily how they are expected to do it (Archbald & Porter, 1994; Weiss, 1999). Unfortunately, teachers in the U.S. report a progressively declining sense of autonomy across a range of job roles in the past decade (Sparks & Malkus, 2015). Providing the opportunity for teachers to exercise their discretion in the classroom can strengthen teacher enthusiasm in instructional activities, and is an avenue through which teachers can participate in school leadership roles (Cerit, 2009). Engaging teachers in making other school decisions can often lead to similar increases in job satisfaction and morale as well (Ingersoll, 2001; Jones, 1997; Koklu, 2012; Rice & Schneider, 1994). After all, having input into school policy and procedural decisions also gives teachers latitude over what they will ultimately be working on in their classrooms. Decision-making participation regarding evaluation procedures, selection of department and team leaders, structuring supervisory hierarchies, hiring and promotion criteria, professional development opportunities, and establishing task-oriented teams are among the dimensions that have been linked to positive teacher satisfaction (Koklu, 2012; Jones, 1997; Rice & Schneider, 1994). Furthermore, the higher the discrepancy between teachers’ desired involvement in 15 school-wide decision making and their actual involvement, the greater the expected decrease in their satisfaction (Jones, 1997). However, the scope and domains of that involvement must be purposefully considered within each school’s context and culture. Taylor and Tashakkori (1995) found that once other predictors of work climate were controlled for, involvement in managerial-related decisions showed only a slight positive relationship with job satisfaction. It is also important to bear in mind that administrators are trained and hired to make certain decisions, and even when teacher morale increases through decision-making participation, it may not lead to improvements in student outcomes (Jones, 1997), which is the ultimate test of what constitutes best school practices. As mentioned above, teachers become teachers in order to work with students, and they value principals who serve as strong leaders in their administrative capacities so that teachers can focus on their own classrooms and students (Bogler, 2001; Van Maelle & Van Houtte, 2012). Students as stressors Yes, teachers generally become teachers because they want to work with students, and they derive a great deal of satisfaction from their relationships with students (Dinham, 2007; Ma & MacMillan, 1999; Shen, 1997). However, when there is a discrepancy between teachers’ desired student interactions and students’ actual behavior, the resultant frustration can damage teacher job satisfaction with comparable strength (Hastings & Bahm, 2003). No wonder then that student discipline problems consistently emerge among the top predictors of teacher job dissatisfaction and attrition (Ingersoll, 2004; Kelly, 2004; Liu & Meyer, 2005; 16 NCTAF, 2007). Student behaviors like aggression, noncompliance, talking and acting out of turn, and defiance, are repeatedly reported to increase teacher stress and motivation to quit (Boyd et al., 2011; Geving, 2007; Kyriacou, 2001). Teachers in the National Center for Education Statistics’ Teacher Follow-Up Survey samples have consistently listed improved student discipline as the second most important step, behind increased salary, that schools might take to encourage teachers to stay (Whitener, Gruber, Lynch, Tingos, & Fondelier, 1997). Considering that salaries are set by the school district, student discipline might be the largest determinant of teacher attrition inside the school building. Even the public at large most frequently cite “lack of discipline” as the biggest problem facing public schools (Cotton, 2001). Student conduct is largely influenced by individual student background characteristics, such as socioeconomic status, English language proficiency, parental education, and familiarity with classroom behavioral expectations that reflect European-American norms (Boykin et al., 2006; Farkas, 2003; Hanushek et al., 2004; Hernandez, 2004; Loeb & Darling-Hammond, 2005). Therefore, student behavior is very much a function of the communities in which schools serve, and the perception of student misconduct is largely driven by exogenous cultural values. School leaders can positively address school-wide behavior, and often do, but many of the influences of students’ behaviors fall outside of the school’s sphere of influence. One aim of the present study was to investigate whether school focus, as a conceivably corrigible characteristic of the school working environment, might buffer the deleterious effects of student misconduct on teachers’ morale. 17 Roles, goals, and confusion As described above, teachers are beset by a wide, shifting, and potentially confounding set of goals. Their expectations are dictated by a broad cast of agents: federal guidelines, state guidelines, district guidelines, school administrators, parents, others teachers, and students. Those parties seldom agree fully on the best methods for instruction, classroom management, schedule planning, and teacher performance evaluations (Petersen, 1997). Teaching practices and teacher roles have long been shaped by policy demands and outside recommendations, but the expectations imposed upon the profession have been particularly expansive during the current era (Valli & Buese, 2007). Recent changes have been less about replacing old role definitions and more about adding new responsibilities to existing ones (Hargreaves, 2000). Teachers, then, have both too much and too little information regarding what they ought to be emphasizing and doing in their schools (Bovender, 2013). Given the array of goals that schools are expected to pursue and the limited amount of time they have to carry out their job tasks, it is not always clear to teachers, students, or even principals which goals should be prioritized; establishing agreement on priorities is at least equally difficult. Widespread role confusion has a consistent negative effect on the morale, job commitment, and productivity of workers across occupations (Abramis, 1994; Gilboa, Shirom, Fried, & Cooper, 2008; King & King, 1990). In short, the same is true for teachers (Hulpia & Devos, 2011; Kyriacou, 2001; Leithwood & McAdie, 2007). Conversely, role clarity can provide support for the teacher subgroups most at risk for attrition. Explicit, consistent procedures help establish a stable work environment for 18 beginning teachers, giving them an opportunity to develop their acumen and bolstering their professional growth, self-efficacy, and job satisfaction as a result (Stockard & Lehman, 2004; Weiss, 1999). Greater clarity and consistency are especially beneficial for teachers in schools serving in high-poverty areas, where typically higher rates of teacher turnover and the extra needs of students often lead to more chaotic school environments (Loeb & Darling-Hammond, 2005; Mehta et al., 2013). From role clarity to school focus Clarity at the organizational level is critical if there is to be clarity at the individual level. It will also be helpful to clarify some terminology and conceptual overlap before situating school focus among extant educational and organizational literature. A helpful first step is to outline the lineage of the term “school focus” as it is used in this study. Holland (1997) expanded on his popular theory of workplace identities by describing the clarity of organizations’ goals and how those goals are communicated. An organization that demonstrates a congruent and stable set of goals, and which makes those goals explicit, is said to have a clear environmental identity, while an organization characterized by amorphous and inconsonant goals is said to have a diffuse environmental identity. The magnitude of an organization’s environmental identity emerges from the collective perspective of employees regarding how consistent and well-understood their job expectations are (Perdue, Reardon, & Peterson, 2007). Gottfredson (2000) refers to that environmental identity 19 as organizational focus.1 Incidentally, the term “environmental identity” is more popular in ecology and environmental psychology in contexts which are not germane to this review or this study. “School focus” here is just shorthand for organizational focus in the school setting for the sake of brevity and conceptual simplicity. School focus, then, is an organizational climate characteristic as experienced by workers in a school. Organizational climate refers to employees’ shared perceptions of workplace policies and procedures and the meaning that they attach to their experiences in the work setting (Ostroff, Kinicki, & Tamkins, 2003; Schneider, Ehrhart, & Macey, 2013). This is distinct from school climate, which is a largely separate body of research. Certainly school focus influences and is influenced by school climate, but there is not a consensus definition or conceptualization for school climate (Cohen, McCabe, Michelli, & Pickeral, 2009; Zullig, Koopman, Patton, & Ubbes, 2010). While schools exist within a complex ecology with interconnected levels of influence (Moore, 2012), it seems fair to assert that school climate research is generally concerned with students’ experiences in the school environment. This review is focused on teachers and schools as workplaces, but the methodology and conceptual framework of the study are primarily grounded in organizational climate research. School focus showed a high positive correlation with a measure of teacher morale across a diverse national sample of schools (Gottfredson, 2000), as discussed further below. Perdue et al. (2007) also found that environmental identity by itself 1 The term “organizational focus” features in a 2009 book by Bart Nooteboom. While his theory similarly references Holland’s work, the etymology of his term appears to be independent of its use here, with only some conceptual overlap. 20 indicates what they consider a healthier working environment, predicting greater employee satisfaction with job supervision and collegiality with coworkers. It makes sense then that teachers in a high focus school would report higher morale; stressful situations in such a school should have a more accessible list of responses, planning and scheduling should be better informed, students and staff should have a clearer understanding of behavioral reward and consequences, and teachers should have a more certain sense of their occupational purpose and performance. Schools with clearly defined values and goals, especially when staff members share those goals, can foster a sense of common purpose and shared efforts, which predict higher loyalty and commitment among staff members (Hulpia & Devos, 2011; Johnson et al., 2012; Kyriacou, 2001; Mehta et al., 2013; Sergiovanni, 1992; Van Maele and Van Houtte, 2012). The responsibility for communicating these values and arranging concordant practices may start with the principal, but school norms arise from multiple associative working relationships between teachers and students (Bryk & Driscoll, 1988). Purposive schools are crucial for concordant teacher practices and student success, but maintaining such schools in an era of extensive reform and restructuring, when the key functions of education are under constant debate, requires meticulous planning and supervision (Petersen, 1997; Conley & You, 2009). 21 Present Study In addition to the purposes outlined in the introduction section, it should be noted that elements of the present study serve in part to expand on the results of a related 2013 study (Bovender), which supported the importance of school focus as a predictor of higher teacher job satisfaction and lower intent to quit. Focus, as an aggregated school-level predictor, significantly predicted teachers’ school-level aggregated satisfaction in one of two years’ samples (γ = .34); unfortunately, there was too little between-school variance in school focus in the second sample to attempt replication of that finding. Teachers’ individual perceptions of their respective schools’ focus showed a very strong positive correlation with individual satisfaction – β = .82 for the 1st year, β = .87 for the 2nd year – when included in a model with student misconduct and teacher demographic variables. In addition, individual teacher perceptions of focus correlated negatively with individual teachers’ reported intent to quit – β = -.56 for the 1st year, β = -.51 for the 2nd year – in a similar model. Limitations of the study sample and measurement restricted the generalizability of those findings and left work to do for contextualizing the potential value of focus relative to other school organizational characteristics. That study used a sample of 45 public elementary schools in a single suburban district within a relatively high-income county (median household income was roughly 90% higher than national average and 50% higher than state average.) In addition, there was a narrow range of ratings regarding job satisfaction, which were substantially concentrated at the high end of possible scores (as were focus ratings), and student 22 problem behavior ratings were similarly concentrated at the low end. As mentioned, in one of the two years’ worth of data, there was insignificant between-school variance in school focus, making those data unusable beyond individual-level analyses. Data used in the present study come from a large national sample with greater variation among teacher-reported climate factors, including teacher morale, school focus, and student misconduct. In this way, the present study adds substantially more power for replicating those earlier results. Additionally, given the relative scarcity of research into school focus as an environmental characteristic of schools’ working environments, a necessary next step is to establish whether focus is an important standalone construct when controlling for other related workplace perceptions. The present study attempts to do just that; many of the principal and administrative behaviors outlined above are included as concurrent predictors of morale and turnover. Further detail regarding the rationale for each hypothesis in the current study is presented alongside the hypotheses below. Hypotheses 1) Teachers’ perceptions of their respective school’s focus will correlate positively with their reported perceptions of school-wide teacher morale and will correlate negatively with the school’s turnover rate, independent of other perceptions of administrator leadership. It would be surprising if school focus did not correlate significantly with school morale; that specific relationship may not have a large body of school-specific literature devoted to it, but similar relationships have been consistently supported in 23 other occupations and work environments. Additionally, school focus itself has shown high positive correlations with morale and job satisfaction (Gottfredson, 2000; Bovender, 2013). The strength of this relationship is likely at least partially explained by the bidrectionality of job satisfaction and employees’ responses regarding perceptions of organizational characteristics (Staw, 1975). Employees who are happy with their jobs are more likely to report favorably on other aspects of the work environment, regardless of whether that satisfaction is directly attributable to each of those characteristics. In addition, schools that function well in one domain of working conditions are also likely to function well in other domains (Boyd et al., 2011). The inclusion of teachers’ perceptions of administrative leadership is intended to establish whether school focus shows this relationship in and of itself. Expectations are largely communicated from the top down through principals and administrators (Leithwood & McAdie, 2007; Shen et al., 2012), and to my knowledge, it has not been shown that focus as a school-level characteristic plays an important role on its own, rather than possibly simply serving as a proxy for overall perceptions of school leadership. Teachers communicate with one another (or do not), work in teams (or do not), and establish a working culture beyond the input of administrators, who may or may not consistently agree on what teachers should prioritize. I contend that administrators’ competence in some areas of responsibility does not necessarily equate to their schools having higher focus, but this hypothesis stands to demonstrate whether teachers agree with this assertion. Furthermore, improving clarity and consistency in roles and goals may be a tenable area for intervention, whereas “improving school leader quality”, while an important aspiration, may be a more 24 diffuse and less immediately actionable endeavor. It’s harder to advocate for working towards the former without establishing it as somewhat independent of the latter. 2) School focus will attenuate the expected negative relationship between classroom disorderliness and teacher morale and attenuate the expected positive relationship between classroom disorderliness and school turnover rates. The crux of this question is whether school focus can serve as a safeguard against the deleterious effects of student misconduct on teacher morale and retention. In a sample of Finnish teachers, Bakker et al. (2007) found that “job resources”, including collegiality and supervisor support, protected against the negative impact of student misconduct on teacher work engagement. This job demands-resources model (Demerouti, Bakker, Nachreiner, & Schaufeli, 2001) has been the framework for similar buffering hypotheses in studies of protective organizational climate factors and positive worker outcomes in other high-stress occupations, including nurses (Dollard et al., 2012), offshore workers (Nielsen et al., 2011), and police officers (Dollard et al., Tuckey, & Dormann, 2012). Student discipline problems consistently emerge among the top predictors of teacher job dissatisfaction and attrition (Ingersoll, 2004; Kelly, 2004; Liu & Meyer, 2005; NCTAF, 2007), so while misconduct is only one element of job stress, it is a very important one. The question then is whether teachers who collectively experience poorer student behavior in their school are less likely to experience poor morale or to attrit out of the school when they have a clear understanding of their priorities and expectations. Here again, school focus seems more amenable to 25 immediate intervention than does school-wide student conduct, which is often influenced by aspects of students’ lives beyond the school environment, though important efforts to improve school-wide student conduct, such as Positive Behavioral Interventions and Supports (see Bradshaw et al., 2008; Barret et al., 2008), are increasingly common, and rightly so. It is clear that schools can expect a wide range of desirable outcomes if they can effectively reduce student misconduct across their school, but if increasing focus can help teachers remain resilient in the face of high student-related job stress, a focus intervention might be a helpful stopgap. 3) The anticipated positive relationship between school focus and school morale will be stronger in schools where raters show greater convergence in their perceptions of school focus, as will the anticipated negative relationship between school focus and employee turnover. In short, inasmuch as staff members endorse a cohesive understanding of what they are expected to do, it would seem to be more meaningful if there was also clarity and consistency in that perception. That perceptual cohesion, referred to in literature as climate strength (Dickson et al., 2006; Schneider et al., 2002), has been found to increase the effects of evaluations of other domains of working environments (Du et al., 2015; González-Romá et al., 2002; Schneider et al., 2002; Schyns & Van Veldhoven, 2010). Because climate strength and organizational focus seem to overlap conceptually – Zohar and Luria’s (2005) definition of climate strength is essentially identical to Gottfredson’s (2000) definition of focus – it stands to reason that the 26 strength of teachers’ perceptions of school focus could be an indicator of the level of school focus itself. Methods Sample The proposed study draws from principal and teacher survey data collected as part of the National Study of Delinquency in Schools (Gottfredson et al., 2000). Data were collected over two years, with two separate principal questionnaires administered in the first and second years, and with teacher and student responses collected during spring of the second year. The data used for the present study come from the 1st-year principal and teacher surveys. The principal and teacher questionnaires were sent to a sample of 847 public, private, and religiously-affiliated (of these, mostly Catholic) secondary schools stratified roughly equally across location (urban, suburban, and rural) and grade levels (middle, and high). Responses for the 1st-year principal survey were received from 558 schools, an overall response rate of 65.9%. Participation rates fell in the 2nd year surveys; teacher responses were received from 403 schools, (47.6%), while principal responses in that year came from 416 schools (49.1%). Rural and middle schools showed the highest relative response rates for both teachers and principals (see Appendix A, Tables 11 and 12). Available data do not contain sufficient information to calculate teacher response rates within each school. For the purposes of these analyses, schools designated as “alternative school for educationally handicapped students”, “alternative school for students with behavioral or adjustment problems”, or “military-style school” were removed from 27 the remaining sample. It is expected that student behavior and school policies would be substantially different in these school environments than in the rest of the sample, potentially confounding the results. Fifty-three schools fit at least one of these criteria (13.2%). Schools with fewer than five teachers responding were also excluded because of the reliance on aggregated data in the analyses. Finally, two additional schools were removed because of significantly higher turnover than others in the sample (>30%), leaving a sample of 348 schools for the teacher analyses. Survey data from all self-reported full-time teachers in these remaining schools, with non- respondents (those teachers who responded to 50% or fewer of the survey items) removed, yielded a final teacher sample of 11,376. There was an average of 32.7 teachers per school included in the final analyses. A breakdown of the final sample is below. Further school and teacher sample characteristics are present in Appendix A. Table 1 Final Sample Sizes by Grade Level and Location Type Location Type Middle Schools High Schools Combined School Levels N Teachers N Schools (Turnover) N Teachers N Schools (Turnover) N Teachers N Schools (Turnover) Rural 2,138 77 (62) 1,376 58 (48) 3,514 135 (110) Suburban 2,238 63 (47) 1,341 38 (29) 3,579 101 (76) Urban 2,279 67 (50) 2,004 45 (32) 4,283 112 (82) Combined Locations 6,655 207 (159) 4,721 141 (109) 11,376 348 (268) Note. Turnover analyses required both teacher and principal questionnaire responses. The number of schools used in those analyses was limited by principal participation in each sample, and appears in parentheses. 28 Measurement The Teacher Questionnaire, administered in the second-year data collection wave, contained 116 survey items relating to teacher background information, teacher practices and experience, student misconduct, perceptions of school climate, and school-wide programs and practices. Contained within the questionnaire were the following teacher response scales which were used in the analyses: Organizational Focus, Classroom Orderliness, Morale, and Administrator Leadership. The Organizational Focus scale (Gottfredson & Holland, 1996) is intended provide a measure of the degree to which an environment has a focused set of consistent and explicit goals, as opposed to conflicting and poorly defined goals. The scale contains 16 items with 4-point Likert-type response options ranging from “False” to “True”. In the first psychometric test of the scale (at that time, referred to as “Environmental Identity”), Smart and Thompson (2001) found an impressive internal reliability, with an alpha coefficient of 0.91, among a sample of faculty in 38 academic departments. More recently, teacher samples demonstrated equally strong alpha coefficients of .90 and .89, respectively (Bovender, 2013). See Table 14 in Appendix B for the full list of scale items and the directionality of their response scores. The Classroom Orderliness scale, adapted from the Effective School Battery (Gottfredson, 1999), contains 14 items relating to class-wide student behaviors, with 5-point responses ranging from “Never” to “Almost Always”. Orderliness refers to the presence of desirable student classroom behaviors (paying attention, complying with teacher requests) and the absence of disruptive externalizing behaviors (teasing 29 and threatening classmates, talking out of turn, distracting classmates). Items reference collective student behavior, such as “Students do what I ask them to do,” and “The classroom activity comes to a stop because of discipline problems.” Naturally, the second example would be reverse-scored when indicating the degree of “orderliness”. However, for conceptual clarity in this study, which considers the impact of student misconduct, all items of the Classroom Orderliness scale were reverse-scored from their original orientation to provide a measure of classroom disorder, and will subsequently be referred to as such. See Table 15 in Appendix B for the full list of scale items and the directionality of their response scores. The Teacher Morale and Administrator Leadership scales of the Effective School Battery were used to assess the sense of collegiality between school staff, the efficacy with which staff members feel they are able to achieve their goals, and teachers’ perceptions of principals’ traits, practices, and performance. The 9 items of the Teacher Morale scale offer true/false responses relating to individual teachers’ perceptions of themselves and the other teaching faculty (see Appendix B, Table 17). The 7 items of the Administrator Leadership scale (Table 16) are also true/false and refer to a variety of principal/administrator behaviors, such as “Our principal is a good representative of our school before the superintendent and the board,” and “The administration is supportive of teachers.” As mentioned, principals responded to different questionnaires in the first and second year of data collection. The 1st-year questionnaire contains 142 items, including background school information that was used here for sample inclusion criteria and division of schools by type and region groups, teacher-principal 30 relationships, and school-community relationships. The 2nd-year questionnaire contains 211 items, relating primarily to school disciplinary policies and implementation, as well as some items relating to principal leadership attitudes and behaviors. As teacher perceptions and turnover were the outcomes of interest, only the 1st-year principal response data were used to provide information about the school, including teacher turnover rates. Data Analyses Analyses used teacher item endorsements from the Organizational Focus, Classroom Disorder, Teacher Morale, and Administrator Leadership scale ratings from the teacher questionnaire described above. A “school focus” score was measured for individual teachers by using the average of each teacher’s responses to items of the Organizational Focus scale. These teacher means were then aggregated and averaged for a school-level score for the school-level analyses. “School morale” scores were likewise derived by average item response per teacher, then averaged across all teachers per school. There is a straightforward need for aggregation where school focus and school morale are concerned, as the group-wide consensus is inherent in the working definition of these variables. Additionally, scale items are worded to elicit teacher ratings about their collective experiences. Aggregated group response data are closely related to the individual-level responses, but differ in meaningful ways (Bliese, 2000; Griffin et al., 2001; Griffith, 2006). Each teacher’s perception of the school’s focus and teacher morale is an important indicator of the school working environment, but it is more meaningful to consider the group’s overall perceptions, wherein these constructs arise from the 31 collective consensus of workplace members (LeBreton & Senter, 2008; Schein, 2010). Many studies have found stronger relationships between workplace perceptions and organization-wide performance than those perceptions and individual performance (Griffith, 2006), highlighting the value of this approach. Teacher-rated Administrator Leadership items were similarly aggregated across all teachers in each school for school-level averages. Aggregating these ratings seems appropriate enough; there may be some administrator characteristics that are more instructive than others regarding the relationships of interest in this study, but administrator leadership is intended to function as a control to the influence that overall leadership perceptions might have on focus ratings. Averaging the schoolwide teacher perceptions of principal/administrator efficacy to a single school-level score seems sufficiently meaningful for that purpose. Finally, Classroom Disorder scale scores came from average item scores by teacher, aggregated for a single school-level score. Unlike the other predictor variables described, classroom disorder as an aggregated characteristic of the school is based on teachers’ individual experiences in their own classroom. While the discrepancy in item wording between this and other predictor variable (“my classroom” versus “our school”) is not ideal for purposes of this study, it is reasonable to expect that there are schools in the sample wherein student misbehavior is more pervasive in many, if not all, classrooms. It is also reasonable to expect that the collective experiences of teachers in a school regarding the behavior of students in their respective classrooms is a meaningful characteristic of the school working environment. Lastly, approximating schoolwide student misconduct from the sum of 32 individual responses, while potentially washing out the important perspectives of teachers who experience different classroom conduct than their coworkers, establishes a common level of analysis for making school-level inferences with due caution. Hypothesis 1 concerns the expected positive relationship between school focus (SCHFOC) and school morale (MORALE), independent of the effects of other potentially related perceptions of administrative leadership qualities (LEADER). As these are all treated as school-level characteristics, the relationship can be assessed using a straightforward regression equation: 0 1 2( ) ( )j j j jMORALE SCHFOC LEADER u      (1) where the morale score for school j is a function of the grand mean of school morale ( 0 ), the regression coefficient for SCHFOC in school j ( 1 ), the regression coefficient for LEADER in school j ( 2 ) and the error for school j ( ju ). Additionally, the school’s turnover rate (see measurement description below) will replace MORALE as the predicted variable to address part two of this hypothesis, the expected negative relationship between SCHFOC and school turnover rates, independent of (LEADER): 0 1 2( ) ( )j j j jTURNOVER SCHFOC LEADER u      (2) Turnover rates for each school are derived from principal reports; principals reported the number of full-time teachers in the current ( 1Nteach ) and previous ( 0Nteach ) school year. Separately they reported the number of teachers new to the school this year ( 1New ). The turnover rate is calculated as: 33 1 1 0 0100[ - ( - )] /TURNOVER New Nteach Nteach Nteach (3) where, in instances that there are more teachers in the current year’s staff than in the previous year’s staff ( 1 0Nteach Nteach ), the number of new teachers not attributable to the overall increased number of staff represent the number of teachers who left the school and were replaced, which is calculated as a proportion of the previous year’s staff. For schools with the same or fewer staff in the current year than in the previous year ( 1 0Nteach Nteach ), the number of new teachers subsumes those previous teachers who left and were replaced and those who were not replaced. Hypothesis 2, the proposed attenuation hypothesis wherein SCHFOC might serve as a buffer against the deleterious effects of student misconduct (DISORDER) on teacher morale and retention, was also analyzed through a multiple regression equation of school-level variables. The standardized cross-product term is included to assess for the presence of moderation, as follows: 0 1 2 3( ) ( ) ( )j j j j j jMORALE DISORDER SCHFOC DISORDER SCHFOC u         (4) where the morale score for school j is a function of the grand mean of school morale ( 0 ), the regression coefficient for DISORDER in school j ( 1 ), the regression coefficient for SCHFOC in school j ( 2 ), the regression coefficient for the j jDISORDER SCHFOC standardized cross-product in school j ( 3 ), and the error for school j ( ju ). As before, an identical equation was run with turnover as the outcome variable. 34 Finally, the third hypothesis posited that within-school agreement about the degree of school focus would strengthen the expected positive relationship between focus and the expected negative relationship between focus and teacher attrition. This within-group consensus of climate perceptions, or climate strength, is typically measured using within-group variance or some dispersion measure of individual climate responses (Chan, 1998). Climate strength is often quantified as the inversion of variance among raters (González-Romá & Peiró, 2014), and the Average Deviation Index is one supported method for capturing that variance (Burke et al., 1999; Dumay, 2009). As such, Hypothesis 3 used the ADI across focus ratings for teachers within a given school (FOC_ADI) as a moderator term alongside school focus with morale and turnover, respectively, as outcome variables. That term is calculated as follows (Burke & Dunlap, 2002): 1 N jk j k j X X AD N     (5) where N is the number of teachers within a school responding to an item j, jkX is teacher k’s rating on item j, and jX is the mean of the scores on item j among teachers in a given school. Then, MORALE, SCHFOC, and FOC_ADI being school- level variables, a regression equation similar to that in equation 4 included those terms and the standardized cross-product term of school focus and schools’ focus ADI: 35 0 1 2 3( ) ( _ ) ( _ )j j j j j jMORALE SCHFOC FOC ADI SCHFOC FOC ADI u         (6) In addition, the same equation was carried out with school teacher turnover rates replacing school morale as the predicted outcome. 36 Results Table 2 shows the bivariate correlations between predictor and outcome variables of interest. Tables 3-6 below present the outcomes of hypothesis testing. Table 2 Bivariate Correlations Between School Morale, Teacher Turnover Rates, and Predictor Variables Outcome Variable Predictor School Morale School Focus Admin. Leadership Classroom Disorder School Morale ─ .846** .775** -.535** Teacher Turnover Rates -.139* -.081 -.101 .257** Note. *p < .05. **p < .01. The first hypothesis was that school focus (SCHFOC) would predict higher school morale and lower teacher turnover, independent of other related perceptions of administrator leadership (LEADER). To address this, I ran a block regression model with LEADER entered into the first block and SCHFOC added in the second block. In addition to standardized beta coefficients for LEADER and SCHFOC, R2 values for each model are presented in Tables 3 and 4 below, as well as the change in R2 which is accounted for by the presence of SCHFOC. 37 Table 3 Standardized Beta Coefficients and R-squared values in the School Morale as Outcome Models β R2 R2Δ LEADER SCHFOC MODEL 1 .775** ─ .601** ─ MODEL 2 .253** .640** .738** .137** Note. *p <.05. **p <.01. Table 4 Standardized Beta Coefficients and R-squared values in the School Turnover Rate as Outcome Models β R2 R2Δ LEADER SCHFOC MODEL 1 -.101 ─ .010 ─ MODEL 2 -.104 .004 .010 .000 Note. *p <.05. **p <.01. Table 3 shows that, as expected, both LEADER and SCHFOC together showed a strong positive correlation with school morale (R2=.738). The addition of SCHFOC also accounted for an increase in R2 of .137 when controlling for LEADER values. With both predictors in the model, SCHFOC demonstrated a stronger relationship with morale than did LEADER. Expected relationships did not emerge when turnover was the outcome variable (Table 4). LEADER and SCHFOC together did not significantly predict school turnover rates, with an overall R2 value of just .010, and the change in R2 accounted for by SCHFOC across all schools was smaller than .000. 38 Table 5 contains results of the proposed attenuation hypothesis, in which the outcome of interest is the regression of classroom disorder (DISORDER) on school morale and turnover rate, respectively. Beta coefficients are presented for the predictors DISORDER, SCHFOC, and the standardized cross-product moderator term (MODERATOR). Table 5 Standardized Beta Coefficients in the School Focus as Moderator Models Outcome Variable Predictor Variables DISORDER SCHFOC MODERATOR MORALE -.229** .753** -.018 TURNOVER .245** .005 -.110* *p <.05. **p <.01. Table 5 shows that classroom disorder did in fact have a significant negative effect on morale and a significant positive effect on turnover, as expected. There was no indication of the hypothesized attenuation effect by SCHFOC on the negative relationship between DISORDER and school morale. SCHFOC did demonstrate a stronger relationship with morale than did DISORDER. While SCHFOC itself showed no predictive relationship with turnover (as in earlier analyses), the presence of SCHFOC appears to have reduced the strength of the positive relationship between DISORDER and turnover. Lastly, Hypothesis 3 was that greater convergence in teacher perceptions of a school’s focus (climate strength of focus) would predict stronger relationships between school focus and morale and between school focus and turnover. In addressing that hypothesis, I ran regression models including the average deviation 39 index (ADI) for each school across all school focus rating items for all teachers in a given school (FOC_ADI). It is important to understand that a higher average deviation index indicates greater divergence among teacher ratings; lower ADI, as it is used in this study, is tantamount to greater climate strength. Table 6 displays beta coefficients for the predictors SCHFOC, FOC_ADI, and the standardized cross- product moderator term (MODERATOR). Table 6 Standardized Beta Coefficients in the Climate Strength as Moderator Models Outcome Variable Predictor Variables SCHFOC FOC_ADI MODERATOR MORALE .718** -.207** .045 TURNOVER -.004 .121* -.021 *p <.05. **p <.01. There was a significant negative relationship, as anticipated, between FOC_ADI and SCHFOC (r = -.656). That is, as teachers in a given school collectively endorsed a greater level of clarity and consistency in the school working environment, they tended to respond to those items with less interrater variability. Interestingly, while interrater agreement did not show a significant moderating effect on the correlation between school focus and either morale or turnover (Table 6), FOC_ADI by itself significantly predicted both outcomes. Schools in which teachers had greater variance in focus item responses had higher teacher turnover and lower aggregate perceptions of school morale. 40 Discussion Higher school focus significantly predicted higher school morale, consistent with expectations. Given the strength of the positive relationship between school focus and individual teacher job satisfaction found previously (Bovender, 2013), as well as extant literature regarding the value of teachers’ role clarity, this is hardly a startling result. Taken in sum with earlier findings, results here effectively demonstrate that this relationship largely holds across elementary, middle, and high schools, is not limited to public schools, and is not specific to urban, suburban, or rural schools. This study was the first to explore whether the positive relationships associated with school focus held once other evaluations of administrative leadership were controlled for. Prior to this study, it was not clear whether teacher ratings reflected focus as a characteristic of the school environment or if focus ratings were a proxy for teachers’ overall perceptions of administrators. There is some conceptual overlap between the Administrator Leadership and School Focus scales in this study, and parsing between the two is not as straightforward as I have described it here. Administrators do not operate in an organizational vacuum, and their practices may be determined by organizational characteristics more than personal characteristics. We can expect that good principals will actively work to create clear expectations (Conley & You, 2009; Reyes & Imber, 1992), but disorganized school environments might it difficult to communicate goals and maintain consistent priorities, regardless of actual leadership practices. Results here at least support describing focus as a characteristic of schools as workplaces. 41 School focus did not have a significant effect on turnover rates. There was an expectedly negative but nonsignificant relationship between the two as a zero-order correlation (Table 2). Once administrative leadership perceptions were controlled for, even that nonsignificant effect disappeared. Individual teachers’ perceptions of school focus have shown a strong negative correlation with teachers’ intent to quit in a smaller sample of elementary schools (Bovender, 2013). Turnover intentions are frequently the most effective predictor of subsequent turnover (Griffeth et al., 2000; Mobley et al., 1977; Richardson et al., 2008; Van Breukelen et al., 2004), so the expectation that low focus would lead to actual turnover seems defensible enough. However, it should be mentioned that in that Bovender (2013) analysis, aggregated school-level focus did not have a significant correlation with turnover intentions. Present evidence does not support the assertion that school focus as an environmental school characteristic increases teacher retention, in and of itself. This does not dismiss the importance of the inverse, that lower turnover can help schools maintain some stability and consistency in teacher norms and expectations, but this effect might be more important on an individual or smaller group level. Beginning teachers, for instance, likely benefit from having higher-tenured colleagues available for mentoring, induction, and socialization to norms (Borman & Dowling, 2008; Ingersoll, 2004; Perry & Hayes, 2011; Shen, 1997). Hypothesis 2, the proposed moderating effect of school focus on the regression of school morale and turnover, was partly supported. One of the aims of the present study was to extend the findings from the Bovender, 2013 study, which did not find direct evidence of this effect, but which was limited by a narrow range of 42 teachers’ student misconduct ratings. Results here suggest that focus may buffer against student misconduct leading to turnover, though the same cannot be said for focus offsetting the negative impact of student misconduct on schoolwide teacher morale. This does not necessarily discount the possibility that some teachers who feel overwhelmed with managing student conduct would be more satisfied if their school provided a consistent, concise set of priorities. On the whole, evidence suggests that increased focus may warrant consideration as an avenue for school-level intervention to support teacher retention in the face of student misconduct, that consummate source of teacher stress and burnout (Friedman, 2000; Schaufeli & Buunk, 2003; Shernoff et al., 2011). That said, it is worth noting that reliance on teacher reports to measure student misconduct in their classroom may have weakened the power to detect a moderating effect of school focus on the relationship between student behavior and morale. The concreteness of items in the Classroom Disorder scale (Appendix B, Table 15) hopefully militates against teacher perceptual bias, but different teachers, in all likelihood, perceive different frequency and severity of behavior problems, even in classrooms with identical student behavior. If school focus does provide the hypothesized buffer, this effect may show up in the behavior ratings themselves. A teacher who is less affected by classroom misconduct because of higher school focus might report that his or her students don’t misbehave as often as the students objectively do; in effect, the teacher’s perceptions of school focus may have already moderated the importance of students’ behavior regarding the teacher’s perception of school morale, but the moderating effect won’t be observable through teacher ratings. 43 Data came from 1st-year principal responses (for turnover) and 2nd-year responses for teachers. As a result, turnover rates came from the year prior to teacher ratings. It would have been preferable to have turnover rates for the same year as teacher responses, though it does seem reasonable to think that school morale, school focus, student conduct, and perceptions of administrative leadership would be mostly stable one year apart, except in instances where there was also principal turnover. Because principal tenure was not included in the 2nd-year principal survey, there was no way to detect principal turnover. The fact that turnover preceded teacher ratings might also suggest that the inferred causality, that focus protects against turnover, is actually the other way around. It seems likely that higher turnover could have led to less focus because new teachers (who were hired to replace the teachers who left after the previous year) were still figuring out how to balance the school’s expected teacher roles and priorities. Another possibility is that newer teachers, freshly trained and inducted into their respective schools’ expectations, felt that their goals were clear but felt less prepared to manage student misconduct. As such, it cannot be definitively stated the buffering hypothesis represents an actual phenomenon, and replication using turnover rates from the same year that teachers rate their perceptions of the school environment is needed before school focus can be fully considered as a protective factor against teacher attrition. Hypothesis 3, concerning the proposed moderator function of interrater agreement, as operationalized by the inverse of the average deviation index (ADI) for school focus items, was not supported. School focus did not appear to have a more 44 substantive effect on morale and turnover in schools where they was greater “focus” about the degree of focus. However, the significant negative relationship between school focus and school focus ADI does seem to provide support for the construct validity of school focus beyond what is given by scale reliabilities. If clarity and consistency in job expectations, etc., exists as a felt experience in the school working environment, it makes intuitive sense that teachers would have a clearer and more consistent shared experience of it when there is more of it. The additional significant negative correlation between focus ADI and morale and significant positive correlation between focus ADI and turnover also make sense if one follows the thinking that the climate strength of focus is largely tantamount to focus itself. It may be that the small, nonsignificant moderation of interrater agreement on the focus- outcome regressions has something to do with this conceptual redundancy – if there is indeed high schoolwide consensus, the degree of consensus may not serve as a moderator – though in the case of focus and turnover, there was not much of a relationship to moderate. As mentioned above, others have examined climate strength as a moderator of different organizational climate elements relating to different workplace outcomes, so far with mixed results (Colquitt et al., 2002; Du et al., 2015; Gonzalez-Roma et al., 2002; Lindell & Brandt, 2000; Rafferty & Jimmieson, 2010; Sanders et al., 2008; Schneider et al., 2002; Schyns & Van Veldhoven, 2010). When and how these relationships emerge and what measurement issues might limit them is not yet clearly established (Schneider et al., 2013). 45 Limitations It should be noted that analyses in which schools’ turnover rates served as the predicted outcome required both teacher and principal responses for a given school (see Table 1). Schools without available turnover rates in this sample may have had important differences from the other schools. In the context of examining perceptions of administrators and expectations about what staff should be doing, this does not seem like a small omission. One could argue that a certain amount of staff morale, school focus, and administrator favorability would be needed for teachers to take on the added task of filling out nonessential paperwork (these were not short surveys), as well as the need for sufficient communication to let teachers know about the surveys in the first place. Indeed, in comparing schools with teacher and principal responses to schools with just teacher responses, the latter group had an average of 7% lower school focus ratings, 13% lower morale ratings, and 7% lower administrative leadership ratings than the former group. Unfortunately, schools with only principal responses could not be used in the analyses presented here. The net result of schools’ nonparticipation (47.6% of schools solicited for the teacher survey in Year 2 responded) is that schools in this study sample, while containing a roughly even representation across location types and grade levels, may be substantively different than schools which self-selected out of the study. For example, the average turnover rate of schools retained in the study sample was 8%, while the national average at the time of data collection was roughly 13%. While this is an obvious downside of 46 conducting such large-scale survey research, the tradeoff is the statistical power made possible by having over 400 schools that did return teacher surveys. Turnover did demonstrate some of the expected relationships with predictor variables, but the meaningfulness of those relationships, as well as the power to detect them, was limited by the reliance on a single year’s concrete turnover rate (Morrell, 2016). Several schools with turnover rates above 30% were excluded from turnover analyses because of their high leverage, but in the absence of multi-year data, it cannot be assumed that these schools had an anomalously high turnover for that one year. There is also no way to tell whether any of the schools in the sample had aberrantly low turnover for the year. The large sample size and the fairly even distribution of schools across grade levels and locations seems sufficient to offset the statistical noise in turnover rates sample-wide, but there are other limitations to this approach that warrant discussion. While actual attrition rates are valuable – after all, for reasons described above, they are an important outcome – the rates are presumed for this study to imply some meaningful level of collective dissatisfaction. However, analyses here effectively aggregated a series of dichotomous outcomes (a teacher leaves or does not), which may or may not have overlapped from teacher to teacher. This approach to turnover measurement treats leavers as a homogenous group (Campion, 1991), but turnover is a highly personal decision often influenced by life circumstances beyond the work environment (Borman & Dowling, 2008; Grissom, Viano, & Selin, 2015). In addition, there can be important differences between those who leave a particular school to teach somewhere else and those who leave the profession entirely, and 47 predictors of the former are not necessarily predictors of the latter (Imazeki, 2005; Kukla-Acevedo, 2009). There is an unequivocal link between individuals’ job dissatisfaction and their intentions to quit their respective jobs (Bovender, 2013; Hom & Kinicki, 2001; Liu & Meyer, 2005), though some teacher turnover research has suggested that expressed quitting intentions have a relatively weak relationship with actual turnover (Ladd, 2011). Still, teachers who do quit frequently report being motivated by job dissatisfaction when deciding to leave (Ingersoll, 2001). Extending that expectation to actual aggregated turnover and aggregated morale requires more caution in drawing inferences. Finally, it may be reasonable to assume that the bulk of teachers who left did so voluntarily (Keigher, 2010; Goldring et al., 2014), but that does not necessarily mean that these departures were avoidable (Campion, 1991; Grissom, 2011). The ideal resolution to this turnover measurement concern would probably be to collect longitudinal data of actual turnover along with teacher-reported quitting intentions. Aggregation across all variables of interest in the present study introduced some ambiguity in parsing the implications of those variables at different levels of analysis. Conceptually, aggregating teacher ratings of school focus and teacher morale seems the most appropriate; both are considered characteristics of the work environment and both questionnaires are composed of appropriately-worded referent- shift consensus items clearly intended to elicit school-level perceptions (Glick, 1985; LeBreton & Senter, 2008). Each teacher has their own perspective and their own experiences that color their view of the school as a whole, but the intragroup consensus view largely defines these constructs and their felt experience within the 48 school (Chan, 1998; Griffith, 2006; Ostroff et al., 2003). Though these views are subjective, where interprofessional working factors are concerned, teachers are the most accurate source of data (Teddlie & Reynolds, 2000). Aggregating perceptions of administrative leadership seems appropriate to maintain school-level considerations, but this does overlook potentially valuable information regarding the connection between teachers’ evaluations of school leaders and their perception of school focus. That relationship may be more apparent at the level of individual response. Classroom disorder is more problematic as an aggregated variable, because wording of items dealt with individuals’ classroom experience. Certainly there are differences in student behavior between schools (Loeb & Darling-Hammond, 2005; Mehta et al., 2013), but the experiences of teachers in their respective classroom may differ within the same school, and those differences may be especially important when each teacher considers staying or leaving his or her respective school. This could have implications for the supposition that school focus helps teachers weather the storm when they are faced with disorderly students. These data are admittedly rather old. Since the time of data collection, the school accountability movement, particularly in the wake of No Child Left Behind (2001), has continued to push standardized test performance to the top of schools’ list of priorities (Conley & You, 2009; Perlstein, 2007; Ravitch, 2010). One could argue that the increased emphasis on testing might lead to clearer and more consistent objectives without necessarily contributing to higher morale. On the other hand, one can argue that the pressures of high-stakes testing supersede the array of other teacher roles and responsibilities without necessarily guiding teachers on how to manage their 49 priorities in response. Regardless, while more recent data would be much preferred, school structures and the nature of working in schools are slow to change relative to many other professions (Evans, 2011). It may not always be beneficial that school practices are slow to change, but for purposes of this study, it lends a reasonable expectation that these results are applicable to teaching now. If this line of research is continued, collecting new data will be essential. Summary and conclusions Results of this study provide adequate evidence to place school focus within the hierarchy of schools’ needs for maintaining or improving morale across many school contexts. Focus is not an all-or-nothing proposition, however, and hypotheses here may have considered focus to be a more fundamental need than warranted. More basic needs like teacher safety and adequate supplies in the school building might need to be met before interpersonal working dynamics become a primary concern (Bolin, 2007; Ingersoll, 2001). Among those workplace dynamics, knowing or not knowing what to do is one part of a larger context. For example, when considering the value of consistent and clearly-expressed goals for teachers, it is important to also consider what the goals are, who determines the goals, how they determine the goals, and the level of input teachers have into the goals they are expected to pursue. Principals and administrators generally set the agenda, but the workplace culture regarding norms and expectations arises from the collective perceptions of all those who work in the school (Schein, 2010). Furthermore, teachers who elect to transfer or quit often cite having too little professional autonomy as a motivation for their choices (Boyd et al, 2011; Ingersoll, 2001), and while role clarity is important, 50 teachers typically want to have some say in what their roles are rather than simply having them handed down (Perie & Baker, 1997; Richardson, Alexander, & Castleberry, 2008). So in a school where teachers’ basic needs are met, there might be a point at which focus starts to feel constricting when there is too much of a good thing. It may be that in schools where student behavior is not too disruptive, and in which administrative and collegial emotional support is adequate enough, that teachers are significantly more satisfied when they know how to structure their time and how to work towards key goals, but that having a high level of focus is not such a fundamental need that it can overcome the negative effects of too much disruptive behavior or too little administrative or collegial support. There has been some recent discussion of “multiple climates” in organizational research (Schneider et al., 2013; Zohar & Hoffman, 2012), and a school’s “focus climate” likely influences and is influenced by other climate elements. However, little work at this point has addressed either the conceptualization or practical implications of multiple climates, in schools or elsewhere (Kuenzi & Schminke, 2009). In addition, despite an expansive literature on the leadership styles, Fulmer and Ostroff (2016) noted in their review earlier this year that specific leader behaviors that facilitate the emergence of group-level climate factors have not been well studied. Results in this nationwide sample, broadly correlational as they may be, provide at least provisional support for the potential value of focusing on focus. I do not expect that school focus interventions can make up for deficits elsewhere in schools’ workplace functioning or that focus could be a panacea for teacher turnover, 51 but the findings in this sample suggest that school focus may function as a protective factor against the work-related stress that leads to teachers leaving. Classroom disorder is only one source of teacher stress, but it is clearly a substantial antecedent of teacher attrition and unique to the school environment. However, as turnover is a complex individual decision, broad correlations can only inform so much. Results in this study, due to complications in operationalizing constructs at different levels of analyses, do little to explain the influence of group- level consensus variables on individual outcomes (Glick, 1985; Lebreton & Senter, 2008). Connecting workplace factors to individual outcomes through multi-level approaches like hierarchical linear modeling (Raudenbush & Byrk, 2002) or multilevel factor analysis (Dunn et al., 2014) is ideal for studying school environments because schools typically have similar functions and structures (Bovender, 2013; Griffith, 2006; Lee, 2000), making between-school differences in teacher outcomes more clearly attributable to differences in the way schools address the same needs (Bryk & Driscoll, 1988). School literature has a robust knowledge base about individual teacher factors that predict individual job satisfaction and individual turnover, and more recently, a fair amount of attention has been paid to system-level characteristics of schools that predict the same. Bringing together the micro and macro-factors that inform teacher experiences (Moore, 2012) can help us move beyond “who”, “what”, “when”, “where”, and “why” of teacher satisfaction and morale to the key question of “how” to introduce positive school change. To that end, clarifying what discrete practices promote focus in the school setting is an important next step if this study is to lead to any practical applications. 52 Some efforts to target school-level focus-related interventions have led to desired outcomes. For example, analogous programs targeting clarity and consistency in student conduct expectations and consequences have been shown to effectively reduce classroom disorder (Gottfredson, 1987; Gottfredson, Gottfredon, & Hybl, 1993). School staff worked together through cooperative planning, problem-solving, and decision-making to establish regular rules and predictable responses to rule infractions or adherence. When well-implemented, students reported more orderly classrooms and teachers reported improved student behavior. It seems reasonable to expect that similarly cooperative efforts between teachers and administrators to establish clearly understood staff priorities and expectations could increase teacher morale and school functioning. Beckard’s model for team building (1972) seems well-suited for conceptualizing that process (as presented in Burke, 2011): 1. Set goals or priorities 2. Analyze or allocate the way work is performed according to team members’ roles and responsibilities 3. Examine the way the group is working – its processes, such as norms, decision making, communications, and so on 4. Examine interpersonal relationships among members Ultimately, any such programmatic interventions will need to be evaluated experimentally, beyond the correlational results reported so far, but in an era of numerous and often conflicting role demands for teachers, I imagine nearly any hypothetical increase in teacher-administration communication and agreement to be positive. At the same time, introducing any additional school change initiatives, 53 focus-oriented or otherwise, necessitates those initiatives being well-vetted and worth the effort, lest they introduce further ambiguous demands and chaos. 54 Appendices Appendix A: Sample characteristics and response rates Table 7 Proportion of Schools by Auspices School Type Proportion of Sample N Schools % of Sample Public 316 90.8 Private 18 5.2 Religious 14 4.0 Table 8 Proportion of Teachers by Age Range Age Range Proportion of Sample N Teachers % of Sample 29 or younger 1292 11.4 30-39 2420 21.4 40-49 3490 30.9 50-59 3522 31.2 60 or older 564 5.0 55 Table 9 Proportion of Teachers by Ethnicity Ethnicity Proportion of Sample N Teachers % of Sample White 9224 81.1 Black 742 6.5 Hispanic 612 5.4 Asian 130 1.1 American Indian 89 0.8 Table 10 Proportion of Teachers by Sex Sex Proportion of Sample N Teachers % of Sample Female 7256 63.8 Male 4081 35.9 Table 11 Teacher Questionnaire School Participation Percentages by Grade Level and Location Type Location Type Middle Schools High Schools Combined School Levels N respond (N sample) % respond N respond (N sample) % respond N respond (N sample) % respond Rural 81 (137) 59.1 75 (145) 51.7 156 (282) 55.3 Suburban 70 (150) 46.7 54 (137) 39.4 124 (287) 43.2 Urban 70 (144) 48.6 53 (134) 39.6 123 (278) 44.2 Combined Locations 221 (431) 51.3 182 (416) 43.8 403 (847) 47.6 56 Table 12 Principal Year 1 Questionnaire Response Percentages by Grade Level and Location Type Location Type Middle Schools High Schools Combined School Levels N respond (N sample) % respond N respond (N sample) % respond N respond (N sample) % respond Rural 95 (137) 69.3 106 (145) 72.4 201 (282) 71.3 Suburban 105 (150) 70.0 85 (137) 62.0 190 (287) 66.2 Urban 88 (144) 61.1 79 (134) 59.0 167 (278) 60.1 Combined Locations 288 (431) 66.8 270 (416) 64.9 558 (847) 65.9 57 Table 13 Survey Participation Rates by School Auspices Survey Public Schools Private Schools Religious Schools N respond (N sample) % respond N respond (N sample) % respond N respond (N sample) % respond Principal Year 1 696 (1041) 66.9 89 (149) 59.7 63 (88) 71.6 Teacher 359 (711) 50.5 15 (31) 48.4 29 (105) 27.6 Note. Principal survey samples included elementary, middle and high schools. Teacher survey samples included only middle and high schools. 58 Appendix B: Scale items and reliabilities Table 14 Teacher Organizational Focus Scale: Items and Reliability Item Corrected Item-total Correlation Squared Multiple Correlation (R2) Alpha if Item Deleted This school clearly signals to faculty and staff what performance is expected of them. .650 .503 .928 Rules and operating procedures are clear and explicit in this school. .687 .569 .927 It is difficult to determine what is expected of a person in this school.a .679 .522 .927 The goals of this school are clear. .685 .555 .927 Everyone understands what behavior will be rewarded in this school. .642 .467 .928 Some persons in positions of power or authority in this school have conflicting expectations for others.a .609 .435 .928 Everyone here is working towards the same ends. .672 .478 .927 In this school, people who accomplish the same thing are rewarded in the same way. .605 .416 .928 People are often confused about what objective they should go for in this school.a .736 .577 .926 In this school, people know what to do and when to do it. .687 .498 .927 People know how to achieve rewards here. .606 .446 .928 People have often said that it is difficult to decide what aims to work towards in this school.a .747 .598 .925 This school simultaneously pursues many conflicting goals.a .681 .498 .927 My school has a clear focus. .793 .671 .925 My school is torn up by leaders with different agendas.a .688 .518 .927 Rules and procedures are often ignored in this school.a .690 .519 .926 Note. Alpha = .931. Respondents were presented with a list of statements to show how well each described their school. Item responses were “false”, “mostly false”, “mostly true”, and “true”. a = Item is reverse scored. 59 Table 15 Teacher Classroom Disorder Scale: Items and Reliability Item Corrected Item-total Correlation Squared Multiple Correlation (R2) Alpha if Item Deleted Students pay attention in class.a .563 .420 .919 Students take things that do not belong to them. .572 .415 .919 Students do what I ask them to do.a .555 .417 .919 Students destroy or damage property. .601 .450 .918 Students talk at inappropriate times. .653 .475 .916 Students make disruptive noises (like yelling, animal noises, tapping, etc.). .740 .577 .913 Students try to physically hurt other people (by tripping, hitting, throwing objects, etc.). .692 .553 .915 Students tease other students. .620 .479 .917 Students make threats to others or curse at others. .691 .566 .915 I spend more time disciplining than I do teaching. .717 .604 .914 Students are distracted by the misbehavior of other students. .705 .548 .914 The classroom activity comes to a stop because of discipline problems. .707 .560 .914 How much of your time in the classroom is directed to coping with disruptive student behavior? .566 .404 .920 How much does the behavior of some students in your classroom keep you from teaching? .704 .580 .915 Note. Alpha = .922. Responses for the first 12 items were “almost always”, “often”, “sometimes”, “seldom”, and “never”. Responses for the next item were “none of my time”, “some time each day”, “about half of my time”, and “most of my time”. Responses for the last item were “a great deal”, “a fair amount”, “not very much”, and “not at all”. a = Item is reverse scored. 60 Table 16 Administrator Leadership Scale: Items and Reliability Item Corrected Item-total Correlation Squared Multiple Correlation (R2) Alpha if Item Deleted There is little administrator- teacher tension in this school. .548 .324 .861 Our principal is a good representative of our school before the superintendent and the board. .644 .434 .853 The principal encourages experimentation in teaching. .482 .247 .864 The principal lets staff and students know when they have done particularly well. .578 .340 .857 The administration is supportive of teachers. .681 .481 .848 The principal of our school is open to staff input. .663 .505 .850 The principal of our school plans effectively. .593 .374 .855 Note. Alpha = .855. Responses of above items were “true” or “false”. 61 Table 17 Teacher Morale Scale: Items and Reliability Item Corrected Item-total Correlation Squared Multiple Correlation (R2) Alpha if Item Deleted Our problems in this school are so big that it is unrealistic to expect teachers to make much of a dent in them.a .313 .107 .792 I feel my ideas are listened to and used in this school. .471 .244 .774 The teaching faculty of our school are apathetic.a .300 .130 .796 The teaching faculty of our school are cohesive. .459 .234 .776 The teaching faculty of our school are enthusiastic. .493 .280 .772 The teaching faculty of our school are frustrated.a .573 .384 .760 The teaching faculty of our school are satisfied. .582 .377 .758 The teaching faculty of our school are tense.a .573 .361 .760 The teaching faculty of our school are unappreciated.a .546 .323 .763 Note. Alpha = .793. Responses of above items were “true” or “false”. a = Item is reverse scored. 62 Appendix C: Descriptive statistics Table 18 Descriptive Values for School Morale by Grade Level and Location Type Location Type Middle Schools High Schools Combined School Levels MEAN (SD) MAX (MIN) MEAN (SD) MAX (MIN) MEAN (SD) MAX (MIN) Rural .655 (.149) .940 (.240) .644 (.153) .940 (.310) .650 (.150) .940 (.240) Suburban .656 (.148) .950 (.340) .632 (.184) 1.00 (.320) .647 (.162) 1.00 (.320) Urban .601 (.158) .930 (.300) .618 (.154) .930 (.240) .608 (.156) .930 (.240) Combined Locations .638 (.153) .950 (.240) .632 (.153) 1.00 (.240) .636 (.156) 1.00 (.240) Note. Possible range of values per item is 0-1. Table 19 Descriptive Values for School Focus by Grade Level and Location Type Location Type Middle Schools High Schools Combined School Levels MEAN (SD) MAX (MIN) MEAN (SD) MAX (MIN) MEAN (SD) MAX (MIN) Rural 2.00 (.359) 2.84 (.880) 1.95 (.426) 2.75 (.970) 1.98 (.389) 2.84 (.880) Suburban 1.96 (.383) 2.73 (.900) 1.90 (.335) 2.65 (1.33) 1.93 (.365) 2.73 (.900) Urban 1.87 (.375) 2.78 (1.07) 1.82 (.342) 2.44 (.990) 1.85 (.361) 2.78 (.990) Combined Locations 1.95 (.374) 2.84 (.880) 1.90 (.378) 2.75 (.970) 1.93 (.376) 2.84 (.880) Note. Possible range of values per item is 0-3. 63 Table 20 Descriptive Values for Classroom Disorder by Grade Level and Location Type Location Type Middle Schools High Schools Combined School Levels MEAN (SD) MAX (MIN) MEAN (SD) MAX (MIN) MEAN (SD) MAX (MIN) Rural 1.72 (.241) 2.24 (1.26) 1.54 (.250) 2.43 (1.08) 1.64 (.260) 2.43 (1.08) Suburban 1.69 (.240) 2.44 (1.30) 1.51 (.309) 2.26 (.830) 1.62 (.280) 2.44 (.830) Urban 1.88 (.275) 2.54 (1.47) 1.57 (.264) 2.09 (1.02) 1.75 (.310) 2.54 (1.02) Combined Locations 1.76 (.264) 2.54 (1.26) 1.54 (.270) 2.43 (.830) 1.67 (.288) 2.54 (.830) Note. Possible range of values per item is 0-4. Table 21 Descriptive Values for Administrator Leadership by Grade Level and Location Type Location Type Middle Schools High Schools Combined School Levels MEAN (SD) MAX (MIN) MEAN (SD) MAX (MIN) MEAN (SD) MAX (MIN) Rural .729 (.165) 1.00 (.250) .796 (.159) 1.00 (.370) .794 (.162) 1.00 (.250) Suburban .789 (.157) .980 (.360) .775 (.145) 1.00 (.410) .784 (.152) 1.00 (.360) Urban .762 (.158) .990 (.260) .761 (.141) .970 (.440) .762 (.151) .990 (.260) Combined Locations .782 (.160) 1.00 (.250) .779 (.149) 1.00 (.370) .781 (.156) 1.00 (.250) Note. Possible range of values per item is 0-1. 64 Table 22 Descriptive Values for School Teacher Turnover Rate by Grade Level and Location Type Location Type Middle Schools High Schools Combined School Levels MEAN (SD) MAX (MIN) MEAN (SD) MAX (MIN) MEAN (SD) MAX (MIN) Rural 7.22 (6.05) 20.83 (0.00) 8.20 (6.12) 21.57 (0.00) 7.66 (6.07) 21.57 (0.00) Suburban 8.91 (5.64) 22.73 (0.00) 7.25 (5.60) 16.67 (0.00) 8.27 (5.65) 22.73 (0.00) Urban 9.27 (6.04) 21.43 (0.00) 6.23 (4.14) 14.42 (0.00) 8.02 (5.52) 21.43 (0.00) Combined Locations 8.40 (5.95) 22.73 (0.00) 7.35 (5.46) 21.57 (0.00) 7.96 (5.76) 22.73 (0.00) 65 Bibliography Abramis, D. J. (1994). Work role ambiguity, job satisfaction, and job performance: Meta-analyses and review. Psychological Reports, 75, 1411-1433. Archbald, D. A. & Porter, A. C. (1994). Curriculum control and teachers’ perceptions of autonomy and satisfaction. Educational Evaluation and Policy Analysis, 16, 21-39. Bakker, A. B., Hakanen, J. J., Demerouti, E., & Xanthopoulou, D. (2007). Job resources boost work engagement, particularly when job demands are high. Journal of Educational Psychology, 99, 274-284. Barrett, S. B., Bradshaw, C. P., & Lewis-Palmer, T. (2008). Maryland statewide PBIS initiative: Systems, evaluation, and next steps. Journal of Positive Behavior Interventions, 10, 105-114. Baughman, K. S. (1996). Increasing teacher job satisfaction: A study of the changing role of the secondary principal. American Secondary Education, 24, 19-22. Beckhard, R. (1972). Optimizing team-building efforts. Journal of Contemporary Business, 1, 23-32. Bektas, F., Ocal, M. F., & Ibrahim, A. (2012). School culture as the predictor of job Satisfaction with respect to teachers’ perceptions: A causal study. New Educational Review, 27, 295- 305. Bierlein, L. A. 1993. Controversial issues in education policy. Newbury Park, CA: Sage. 66 Bliese, P. (2000). Within-group agreement, non-independence, and reliability: Implications for data aggregation and analysis. In K. J. Klein & S. W. Kozlowski (Eds.). Multilevel theory research and methods in organizations: Foundations, extension, and new directions. San Francisco: Jossey-Bass. Bogler, R. (2001). The influence of leadership style on teacher job satisfaction. Educational Administration Quarterly, 37, 662-683. Bolin, F. (2007). A study of teacher job satisfaction and factors that influence it. Chinese Education and Society, 40, 47-64. 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, 367-409. Bovender, W. (2013). Organizational focus as a moderator of the relation between student externalizing behavior and teacher job satisfaction. Unpublished master’s thesis, University of Maryland, College Park. 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, 303-333. Boykin, A. W., Tyler, K. M., Watkins-Lewis, K. M., & Kizzie, K. (2006). Culture in the sanctioned classroom practices of elementary school teachers serving low- income African American students. Journal of Education of Students Placed At-Risk, 11, 161–173. 67 Bradshaw, C. P., Koth, C. W., Bevans, K. B., Ialongo, N., & Leaf, P. J. (2008). The impact of school-wide positive behavioral interventions and supports (PBIS) on the organizational health of elementary schools. School Psychology Quarterly, 23, 462-473. Bryk, A. S., & Driscoll, M. E. (1988). The high school as community: Contextual influences and consequences for students and teachers. Madison, WI: National Center on Effective Secondary Schools, University of Wisconsin- Madison. (ERIC Document Reproduction Service No. ED 302539). Burke, M. J., Finkelstein, L. M., & Dusig, M. S. (1999). On average deviation indices For estimating interrater agreement. Organizational Research Methods, 2, 49- 68. Burke, M. J., & Dunlap, W. P. (2002). Estimating interrater agreement with the Average deviation index: A user’s guide. Organizational Research Methods, 5, 159–172. Burke, W. W. (2011). Organization Change: Theory and Practice. Thousand Oaks, CA: Sage. Caprara, G. V., Barbaranelli, C., Steca, P., & Malone, P. S. (2006). Teachers’ self- Efficacy beliefs as determinants of job satisfaction and students’ academic achievement: A study at the school level. Journal of School Psychology, 44, 473-490. Cerit, Y. (2009). The effects of servant leadership behaviors of school principals on teachers’ job satisfaction. Educational Management Administration & Leadership, 37, 600-623. 68 Chan, D. (1998). Functional relations among constructs in the same content domain at Different levels of analysis: A typology of composition models. Journal of Applied Psychology, 83, 234-246. Cohen, J.C., McCabe, E.M., Michelli, N.M., & Pickeral, T. (2009). School climate: Research, policy, practice, and teacher education. Teachers College Record, 111, 180–213. Colquitt, J. A., Noe, R. A., & Jackson, C. L. (2002). Justice in teams: Antecedents and consequences of procedural justice climate. Personnel Psychology, 58, 83-109. Conley, S., & Woosley, S. (2000). Teacher role stress, higher order needs and work outcomes. Journal of Educational Administration, 38, 179-201. Conley, S., & You, S. (2009). Teacher role stress, satisfaction, commitment, and intentions to leave: A structural model. Psychological Reports, 105, 771-786. Conley, S. & You, S. (2014). Role stress revisited: Job structuring antecedents, work outcomes, and moderating effects of locus of control. Educational Management Administration & Leadership, 42, 184-206. Cotton, K. (2001). Schoolwide and classroom discipline. Retrieved October 17, 2015: http://educationnorthwest.org/sites/default/files/SchoolwideandClassroomDiscipline.pdf Cox, S., Parmer, T., & Tourkin, S. (2007). Documentation for the 2004-2005 Teacher Follow-up Survey (NCES 2007-349). U.S. Department of Education. Washington (DC): National Center for Education Statistics. 69 Demerouti, E., Bakker, A. B., Nachreiner, F., & Schaufeli, W. B. (2001). The job demands-resources model of burnout. Journal of Applied Psychology, 86, 499-512. Dickson, M. W., Resick, C. J., & Hanges, P. J. (2006). When organizational climate is unambiguous, it is also strong. Journal of Applied Psychology, 91, 351-364. Dinham, S. (2007). How schools get moving and keep improving: Leadership for Teacher learning, student success and school renewal. Australian Journal of Education, 51, 263-275. Dollard, M. F., Tuckey, M. R., & Dormann, C. (2012). Psychosocial safety climate moderates the job demand-resource interaction in predicting workgroup distress. Accident Analysis and Prevention, 45, 694-704. Dollard, M. F., Opie, T., Lenthall, S., Wakerman, J., Knight, S., Dunn, S., et al., (2012). Psychosocial safety climate as an antecedent of work characteristics and psychological strain: A multilevel model. Work and Stress, 26, 385-404. Du, J., Shin, Y., & Choi, J. N. (2015). Convergent perceptions of organizational efficacy among team members and positive work outcomes in organizational teams. Journal of Occupational and Organizational Psychology 88, 178–202. Dumay, X. (2009). Origins and consequences of schools’ organizational culture for student achievement. Educational Administration Quarterly, 45, 523-55. Dunn, E. C., Masyn, K. E., Jones, S. M., Subramanian, S. V., & Koenen, K. C. (2014). Measuring psychosocial environments using individual response: An application of multilevel factor analysis to examining students in schools. Prevention Science, 16, 718-733. 70 Elma, C. (2013). The predictive value of teachers’ perception of organizational justice on job satisfaction. Eurasian Journal of Educational Research, 51, 157-176. Evans, L. (1997). Addressing problems of conceptualization and construct validity in researching teachers’ job satisfaction. Educational Research, 39, 319-331. Farkas, G. (2003). Racial disparities and discrimination in education: What do we know, how do we know it, and what do we need to know? Teachers College Record, 105, 1119-1146. Farkas, S., Johnson, J., & Foleno, T. (2000). A sense of calling: Who teaches and why. New York: Public Agenda. Retrieved August 19th, 2012 from http://www.publicagenda.org/files/pdf/sense_of_calling.pdf Fulmer, C. A. & Ostroff, C. (2016). Convergence and emergence in organizations: An integrative framework and review. Journal of Organizational Behavior, 37, 122-145. Geving, A. M. (2007). Identifying the types of student and teacher behaviours associated with teacher stress. Teaching and Teacher Education, 23, 624–640. Gigante, N. A., & Firestone, W. A. (2008). Administrative support and teacher leadership in schools. Journal of Educational Administration, 46, 302-331. Gilboa, S., Shirom, A., Fried, Y., & Cooper, C. (2008). A meta-analysis of work demand stressors and job performance: Examining main and moderating effects. Personnel Psychology, 61, 227-271. Glick, W. H. (1985). Conceptualizing and measuring organizational and psychological climate: Pitfalls of multi-level research. Academy of Management Review, 10, 601-610. 71 Goldring, R., Taie, S., & Riddles, M. (2014). Teacher attrition and mobility: Results from the 2012-13 Teacher Follow-up Survey. (NCES 2014-077, U.S. Department of Education, National Center for Education Statistics). Washington, DC: U.S. Government Printing Office. González-Romá, V., Peiró, J. M., & Tordera, N. (2002). An examination of the antecedents and moderators influences of climate strength. Journal of Applied Psychology, 87, 465–473. González-Romá, V., & Peiró, J. M. (2014). Climate and culture strength. In B. Schneider & K. Barbera (Eds.), The Oxford handbook of organizational climate and culture, 496-531. New York, NY: Oxford University Press. Gottfredson, D. C. (1987). An evaluation of an organization development approach to reducing school disorder. Evaluation Review, 11, 739-763. Gottfredson, D. C., Gottfredson, G. D., & Hybl, L. G. (1993). Managing adolescent behavior: A multiyear, multischool study. American Educational Research Journal, 30, 179-215. Gottfredson, G. D., Gottfredson, D. C., Czeh, E. R., Cantor, D., Crosse, S. B., & Hantman, I. (2000). National study of delinquency prevention in schools. Final report for the National Institute of Justice, U.S. Department of Justice, Grant #96-MN-MN-008. Ellicott City, MD: Gottfredson Associates, Inc. Gottfredson, G. D., & Holland, J. L. (1996b). Organizational focus questionnaire. Ellicott City, MD: Gottfredson Associates, Inc. Gottfredson, G. D. (1999). User’s manual for the Effective School Battery. Ellicott City, MD: Gottfredson Associates, Inc. 72 Gottfredson, G. D. (2000, August). Environmental Focus in a Large National Sample of Schools. Paper presented at the annual meeting of the American Psychological Association, Washington, DC. Gottfredson, G. D., Gottfredson, D. C., Payne, A. A., & Gottfredson, N. C. (2005). School climate predictors of school disorder: Results from a national study of delinquency prevention in schools. Journal of Research in Crime and Delinquency, 42, 412-444. Griffin, M. A., Mathieu, J. E., & Jacobs, R. R. (2001). Perceptions of work contexts: Disentangling influences at multiple levels of analysis. Journal of Occupational and Organizational Psychology, 74, 563-579. Griffith, J. (2006). A compositional analysis of the organizational climate- performance relation: Public schools as organizations. Journal of Applied Social Psychology, 36, 1848-1880. Griffeth, R. W., Hom, P. W., & Gaertner, S. (2000). A meta-analysis of antecedents and correlates of employee turnover: Updates, moderator tests, and research implications for the next millennium. Journal of Management, 26, 463-488. 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, 2552-2585. Hanushek, E. A., Kain, J. F., & Rivkin, S. G. (2004). Why public schools lose teachers. Journal of Human Resources, 39, 326-354. Hargreaves, A. (2000). Four ages of professionalism and professional learning. Teachers and Teaching: History and Practice, 6, 151-182. 73 Hartnell, C. A., Ou, A. Y., & Kinicki, A. (2011). Organizational culture and organizational effectiveness: a meta-analytic investigation of the competing values framework. Journal of Applied Psychology, 96, 677-694. Hastings, R. P., & Bham, M. S. (2003). The relationship between student behavior and patterns of teacher burnout. School Psychology International, 24, 115- 127. Hernandez, D. J. (2004). Demographic change and the life circumstances of immigrant families. Future of Children, 14, 17-47. Holland, J. L. (1997). Making vocational choices: A theory of vocational personalities and work environments (3rd Ed.). Odessa, FL: PAR, Inc. Hom, P. W. & Kinicki, A. J. (2001). Toward a greater understanding of how dissatisfaction drives employee turnover. Academy of Management Journal, 44, 975-987. House, J.S. (1981). Work Stress and Social Support. Reading, MA: Addison-Wesley. Hulpia, H., Devos, G., & Van Keer, H. (2011). The relation between school leadership from a distributed perspective and teachers’ organizational commitment: Examining the source of the leadership function. Educational Administration Quarterly, 47, 728-771. Hussar, W. J., and Bailey, T. M. (2013). Projections of Education Statistics to 2022 (NCES 2014-051). U.S. Department of Education, National Center for Education Statistics. Washington, DC: U.S. Government Printing Office Imazeki, J. (2005). Teacher salaries and teacher attrition. Economics of Education Review, 24, 431-449. 74 Ingersoll, R. M. (2001). Teacher turnover and teacher shortages: an organizational analysis. American Educational Research Journal, 38, 499-534. Ingersoll, R. M. (2003). Is there really a teacher shortage? Seattle: University of Washington, Center for the Study of Teaching and Policy. Ingersoll, R. M. (2004). Why do high-poverty schools have difficulty staffing their classrooms with qualified teachers? Retrieved June 13, 2012, from http://www.americanprogress.org/kf/ingersoll-final.pdf. Ingersoll, R.M. and May, H. (2011). Recruitment, Retention and the Minority Teacher Shortage. Consortium for Policy Research in Education. CPRE Research Report #RR-69. Johnson, S. M., Kraft, M. A., & 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, 1-39. Jones, R. E. (1997). Teacher participation in decision-making – Its relationship to staff morale and student achievement. Education, 1, 76-83. Judge, T. A. & Kammeyer-Mueller, J. D. (2012). Job attitudes. Annual Review of Psychology, 63, 341-367. Keigher, A. (2010). Teacher attrition and mobility: Results from the 2008-09 Teacher Follow-Up Survey (NCES 2010-353, U.S. Department of Education, National Center for Education Statistics). Washington, DC: U.S. Government Printing Office. 75 Kelly, S. (2004). An event history analysis of teacher attrition: Salary, teacher tracking, and socially disadvantaged schools. Journal of Experimental Education, 72, 195-220. Kena, G., Hussar, W., McFarland, J., de Brey, C., Musu-Gillette, L., Wang, X. et al. (2016). The condition of education 2016 (NCES 2016-144, U.S. Department of Education, National Center for Education Statistics). Washington, DC: U.S. Government Printing Office. King, L. A. & King, D. W. (1990). Role conflict and role ambiguity: A critical assessment of construct validity. Psychological Bulletin, 107, 48-64. Kloep, M. & Tarifa, F. (1994). Working conditions, work style, and job satisfaction among Albanian teachers. International Review of Education, 40, 159-172. Koklu, M. (2012). Participations in decision making, desires for participation, job Satisfactions and conflict management styles of secondary education teachers. Education and Science, 37, 208-223. Kozlowski, S. W. J., & Klein, K. J. (2000). A multilevel approach to theory and research in organizations: contextual, temporal, and emergent processes. In K. J. Klein & S. W. J. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations (pp. 3–90). Kuenzi, M. & Schminke, M. (2009). Assembling fragments into a lens: A review, critique, and proposed research agenda for the organizational work climate literature. Journal of Management, 35, 634-717. 76 Kukla-Acevedo, S. (2009). Leavers, movers, and stayers: The role of workplace conditions in teacher mobility decisions. Journal of Educational Research, 102, 443-452. Kyriacou, C. (2001). Teacher Stress: directions for future research. Educational Review, 53, 27-35. Ladd, H. F. (2011). Teachers’ perceptions of their working conditions: How predictive of planned and actual teacher movement? Educational Evaluation and Policy Analysis, 33, 235-261. Landers, E., Alter, P., & Servilio, K. (2008). Students’ challenging behavior and teachers’ job satisfaction. Beyond Behavior, 18, 26-33. LeBreton, J. M. & Senter, J. L. (2008). Answers to twenty questions about interrater reliability and interrater agreement. Organizational Research Methods, 11, 815-852. Lee, V. E. (2000). Using hierarchical linear modeling to study social contexts: The case of school effects. Educational Psychologist, 35, 125-141. Leithwood, K., & McAdie, P. (2007). Teaching working conditions that matter. Education Canada, 47, 47-72. Lindell, M. K. & Brandt, C. J. (2000). Climate quality and climate consenses as mediators of the relationship between organizational antecedents and outcomes. Journal of Applied Psychology, 85, 331-345. 77 Littrell, P. C., Billingsley, B. S., & Cross, L. H. (1994). The effects of principal support on special and general educators’ stress, job satisfaction, school commitment, health, and intent to stay in teaching. Remedial and Special Education, 15, 297-310. Liu, X. X., & Meyer, J. P. (2005). Teachers’ perspective of their jobs: A multilevel analysis of the Teacher Follow-Up Survey for 1994-95. Teachers College Record, 107, 985-1003. Loeb, S., Darling-Hammond, L, & Luczak, J. (2005). How teaching conditions predict teacher turnover in California schools. Peabody Journal of Education, 80, 44-70. Ma, X., & MacMillan, R. B. (1999). Influences of workplace conditions on teachers’ job satisfaction. Journal of Educational Research, 93, 39-47. Marvel, J., Lyter, D. M., Peltola, P., Strizek, G. A., and Morton, B. A. (2006). Teacher attrition and mobility: Results from the 2004–05 Teacher Follow-Up Survey (NCES 2007–307, U.S. Department of Education, National Center for Education Statistics). Washington, DC: U.S. Government Printing Office. Mehta T. G., Atkins, M. S., & Frazier, S. L. (2013). The organizational health of \ urban elementary schools: School health and teacher functioning. School Mental Health, 5, 144-154. Mobley, W. H., Horner, S. O., & Hollingsworth, A. T. (1978). An evaluation of precursors of hospital employee turnover. Journal of Applied Psychology, 62, 237-240. 78 Montgomery, C. (2005). A Meta-Analysis for Exploring the Diverse Causes and Effects of Stress in Teachers. Canadian Journal of Education, 28, 458-486. Moore, C. M. (2012). The role of school environment in teacher dissatisfaction among U.S. public school teachers. SAGE Open. doi:10.1177/2158244012438888. Morrell, K. (2016). Understanding and measuring employee turnover. In G. Saridakis & C. L. Cooper (Eds.), Research Handbook on Employee Turnover, 26-58. Cheltenham, UK: Elgar. National Commission on Teaching and America’s Future. (2007). The high cost of teacher turnover. Washington, DC: Author. Newmann, F., Rutter, R., & Smith, M. (1989). Organizational factors that affect school sense of efficacy, community and expectations. Sociology of Education, 62, 221-238. Nielsen, M. B., Mearns, K., Matthiesen, S. B., & Eid, J. (2011). Using the job demands-resources model to investigate risk perception, safety climate and job satisfaction in safety critical organizations. Scandinavian Journal of Psychology, 52, 467-475. No Child Left Behind Act of 2001, U.S.C. 6319 § (2008). Nooteboom, B. (2009). A cognitive theory of the firm: Learning, governance and dynamic capabilities. Cheltenham, UK: Elgar. 79 Ostroff, C., Kinicki, A. J., & Tamkins, M. M. (2003). Organizational culture and climate. In Handbook of Psychology: Industrial and Organizational Psychology, ed. W. C. Borman, D. R. Ilgen, & R. J. Klimoski, vol. 12, 565- 593. Pearson, L. C., & Moomaw, W. (2005). The relationship between teacher autonomy and stress, work satisfaction, empowerment, and professionalism. Educational Research Quarterly, 29, 38-54. Perdue, S. V., Reardon, R. C., & Peterson, G. W. (2007) Person-environment congruence, self-efficacy, and environmental identity in relation to job satisfaction: A career decision theory perspective. Journal of Employment Counseling, 44, 29-38. Perie, M., & Baker, D. P. (1997). Job satisfaction among America’s teachers: Effects of workplace conditions, background characteristics, and teacher compensation (NCES 97-471). Washington, DC: U.S. Government Printing Office. Perrachione, B. A., Rosser, V. J., & Petersen, G. J. (2008). Why do they stay? Elementary teachers’ perceptions of job satisfaction and retention. The Professional Educator, 32, 1- 18. Perry, B., & Hayes, K. (2011). The effect of a new teacher induction program on new teacher program on new teachers’ reported teacher goals for excellence, mobility, and retention rates. International Journal of Educational Leadership Preparation, 6, 1-12. 80 Petersen, W. O. 1997, February. Principals’ values: Coming to shared purposes through a values-laden sense of identity? Paper presented at the annual meeting of the Eastern Education Research Association, Hilton Head, SC. Rafferty, A. E. & Jimmieson, N. L. (2010). Team change climate: A group-level analysis of the relationships among change information and change participation, role stressors, and well-being. European Journal of Work and Organizational Psychology, 19, 551-586. Raudenbush, S. W., & Byrk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods. Thousand Oaks, CA: Sage. Ravitch, D. (2010). The death and life of the great American school system: How testing and choice are undermining education. New York, NY: Basic Books. Reyes, P., & Imber, M. (1992). Teachers’ perceptions of the fairness of their workload and their commitment, job satisfaction, and morale: Implications for teacher evaluation. Journal of Personnel Evaluation in Education, 5, 291-301. Rice, E. M. & Schneider, G. T. (1994). A decade of teacher empowerment: An empirical analysis of teacher involvement in decision making, 1980-1991. Journal of Educational Administration, 32, 43-58. Richardson, B. K., Alexander, A., Castleberry, T. (2008). Examining teacher turnover in low- performing, multi-cultural schools: Relationships among emotional labor, communication symmetry, and intent to leave. Communication Research Reports, 25, 10-22. 81 Sanders, K., Dorenbosch, L., & de Reuver, R. (2008). The impact of individual and shared employee perceptions of HRM on affective commitment – Considering climate strength. Personnel Review, 37, 412-425. Schein, E. H. (2010). Organizational culture and leadership (4th Ed.). San Francisco, CA: Jossey-Bass. Schneider, B., Ehrhart, M. G., & Macey, W. H. (2013). Organizational climate and culture. Annual Review of Psychology, 64, 361-388. Schneider, B., Salvaggio, A. N., & Subirats, M. (2002). Climate strength: A new direction for climate research. Journal of Applied Psychology, 87, 220–229. Schyns, B., & Van Veldhoven, M. (2010). Group leadership climate and individual organizational commitment. Journal of Personnel Psychology, 9, 57-68. Scott, C., Stone, B., & Dinham, S. (2001). International patterns of teaching discontent. Education Policy Analysis Archives, 9. Retrieved November 15, 2012 from http://epaa.asu.edu/epaa/v9n28.html. Sergiovanni, T. J. 1992. Moral leadership: Getting to the heart of school improvement. San Francisco: Jossey-Bass. Shen, J. (1997). Teacher retention and attrition in public school: Evidence from SASS91. Journal of Educational Research, 91, 81-88. Shen, J., Leslie, J. M., Spybrook, J. K., & Ma, X. (2012). Are principal background and school processes related to teacher job satisfaction? A multilevel study using Schools and Staffing Survey 2003-04. American Educational Research Journal, 49, 200-230. 82 Shields, P. M., Humphrey, D. C., Wechsler, M. E., Riel, L. M., Tiffany-Morales, J., Woodworth, K., Young, V. M., & Price, T. (2001). The status of the teaching profession, 2001. Santa Cruz, CA: The Center for the Future of Teaching and Learning. Skaalvik, E. M. & Skaalvik, S. (July 2011). Teachers’ feeling of belonging, exhaustion, and job satisfaction: The role of school goal structure and value consonance. Anxiety, Stress, & Coping, 24, 369-385. Skaalvik, E. M. & Skaalvik, S. (August 2011). Teacher job satisfaction and motivation to leave the teaching profession: Relations with school context, feeling of belonging, and emotional exhaustion. Teaching and Teacher Education, 27, 1029-1038. Smart, J. C. & Thompson, M. D. (2001). The Environmental Identity Scale and differentiation among environmental models in Holland's theory. Journal of Vocational Behavior, 58, 436-452 Sparks, D. & Malkus, N. (2015). Public school teacher autonomy in the classroom across school years 2003–04, 2007–08, and 2011–12 (NCES 2015-089, U.S. Department of Education, National Center for Education Statistics).Washington, DC: U.S. Government Printing Office. Staw, B. M. (1975). Attribution of the "causes" of performance: A new alternative interpretation of cross-sectional research on organizations. Organizational Behavior and Human Performance, 13, 414-432. 83 Stockard, J., & Lehman, M. (2004). Influences on the satisfaction and retention of 1st- year teachers: The importance of effective school management. Educational Administration Quarterly, 40, 742-771. Taylor, D. L., & Tashakkori, A. (1995). Decision participation and school climate as Predictors of job satisfaction and teachers’ sense of efficacy. The Journal of Experimental Education, 63, 217-230. Theobald, N. (1990). An examination of the influence of personal, professional and school district characteristics on public school teacher retention. Economics of Education Review, 9, 241-250. Tickle, B. R., Chang, M., & Kim, S. (2011). Administrative support and its mediating effect on US public school teachers. Teaching and Teacher Education, 27, 343-349. Vail, K. (2005). Create great school climate. Education Digest, 71, 4-11. Valli, L. & Buese, D. (2007). The changing roles of teachers in an era of high-stakes accountability. American Educational Research Journal, 44, 519-558. Van Breukelen, W., Van Der Vlist, R., & Steensma, H. (2004). Voluntary employee turnover: Combing variables from the ‘traditional’ turnover literature with the theory of planned behavior. Journal of Organizational Behavior, 25, 893-914. Van Maele, D. & Van Houtte, M. (2012). The role of teacher and faculty trust in forming teachers’ job satisfaction: Do years of experience make a difference? Teaching and Teacher Education, 28, 879-889. 84 Weiss, E. (1999). Perceived workplace conditions and first-year teachers’ morale, career choice commitment, and planned retention: A secondary analysis. Teaching and Teacher Education, 15, 861-879. Whitener, S. D., Gruber, K., Lynch, H., Tingos, K., & Fondelier, S. (1997). Characteristics of stayers, movers, and leavers: Results from the Teacher Follow-Up Survey: 1994-95 (NCES 97450, U.S. Department of Education, National Center for Education Statistics). Washington, DC: U.S. Government Printing Office. Woods, R. C. & Montagno, R.V. (1997) Determining the negative effect of teacher attendance on student achievement. Education, 118, 307-316. Zohar, D. & Hoffman, D. H. (2012). Organizational culture and climate. In Kozlowski, S. W. J. (Ed.), Oxford Handbook of Organizational Psychology. New York: Oxford University Press. Zohar, D. & Luria, G. (2005). A multi-level model of safety climate: Cross-level relationships between organization and group-level climates. Journal of Applied Psychology, 90, 616-628. Zullig, K. J., Koopman, T. M., Patton, J. M., & Ubbes, V. A. (2010). School climate: Historical review, instrument development, and school assessment. Journal of Psychoeducational Assessment, 28, 139-152.