Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13
Topics
Discussion
Announcements
 

Examining Elementary Teachers’ Risk for Occupational Stress: Associations With Teacher, School, and State Policy Variables


by Richard Lambert, Christopher J. McCarthy, Paul G. Fitchett & Maytal Eyal - 2018

Background/Context: It is widely understood that teachers are plagued by a myriad of challenges that ultimately affect their stress levels, job satisfaction, and effectiveness at work. Teacher stress can lead to burnout, lowered occupational commitment, and an eventual decision to leave the field. An important question for the field is how best to understand which teachers are most vulnerable to stress. This study used Lazarus and Folkman’s transactional theory, which is the dominant model within the stress literature, to examine teachers’ stress vulnerability.

Objective: This study examined how elementary teacher appraisals of their classroom environment contribute to their risk for stress in the context of individual, classroom, and school characteristics, as well as state-level policy factors. Further, this study looked at how these factors are associated with teachers’ occupational stress, burnout, and commitment to teaching.

Participants: Participants were 11,850 full-time public school elementary teachers (Grades 1–5) who responded to the National Center for Education Statistics 2007/2008 Schools and Staffing Survey (SASS).


Research Design: Secondary data from the SASS were employed. The Rasch rating scale model was used to form scores on the Appraisal Index (teachers’ ratio of experiencing resources versus demands), as well as the Classroom Control and Burnout scales.


Data Collection and Analysis: Multilevel modeling was used. Each model had two levels with teachers nested within the state where they work. Two types of models were estimated using hierarchical linear modeling (HLM) software and restricted maximum likelihood procedures. The SASS teacher final sampling weight was then normalized and applied to all analyses within the HLM software. All results were reported with robust standard errors.


Findings: Teachers classified as at risk for stress based on SASS items about classroom demand and resources were found to be more likely to report lower job satisfaction and burnout symptoms, as well as reduced occupational commitment. Professional characteristics, school context, and the policy climate in which teachers work were also associated with teachers being at risk for stress.

Conclusions/Recommendations: Given these connections between occupational stressors, teacher appraisals of the classroom environment, and occupational outcomes, these results suggest that education stakeholders should be mindful of the climate and context in which public policies are enacted.



Teachers are the primary actors responsible for formal student learning, and it is widely understood that their jobs are challenging (Yang, Ge, Hu, Chi, & Wang, 2009). Yet, while some teachers thrive in the classroom, many experience sustained work stress that can reduce job satisfaction (Gilbert, Adesope, & Schroeder, 2014) and impair teaching effectiveness, particularly with respect to classroom management (Aloe, Shisler, Norris, Nickerson, & Rinker, 2014). In the long run, teacher stress can lead to burnout (Friedman, 2006), lowered occupational commitment (Klassen & Chiu, 2011), and an eventual decision to leave the field (McCarthy, Lambert, & Reiser, 2014).


Although teacher stress is widely recognized, research often fails to include a central tenet of most theoretical models: namely, that stress results when an individual appraises the magnitude of demands he or she encounters as exceeding available resources (Meurs & Perrewé, 2011). According to Folkman, Lazarus, Dunkel-Schetter, DeLongis, and Gruen (1986), appraisal is a cognitive process that involves evaluating demands that have significance for our well-being against available resources for meeting those demands. Appraisal is a central theoretical construct in Lazarus and Folkman’s (1984) transactional theory and is conceptualized as involving two complementary elements: evaluating demands accurately in terms of what they mean for our well-being, while also putting the best possible light on events to maintain hope and cope effectively. Lazarus (2001) summarized the appraisal process in this way: “appraisal is a compromise between life as it is and what one wishes it to be, and efficacious coping depends on both” (p. 41). Applied to the context of teaching, this theory suggests that educators are most vulnerable to stress when appraising classroom demands (for example, students engaging in disruptive behavior) as exceeding resources available to them (in this case, their capacity to use effective classroom management skills).


The transactional framework is increasingly being applied to understanding teachers’ job stress (Spilt, Koomen, & Thijs, 2011), particularly with respect to how educators’ appraisals of the classroom are associated with their experience of stress (Bakker, Hakanen, Demerouti, & Xanthopoulou, 2007; van den Tooren, de Jonge, Vlerick, Daniels, & Van de Ven, 2011). Consistent with Lazarus’s (2001) definition noted earlier, a teacher’s appraisals are presumed to be based on objective features of the classroom leavened with the teacher’s unique interpretation of his or her everyday classroom reality. Accordingly, the term risk for stress is used in this study to refer to teachers who appraise their classroom demands as insufficient vis-à-vis their classroom resources, putting them at greater risk for eventual stress symptoms and lowered occupational health (McCarthy, Lambert, Lineback, Fitchett, & Baddouh, 2016).


Currently, the policy implications stemming from the transactional conceptualization of teacher stress are limited by several factors. First, while transactional theory has been the dominant model of stress for decades, the construct of appraisal historically has been challenging to define and operationalize (Matheny, Aycock, Curlette, & Junker, 1993). Specifically, measuring an individual teacher’s perceptions of his or her demands and resources poses significant measurement challenges (Hobfoll, 1989) because appraisals of demands and resources need to be compared in order to understand a given person’s risk for stress. Some researchers have used proxy variables for stress, such as number of hours worked (Grissom, Nicholson-Crotty, & Harrington, 2014). Such parsimonious measures fail to accurately conceptualize the appraisal process. Lacking an accurate operationalization of the transactional theory appraisal process limits clear takeaways for policy makers and administrators.


A second obstacle is the current reliance in the policy literature on educational production models, which examine various inputs (salary, working conditions) that are presumed to result in outputs, such as job satisfaction and occupational commitment (Hanushek, 2008; Monk, 1989). This research relies on, but does not account for, the self-reported nature of these data, which has been criticized for lacking external reliability and being susceptible to social desirability bias (Boyd et al., 2011). Transactional theory, on the other hand, posits that appraisal include both objective elements and individual interpretations. Understanding self-reports about inputs, such as excessive administrative tasks, may offer more avenues of intervention for local administrators and policy makers, who may have limited influence over system factors but could benefit from identifying which teachers are experiencing the highest demand levels when considering resource allocation.


Accounting for differences in teacher appraisals of their classroom raises a third issue with important policy implications: namely, to what extent is a teacher’s risk for stress accounted for by assessing his or her appraisals of the classroom, and to what extent is it accounted for by other factors, such as teacher, classroom, and school characteristics, and broader contextual factors, such a state policy (Lambert, McCarthy, Fitchett, Lineback, & Reiser, 2015)? The current study was designed to address these issues. The study seeks to examine how teacher appraisals of the classroom environment contribute to elementary teachers’ risk for stress, specifically within the context of individual, classroom, and school characteristics, as well as state-level policy factors. These data, with the exception of state-level policy factors, were drawn from the Schools and Staffing Survey (SASS). This survey, the largest and most comprehensive data source available on teachers and schools (Ingersoll & Smith, 2003), is administered by the National Center for Education Statistics (NCES). Previous research using SASS data has shown that items from the SASS survey that address classroom demands and resources can be used to reliably classify elementary teachers according to their risk for stress (Lambert et al., 2015). The SASS also includes items that assess working conditions and vocational concerns, which, along with Education Week’s annual Quality Counts survey, have been used to examine associations between teachers’ stress and state-level factors. To situate this study in the current literature, we will provide a brief review of research on teacher working conditions and how state policy climate affects working conditions. Further, we will review the relevant literature assessing teacher stress using transactional theory.


RESEARCH EXAMINING TEACHER WORKING CONDITIONS


Though teacher pay is often the most publicized policy tool for improving teacher retention and overall occupational well-being, numerous studies point toward teachers’ perceptions of workplace climate as a more robust predictor of various vocational factors, such as job satisfaction, risk for occupational burnout, and professional commitment (Ingersoll, 2001; Ladd, 2011; Liu & Ramsey, 2008; Loeb, Darling-Hammond, & Luczak, 2005; McCarthy, Lambert, O’Donnell, & Melendres, 2009). Teachers who view their school favorably are more likely to express a commitment toward teaching and overall satisfaction in their occupation. In general, the research examining teachers’ attitudes related to their respective school environments can be delineated into three overlapping categories: school leadership, professional authority, and instructional supports.


The most recognizable leader in schools—principals—set building-level policies (e.g., student discipline protocols and the allocation of instructional supports) that are foundational for the perceived workplace culture in schools. In American school systems, principals hold enormous power over teachers’ course and grade-level assignments, discipline policies, and other general personnel decisions (Johnson, 2006). Teachers who perceive themselves as equitable stakeholders in school-level decisions report a more favorable view of their principals and, by extension, their workplace climate (Johnson et al., 2014; Ladd, 2011). School leaders who lean toward more democratic approaches of administration include teachers in building-level policy making, curtail superfluous bureaucracy, and effectively communicate their vision. Conversely, teachers under the thumb of less transparent leadership are more likely to leave teaching, to seek out employment in other schools, and to report less satisfaction with their job (Johnson, 2006; Liu & Ramsey, 2008).  


Concurrently, principals, as the “brokers” of schools’ policy (Johnson, 2006, p. 15), make decisions on how to delegate professional autonomy to teachers. Wanting to professionalize their occupational lives (Ingersoll & Merrill, 2011), PK–12 educators seek control over educational policies that affect their individual classroom communities, while also attempting to exert greater school-level influence (Good & Brophy, 2000; Goodlad, 2004). Teachers who report holding greater influence over their classroom decision making are less likely to leave teaching and less frequently experience occupational burnout (Farber, 1991; Pearson & Moomaw, 2005).


Instructional supports are a third crucial component of teachers’ workplace perceptions. In-school instructional supports such as teaching materials, induction, professional development, and teacher mentoring are associated with lower teacher attrition (Ingersoll, 2001; Johnson, 2004, 2006; Ladd, 2011; Smith & Ingersoll, 2004). These supports help offset the bureaucratic and instructional demands of school, thereby improving job satisfaction and occupational commitment.


However, prepackaged school leadership approaches and one-dimensional policies toward the distribution of instructional supports are insufficient. Research suggests that principals foster a leadership style tailored to the unique school ecologies, addressing the varying instructional needs of staff to foster positive teacher morale and maintain a constructive workplace climate (Johnson et al., 2014; Loeb et al., 2005). Thus, how teachers perceive resources (and the demands) of schools varies among staff, making it difficult for school leadership to craft blanket policies that address the occupational concerns of the teaching workforce.


WORKING CONDITIONS WITHIN A POLICY CONTEXT


State policy contexts both directly and indirectly influence the work of teachers and ultimately affect how teachers perceive their work environments. Weick (1976) argued that schools are “loosely coupled systems” (p. 3), which has traditionally made it difficult to coordinate learning outcomes and manage staff. In an attempt to regulate school environments, contemporary educational policy has increasingly turned toward controlling the work of teachers (Ingersoll, 2009). The result is primarily a top-down organizational structure in which state and federal mandates dictate resources and money allocation, drive curriculum, set personnel standards, and establish mandated testing protocols in various subject areas. In turn, principals and others in school leadership roles exert the greatest direct impact on teachers’ working conditions by enacting policy within schools. These policies are often subsumed under the umbrella of accountability, holding various public education actors responsible for how successful these policies are enacted. The purpose is to purportedly maintain (or improve) the overall educational quality in the nation’s public schools (Lauen & Tyson, 2009; Plank & Keesler, 2009).


State accountability policies have received substantial scrutiny over the years for micro-managing classroom time and placing increased pressure on teachers’ performance (Ingersoll, 2007). D. K. Cohen and Moffitt (2009) suggested that, for policies to be successful, there must be alignment between what they referred to as “aims” and “capabilities.” In other words, the purposes (i.e., aims) of a given policy initiative must be supported by available resources (i.e., capabilities). Policies with goals misaligned to available resources will not succeed, while potentially placing an undue burden on school personnel (i.e., teachers).


Research findings vary as to whether standardized testing and accountability systems positively or negatively affect teaching conditions. One line of research and theory suggests that increased accountability structures undermine teachers’ instructional freedom, thereby contributing to higher levels of burnout, stress, attrition, and job dissatisfaction (Apple, 2004; Cochran-Smith & Lytle, 2006; Crocco & Costigan, 2007). Some researchers have posited that pressures to cover standards-based curricula often countermand teacher autonomy over classroom instruction (Darling-Hammond & Rustique-Forrester, 2005). Though accountability is intended to spur improvement among schools, it also undermines teacher quality by inherently disincentivizing qualified teachers from working in schools that have been placed under sanction or stigmatized as low performing (Ingersoll, 2001). Thus, it may reinforce a system in which the least experienced, less qualified teachers are allocated to high-needs classrooms (Darling-Hammond & Rustique-Forrester, 2005).


However, other studies comparing teacher survey data before and after the inception of No Child Left Behind infer that accountability policies have a minimal effect on teachers’ perceptions of their workplace environment (Grissom et al., 2014; Lambert et al., 2015). Other research suggests that quality of teaching, including an emphasis on more rigorous instructional practices, can be improved through the use of structured assessments. For example, in states like Connecticut, Kentucky, and Vermont, studies have reported improvement in teaching tied to a combination of investments in professional development for teachers, equalization of school resources, and standardized assessments for students (Darling-Hammond & Rustique-Forrester, 2005). Another study points to stronger standards, testing, and accountability systems as a factor reducing the likelihood of teachers leaving their schools (Smith, 2007).


The majority of the aforementioned research has made use of education production function models for analysis—theorizing that various teacher-, school-, and policy-level inputs predict observable outputs (Monk, 1989). Because these analyses operationalize teacher self-reported data as actual working conditions (or inputs) rather than perceptions of occupational climate, findings fail to account for reporting biases (Boyd et al., 2011). Privileging teachers’ perceptions of their workplace environment using transactional theory may allow for a fuller understanding of teachers’ risk for stress and potentially offer a reliable and parsimonious approach toward identifying teachers who are at risk for burnout, attrition, and other vocational concerns that can result from stress.


MEASURING TEACHERS’ RISK FOR STRESS AND OCCUPATIONAL HEALTH


The transactional perspective, which posits appraisals of demands and resources as central to understanding risk for stress, may be particularly appropriate for understanding elementary teachers’ stress, given that these teachers work in intact classrooms and encounter similar classroom demands and resources on a daily basis (McCarthy et al., 2009). Lambert, McCarthy, O’Donnell, and Wang (2009) developed the Classroom Appraisal of Resources and Demands (CARD) to measure elementary teachers’ perceptions of demands in the following categories: students with problematic behaviors; other student-related demands, such as poor attendance; administrative demands; and lack of instructional resources (sample item: availability of instructional supplies). The Resources section of the CARD asks elementary teachers about the helpfulness of the following types of classroom resources: availability of school support personnel (administrators and classroom aides), other adults (community volunteers), instructional support materials, and specialized instructional resources, such as for children performing below grade level. If a given resource is not available, the respondent is given the option of answering “not applicable.” Evidence for the reliability of the CARD was reported by Lambert et al. (2009).


Teachers are classified using the CARD difference score between their ratings of demands and resources in the classroom; this is labeled an Appraisal Index because it refers to the overall appraisal of teachers’ resources vis-à-vis their demands. The Appraisal Index is a difference score between resources and demands that places teachers into three groups according to their risk for stress (see McCarthy et al., 2016, for a review of the classification procedures): (1) teachers perceiving classroom resources as greater than demands (labeled the Resourced group); (2) teachers perceiving classroom demands as equal to resources (labeled the Balanced group); and (3) teachers perceiving classroom demands as greater than resources (labeled the Demands group). According to Lazarus and Folkman’s (1984) transactional model of stress, this last group, Demands, is theorized to be most vulnerable to stress (McCarthy et al., 2016).


The theoretical proposition that Demands group teachers are most vulnerable to stress was supported in a review of 18 studies using the CARD, which demonstrated that teachers’ scores on the CARD were associated with numerous indicators of occupational health. Specifically, Demands group teachers were found to be (a) lower in job satisfaction, which refers to a person’s attitude toward her or his work (Brayfield & Rothe, 1951); (b) higher in burnout, which includes symptoms such as emotional exhaustion, distancing oneself from others at work, and a reduced sense of accomplishment (Schaufeli & Enzmann, 1998); and (c) reduced occupational commitment, defined by Jepson and Forrest (2006) as “dedication and loyalty to the teaching profession” (p. 188). Each of these constructs—satisfaction, burnout, and occupational commitment—can be viewed as part of a developmental sequence of how stress may affect teachers. Specifically, stress can impact a teacher’s job satisfaction, leading to burnout and, ultimately, a decision to leave teaching (McCarthy, Lambert, Crowe, & McCarthy, 2010).


Although the CARD has been used with local samples of elementary teachers, evidence has shown that the classification approach of assessing teacher demands and resources in the classroom can be replicated with the SASS, which surveys teachers on their workplace climate (Lambert et al., 2015). Although the specific classroom demands and resources items from the CARD are not included in the SASS survey, SASS items asking about classroom demands and resources are conceptually similar, and the classification strategy of creating difference scores to arrive at the Appraisal Index was supported by large group differences between the resulting appraisal groups in the expected directions. These group differences replicated findings from earlier CARD research (McCarthy et al., 2016). For example, assignment of teachers to categories according to their risk for stress was supported through predicted associations with other variables available in the SASS data set, specifically those having to do with job satisfaction, occupational commitment, and burnout symptoms. An important question is, therefore, How much of a teacher’s risk for stress is associated with classroom appraisals, and to what extent do teachers’ professional characteristics, school context, and state policy contribute to their risk for stress, as well their risk for burnout and reduced occupational health?


GOALS OF THE CURRENT STUDY


The current study extended existing research assessing elementary teachers’ risk for stress in several ways: by attempting to replicate findings from school-system-specific studies using national data, by examining previous findings in a multivariate context, and by examining sources of teacher stress at multiple levels, including teacher, classroom, and school characteristics, as well as the state policy climate context. Multiple outcome measures were chosen to investigate the relationships between potential stressors and teacher subjective experiences across the proposed sequence, from occupational stress, to burnout, to leaving the profession. The following research questions guided the analyses: (1) To what extent are teacher characteristics, classroom/school context, and state policy climate associated with teachers’ risk for occupational stress? (2) To what extent are teacher characteristics, risk for stress, school/classroom context, and state policy climate associated with teachers’ occupational burnout? (3) To what extent are teacher characteristics, risk for stress, classroom/school context, and state policy climate associated with teachers’ occupational commitment?


METHODS


PARTICIPANTS


The participants in this study (N = 11,850) were all full-time public school elementary teachers who responded to the 2007–2008 SASS. The first version of the CARD was developed to assess teacher appraisals of classroom demands and resources in elementary-grade classrooms. Similarly, elementary teachers were chosen as the focus of this study because they tend to have a single group of children with whom they spend almost all their instructional time. Middle and secondary teachers do not generally serve a single intact classroom of children; therefore, appraisals of both classroom demands and resources involve a potentially different psychological process (McCarthy et al., 2009). Teacher demographic and professional background information, along with school characteristics, is summarized in Table 1. Given the complex, multistage sampling procedures, the purposeful oversampling, and the varying nonresponse rates across subgroups of teachers, weighting is required to ensure that the results are nationally representative. The teacher final sampling weights provided by NCES were normalized and applied to the results in Table 1. A large majority of the teachers had worked in education for two or more years (94.3%), and the remaining 5.7% were new to teaching as defined by having completed less than two years of full-time teaching. The average years of experience in education was 13.48 years (SD = 10.11).


Table 1. Demographic Characteristics of the Sample

 

 

Weighted

Demographic Variable

Category

%

 

 

 

 

 

 

Urbanicity of school

Central city

27.0

 

Urban fringe

35.5

 

Small town or rural

37.6

 

 

 

Work in a Title I school

Yes

54.5

 

 

 

Years of teaching experience

Less than two

5.7

 

Two or more

94.3

 

 

 

New to current school

Yes

14.4

 

 

 

Alternatively certified

Yes

11.2

 

 

 

No certification

Yes

6.6

 

 

 

Subject area specialist

Yes

35.3

 

 

 

Special education teacher

Yes

11.1

 

 

 

Gender

Male

15.4

 

Female

84.6

 

 

 

Race

Native American

1.1

 

Asian or Pacific Islander

1.9

 

African American

8.0

 

European American

89.0

 

 

 

Hispanic

Yes

7.7

 

 

 

 

 

 


A substantial minority of teachers were in their first two years at a new school (14.4%). This sample includes special education teachers (11.1%) and subject area specialists (35.3%), defined as enrichment teachers and teachers who focus on a particular subject across multiple classrooms. Alternatively certified teachers constituted 11.2% of the sample, and teachers with no certification made up 6.6% of the sample. Race and ethnicity percentages were as follows: European American (89.0%), African American (8.9%), Asian or Pacific Islander (1.9%), Native American (1.1%), and Hispanic (7.7%).


These teachers worked in 3,810 schools located in all 50 states and the District of Columbia. The majority of the teachers worked in Title I schools (54.5%). With respect to urbanity, the schools were located in central city urban locations (27.0%), urban fringe or suburban locations (35.5%), and small town or rural locations (37.6%). The majority (84.5%) of teachers in this sample taught at least one child with an individual education program (IEP), and the remaining 15.5% of teachers taught no children with an IEP. The mean percentage of a respondent’s teaching load that included an IEP was 14.29% (SD = 20.19). In addition, the majority (54.9%) of teachers in this sample taught at least one child with limited English proficiency (LEP), and the remaining 45.1% of teachers taught no children with LEP. The mean percentage of a respondent’s teaching load that included LEF was 10.54% (SD = 20.81).


MEASURES


2007–2008 Schools and Staffing Survey


SASS used a complex multistage sampling procedure in which schools were sampled, and then samples of teachers were selected from within each sampled school. The primary sampling unit for SASS is the school. From the Common Core of Data, a comprehensive listing of all schools in the United States, NCES selects a random sample of schools. Instructions are then sent to each school regarding how to randomly sample teachers within the school. Purposeful oversampling of specific subgroups of schools is also included in the sampling plan. The sample of schools in SASS was designed to collect data regarding teacher perceptions of school climate, overall employment and working conditions, and descriptive data about school contexts throughout the nation (Tourkin et al., 2010). It is designed to create a nationally representative sample of teachers and to oversample low-incidence subgroups. From the SASS data, we operationalized several items that we referred to as teacher characteristics: total years of experience, race indicators, gender, new to school, new to teaching, subject area specialist, special education indicator, alternative license designation, certification designation, and classroom control.1 In addition, classroom and building contextual items included the classroom percentages of students designated with LEP, students designated with IEPs, and urbanicity and Title I status of the school. Finally, the dependent variables from this study were created from SASS data. Examining teachers’ risk for stress was examined in two ways. First, an Appraisal Index was created to examine teachers’ risk for being stressed (as described in the procedures in the next section). Second, a classification system was devised to classify teachers as being Resourced (least at risk for stress), Balanced, or Demanded (most at risk for stress). Burnout was operationalized from seven Likert-type items located in the teacher workplace attitudes inventory.2 Professional commitment was scaled as two dichotomous variables: would become a teacher again (T0320), and intention to remain in teaching (T0321). Further description of how key variables were scaled is described in the next section.


Quality Counts


Quality Counts is a yearly audit of K–12 public education quality measures conducted by the professional periodical Education Week. Scores are derived from Education Week’s external evaluation of various policy indicators within a given state. Gathering data from each of the 50 states and the District of Columbia, Quality Counts reports policy climate information across six areas: (1) Transitions and Alignment, (2) Standards, Accountability, and Assessment, (3) The Teaching Profession, (4) Chance for Success, (5) K–12 Achievement, and (6) School Finance. Each of these areas was associated with a scale (60–100) that translated into a state grade (A–F).


Transitions and Alignment refers to student transitions from one educational setting to another. Items used to create the scale include alignment between kindergarten standards and grade school standards and expectations; college readiness among students; and economic/workforce readiness. Standards, Accountability, and Assessment scale scores examine high standards for learning, accurate and informative assessments, and rigorous overall accountability structures within a given state. The Teaching Profession scale measures teaching quality by assessing factors such as initial licensure requirements for prospective teachers, methods used to evaluate teacher performance, and data systems used to monitor quality. The Teaching Profession also evaluates incentives to improve teaching quality, the allocation of teachers among schools, and physical and emotional support offered to teachers across states. Chance for Success consists of a number of socioeconomic, cultural capital, and school performance attainment items, including family income, parent education, parental employment, linguistic integration, preschool enrollment, kindergarten enrollment, fourth-grade reading as measured by the National Assessment for Educational Progress (NAEP), eighth-grade mathematics (NAEP scores), high school graduation, young adult education, adult educational attainment, annual income, and steady postschool employment. Critics claim that many of these items are not directly attributable to public policy and are not directly aligned with success in schools (Raymond, 2010). We determined that this scale offered potentially valuable insight into how the confluence of educational conditions both inside and out of school is associated with teachers’ risk for occupational stress and decided to include it in the study. The K–12 achievement score uses mathematics and reading data from the NAEP to measure proficiency percentages, growth, and socioeconomic performance gaps. The School Finance grade is calculated using a variety of indicators, including per pupil expenditure and the ratio of district funding to local property wealth. Each of these six areas was included in the statistical models to examine the association between teachers’ risk for stress and the education policy climate.


PROCEDURES


The same CARD scoring strategy used in previous SASS research was applied to the teacher responses in the current sample. This procedure involved creating scale scores for Classroom Demands and Classroom Resources. Both the CARD and SASS items that comprise these scales were detailed by Lambert et al. (2015). A specific case of the one parameter item response theory (IRT) model, the Rasch rating scale model, was used through the WINSTEPS software package to combine the SASS responses for scale into scale scores and estimate ability parameters for each teacher. The resulting Demands and Resources scale scores were moderately correlated (r = -.476).


The Classroom Demands (α = .863) and Resources (α = .827) scales yielded scores with adequately internal consistency reliability. To match the previous protocol for classifying teachers using the CARD, an Appraisal Index score was created based on the difference between the Demands and Resources scale scores (reliability = .895). This reliability coefficient is based on the formula for the reliability of a difference score. Next, a 95% confidence interval was formed around no difference between the Demands and Resources scale scores (McCarthy et al., 2016). Teachers who provided difference scores greater than the upper limit of this interval were classified in the Demands group, those who provided difference scores below the lower limit of the confidence interval were classified in the Resourced group, and those with difference scores within the interval were classified in the Balanced group. A similar process from previous research (Lambert et al., 2015) using the Rasch rating scale model was used to form Classroom Control and Burnout scale scores. These scales yielded scores with adequate internal consistency reliability within the given sample (Classroom Control α = .726, Burnout α = .810).


DATA ANALYSIS STRATEGY


Multilevel modeling was used to address the research questions. Each model had two levels, with teachers nested within the state where they work. Level 1 models examined the association between teacher, classroom, and school characteristics and within state variance in each teacher-level outcome. All Level 1 predictor variables were group mean centered. Level 2 models examined the association between state policy climate variables and between state variance in state-level means for each of the outcomes. A random intercepts model was included at Level 2 to examine associations with state means, and all Level 2 predictors were grand mean centered. Two types of models were estimated using the HLM software: models for continuous outcomes and models for dichotomous outcomes. The models for continuous outcomes used nested regression models, and the models for dichotomous outcomes used linked logistic regression functions with Level 1 variance = 1/[ϕij(1-ϕij)]. Both models were estimated using restricted maximum likelihood procedures. Overall, the number of teachers per school was too small to justify using a three-level model (M = 3.110, SD = 1.376). The majority of these teachers (64.2%) worked in schools that contributed only one, two, or three teachers to the sample. Therefore, school characteristics were modeled as Level 1 fixed effects.


Given the complex, multistage sampling procedures, the purposeful oversampling, and the varying nonresponse rates across subgroups of teachers, the SASS teacher final sampling weight was normalized and applied to all analyses within the HLM software. In addition, all results are reported with robust standard errors as estimated by the HLM software. Given that the SASS sampling plan does not exactly match the clustering structure (teachers within states) that was used with the current analyses, the robust standard errors are an approximation of the correct values. Some of the models also include membership in the Demands and Resourced groups as predictors. These variables are coded 0 / 1, and 1 indicates membership in the respective groups. The baseline condition is the Balanced group (see the appendix).


RESULTS

 

Research questions were examined according to a theorized progression that might lead to a decision to leave the teaching profession (McCarthy et al., 2010). Specifically, the first research question examined predictors of risk for stress (Appraisal Index and appraisal group classification), which is hypothesized as an early warning indicator that teachers are experiencing excessive demand levels. The second research question then examined predictors of burnout, which is hypothesized to result from extended exposure to stressful working conditions. Finally, Research Question 3 examined teachers’ intentions to remain in the profession and whether they would become a teacher again, which could represent outcomes associated with prolonged stress and burnout.


RESEARCH QUESTION 1: PREDICTING TEACHERS’ RISK FOR STRESS


Research Question 1 asked if teacher professional characteristics, school context, and state policy climate were associated with teachers’ risk for occupational stress. As described earlier, risk for stress was operationalized both as continuous variables using scores on the Appraisal Index and as a categorical variable using cutoffs for the Appraisal Index based on previous research (Lambert et al., 2015). Measuring risk for stress using the Appraisal Index provides a continuum for stress vulnerability, whereas using categorizations provides a theoretically meaningful and parsimonious method for understanding which teachers are most vulnerable to stress.


Table 2 contains the results of the model using the Appraisal Index scores as a continuous independent variable. The overwhelming majority of the variance in Appraisal Index scores was found in the null model to be within states (97.6%), and between-state variance comprised the remaining 2.4%. Several classroom and school characteristics were statistically significantly associated with Appraisal Index scores. Classroom percentage of children with LEP was positively associated, indicating that higher classroom percentages tend to be associated with higher Appraisal Index scores, which indicate more risk for occupational stress. Teachers working in urban and rural schools, rather than suburban schools, also reported higher Appraisal Index scores, as did teachers working in Title I schools.


Table 2. Association Between Teacher- and State-Level Variables and Appraisal Index Scores

 

 

 

 

 

Model

Variable

Coefficient

SE

 

 

 

 

 

 

 

 

 

 

 

Level 1: Contextual variables

Classroom % with an IEP

0.002

0.001

 

 

Classroom % with LEP

0.003

0.001

***

 

Urban school setting

0.329

0.048

***

 

Rural school setting

0.094

0.036

**

 

Title I school setting

0.263

0.030

***

 

 

 

 

 

Level 1: Teacher characteristics

Total years of experience

-0.003

0.001

*

 

African American

0.074

0.074

 

 

Asian American

0.053

0.097

 

 

Pacific Islander

0.014

0.183

 

 

Native American

0.099

0.111

 

 

Hispanic

-0.023

0.029

 

 

Male

0.042

0.034

 

 

New to school

-0.305

0.036

***

 

New to teaching

0.009

0.048

 

 

Subject area specialist

0.402

0.034

***

 

Special education teacher

0.196

0.043

***

 

Alternatively certified

0.095

0.050

 

 

No certification

-0.042

0.063

 

 

Classroom Control

-0.354

0.017

***

 

 

 

 

 

Level 2: State policy climate

Intercept

-1.014

0.034

***

 

Chance for Success

0.003

0.006

 

 

K–12 Achievement

-0.004

0.006

 

 

The Teaching Profession

-0.003

0.004

 

 

School Finance

-0.003

0.003

 

 

Transitions and Alignment

-0.008

0.004

*

 

Accountability

0.011

0.004

*

 

 

 

 

 

***p < .001. **p < .01. *p < .05.

 

 

 


Several teacher characteristics were also statistically significantly associated with Appraisal Index scores. More years of experience tended to be associated with lower Appraisal Index scores, indicating less risk for stress. Being new to a school and higher Classroom Control scores were also associated with lowered risk for stress, whereas being a subject area specialist or special education teacher was associated with higher Appraisal Index scores. The pseudo R2 value for the Level 1 model was .144, indicating that Level 1 teacher and classroom/school context variables accounted for a moderate amount of within-state variance.3


Several state policy climate variables were statistically significantly associated with Appraisal Index state average scores. The Transitions and Alignment scores were negatively associated with state average Appraisal Index scores. States with higher Standards, Assessment, and Accountability scores tended to have higher average Appraisal Index scores. The pseudo R2 value for this model was .597, indicating that the state policy climate variables accounted for the majority of the between-state variance.


Research Question 1 was also addressed using Appraisal Index scores to place teachers into one of three groups: Demands, Balanced, and Resourced. Table 3 contains the results of the model using membership in the Demands groups as the dependent variable. Findings were consistent with those obtained when using the Appraisal Index as a continuous variable. This strategy identified 28.1% of the sample as at risk for stress, which is consistent with previous findings using the CARD and SASS data (Lambert et al., 2015; McCarthy et al., 2016).


Table 3. Association Between Teacher- and State-Level Variables and Being Classified in the Demands Group


 

 

 

 

 

 

95%

95%

 

 

 

 

 

Odds

Lower

Upper

Model

Variable

Coefficient

SE

 

Ratio

Limit

Limit

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Level 1: Contextual variables

Classroom % with an IEP

0.004

0.002

*

1.004

1.001

1.007

 

Classroom % with LEP

0.005

0.002

*

1.005

1.001

1.009

 

Urban school setting

0.602

0.097

***

1.826

1.509

2.210

 

Rural school setting

0.088

0.103

 

1.092

0.893

1.336

 

Title I school setting

0.480

0.071

***

1.616

1.407

1.856

 

 

 

 

 

 

 

 

Level 1: Teacher characteristics

Total years of experience

-0.011

0.003

***

0.989

0.983

0.995

 

African American

0.073

0.165

 

1.076

0.778

1.487

 

Asian American

0.122

0.226

 

1.129

0.725

1.760

 

Pacific Islander

0.106

0.191

 

1.112

0.764

1.618

 

Native American

0.198

0.231

 

1.219

0.775

1.919

 

Hispanic

-0.169

0.077

*

0.844

0.726

0.982

 

Male

0.059

0.130

 

1.061

0.823

1.367

 

New to school

-0.498

0.103

***

0.608

0.497

0.744

 

New to teaching

0.054

0.102

 

1.055

0.864

1.289

 

Subject area specialist

0.670

0.127

***

1.955

1.524

2.508

 

Special education teacher

0.322

0.074

***

1.380

1.194

1.596

 

Alternatively certified

0.211

0.078

**

1.234

1.059

1.439

 

No certification

-0.199

0.126

 

0.820

0.640

1.050

 

Classroom Control

-0.646

0.049

***

0.524

0.476

0.577

 

 

 

 

 

 

 

 

Level 2: State policy climate

Intercept

-1.764

0.067

***

0.171

0.150

0.196

 

Chance for Success

0.004

0.008

 

1.004

0.988

1.021

 

K-12 Achievement

-0.004

0.008

 

0.996

0.980

1.013

 

The Teaching Profession

-0.011

0.006

 

0.989

0.977

1.001

 

School Finance

-0.003

0.005

 

0.997

0.987

1.006

 

Transitions and Alignment

-0.011

0.006

 

0.989

0.977

1.002

 

Accountability

0.028

0.007

***

1.028

1.014

1.042

 

 

 

 

 

 

 

 

*** p < .001. **p < .01. *p < .05.

 

 

 

 

 

 



First, several classroom and school characteristics were significantly associated with membership in the Demands group. Both classroom percentages of children with IEPs and with LEP were associated with higher probabilities of membership in the Demands group. Teachers working in Title I schools also had a much higher probability of membership in the Demands group than all other teachers (odds ratio = 1.616), as did teachers working in urban schools, who had a much higher probability of membership in the Demands group than those in suburban schools (odds ratio = 1.826).


Several teacher characteristics were statistically significantly associated with membership in the Demands group. Being a more experienced teacher, being new to a school, reporting higher levels of Classroom Control, and identifying as Hispanic resulted in a lower probability of Demands group membership. Conversely, being alternatively certified, being a subject area specialist (odds ratio = 1.955), or being a special education teacher (odds ratio = 1.380) was associated with higher probabilities of membership in the Demands group. One state policy climate variable was statistically significantly associated with Appraisal Index state average scores. States with higher Standards, Assessment, and Accountability scores tended to have higher average probabilities of membership in the Demands group. The pseudo R2 value for this model was .743, indicating that the state policy climate variables accounted for the majority of the between-state variance.


RESEARCH QUESTION 2: EXAMINING PREDICTORS OF BURNOUT SYMPTOMS


The overwhelming majority of the variance in Burnout scores was found in the null model to be within states (98.1%), whereas between-state variance comprised the remaining 1.9%. The pseudo R2 value for the Level 1 model was .387. Membership in the Demands group was associated with Burnout scores that were, on average, .726 standard deviation units higher than the Balanced group, whereas membership in the Resourced group was associated, on average, with Burnout scores that were .792 standard deviation units lower than the Balanced group.


Several teacher characteristics were also significantly associated with Burnout scale scores (see Table 4). Identifying as an African American or Asian American teacher was associated with higher Burnout scores than identifying as a White teacher. Special education teachers also scored higher on the Burnout scale than regular classroom teachers. Teachers with higher Classroom Control scores tended to have lower Burnout scores.


Table 4. Association Between Teacher- and State-Level Variables and Burnout

 

 

 

 

 

Model

Variable

Coefficient

SE

 

 

 

 

 

 

 

 

 

 

 

Level 1: Contextual variables

Classroom % with an IEP

-0.001

0.001

 

 

Classroom % with LEP

-0.001

0.001

 

 

Urban school setting

0.050

0.053

 

 

Rural school setting

0.043

0.025

 

 

Title I school setting

-0.044

0.034

 

 

 

 

 

 

Level 1: Teacher characteristics

Total years of experience

0.003

0.002

 

 

African American

0.145

0.070

*

 

Asian American

0.221

0.104

*

 

Pacific Islander

-0.051

0.090

 

 

Native American

-0.055

0.098

 

 

Hispanic

0.108

0.082

 

 

Male

0.011

0.028

 

 

New to school

-0.029

0.054

 

 

New to teaching

-0.219

0.088

*

 

Subject area specialist

0.019

0.022

 

 

Special education teacher

0.115

0.055

*

 

Alternatively certified

-0.063

0.072

 

 

No certification

-0.085

0.048

 

 

Classroom Control

-0.144

0.012

***

 

Demands group

0.726

0.031

***

 

Resourced group

-0.792

0.037

***

 

 

 

 

 

Level 2: State policy climate

Intercept

-0.172

0.028

 

 

Chance for Success

-0.011

0.007

 

 

K–12 Achievement

0.008

0.005

 

 

The Teaching Profession

0.006

0.003

*

 

School Finance

-0.003

0.004

 

 

Transitions and Alignment

-0.001

0.003

 

 

Accountability

-0.005

0.004

 

 

 

 

 

 

***p < .001. **p < .01. *p < .05.

 

 

 


The pseudo R2 value for this model was .641, indicating that the state policy climate variables accounted for the majority of the between-state variance. Only one of the state policy climate variables, The Teaching Profession, was statistically significantly associated with burnout. States with higher scores tended to also have higher state average Burnout scores.


RESEARCH QUESTION 3: EXAMINING PREDICTORS OF PROFESSIONAL COMMITMENT


Research Question 3, examining if professional characteristics, risk for stress, school context, and state policy climate are associated with teachers’ occupational commitment, was addressed using both the “willingness to become a teacher again” and “intention to return again” items from the SASS. Table 5 shows the results of the model, using willingness to become a teacher again as the dependent variable, and Table 6 shows the results of the model using intention to return to teaching for the next academic year as the dependent variable.


Table 5. Association Between Teacher- and State-Level Variables and Respondents Indicating They Would Become a Teacher Again


 

 

 

 

 

 

95%

95%

 

 

 

 

 

Odds

Lower

Upper

Model

Variable

Coefficient

SE

 

Ratio

Limit

Limit

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Level 1: Contextual variables

Classroom % with an IEP

0.004

0.002

*

1.004

1.000

1.008

 

Classroom % LEP

0.003

0.002

 

1.003

0.999

1.007

 

Urban school setting

-0.204

0.122

 

0.815

0.642

1.035

 

Rural school setting

-0.147

0.112

 

0.863

0.694

1.075

 

Title I school setting

0.266

0.113

*

1.305

1.046

1.627

 

 

 

 

 

 

 

 

Level 1: Teacher characteristics

Total years of experience

-0.023

0.004

***

0.978

0.970

0.986

 

African American

-0.003

0.126

 

0.997

0.779

1.276

 

Asian American

0.032

0.348

 

1.033

0.522

2.043

 

Pacific Islander

1.288

0.621

*

3.625

1.073

12.248

 

Native American

0.008

0.172

 

1.008

0.720

1.413

 

Hispanic

-0.237

0.096

*

0.789

0.654

0.952

 

Male

-0.203

0.129

 

0.817

0.634

1.053

 

New to school

0.221

0.134

 

1.247

0.958

1.624

 

New to teaching

0.634

0.412

 

1.886

0.841

4.227

 

Subject area specialist

0.092

0.097

 

1.096

0.907

1.324

 

Special education teacher

-0.384

0.210

 

0.681

0.451

1.027

 

Alternatively certified

0.012

0.165

 

1.012

0.733

1.399

 

No certification

0.168

0.188

 

1.183

0.819

1.710

 

Classroom Control

0.340

0.060

***

1.404

1.248

1.578

 

Demands group

-0.726

0.088

***

0.484

0.407

0.575

 

Resourced group

0.772

0.092

***

2.164

1.807

2.590

 

 

 

 

 

 

 

 

Level 2: State policy climate

Intercept

2.058

0.139

***

7.834

5.918

10.370

 

Chance for Success

0.041

0.016

*

1.042

1.009

1.076

 

K–12 Achievement

-0.019

0.013

 

0.982

0.956

1.008

 

The Teaching Profession

-0.005

0.006

 

0.995

0.983

1.008

 

School Finance

0.011

0.012

 

1.011

0.987

1.035

 

Transitions and Alignment

-0.003

0.009

 

0.997

0.978

1.015

 

Accountability

0.008

0.010

 

1.008

0.987

1.029

 

 

 

 

 

 

 

 

***p < .001. **p < .01. *p < .05.

 

 

 

 

 

 


Table 6. Association Between Teacher- and State-Level Variables and Intention to Return to Teaching


 

 

 

 

 

 

95%

95%

 

 

 

 

 

Odds

Lower

Upper

Model

Variable

Coefficient

SE

 

Ratio

Limit

Limit

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Level 1: Contextual variables

Classroom % with an IEP

0.003

0.002

*

1.003

1.000

1.006

 

Classroom % with LEP

0.001

0.002

 

1.001

0.998

1.004

 

Urban school setting

-0.112

0.100

 

0.894

0.735

1.088

 

Rural school setting

-0.074

0.070

 

0.929

0.810

1.065

 

Title I school setting

0.154

0.071

 

1.166

1.015

1.339

 

 

 

 

 

 

 

 

Level 1: Teacher characteristics

Total years of experience

0.027

0.005

***

1.027

1.018

1.037

 

African American

-0.151

0.129

 

0.860

0.668

1.107

 

Asian American

0.217

0.221

 

1.242

0.805

1.917

 

Pacific Islander

-0.921

0.811

 

0.398

0.081

1.951

 

Native American

-0.005

0.341

 

0.995

0.510

1.940

 

Hispanic

-0.025

0.117

 

0.975

0.776

1.226

 

Male

0.170

0.102

 

1.185

0.971

1.447

 

New to school

-0.099

0.091

 

0.906

0.758

1.082

 

New to teaching

0.597

0.183

***

1.817

1.270

2.598

 

Subject area specialist

-0.099

0.059

 

0.906

0.808

1.016

 

Special education teacher

-0.331

0.132

*

0.718

0.554

0.931

 

Alternatively certified

-0.198

0.090

*

0.821

0.688

0.979

 

No certification

0.090

0.149

 

1.094

0.817

1.466

 

Classroom Control

0.114

0.035

***

1.121

1.048

1.200

 

Demands group

-0.503

0.083

***

0.605

0.514

0.712

 

Resourced group

0.425

0.073

***

1.529

1.324

1.766

 

 

 

 

 

 

 

 

Level 2: State policy climate

Intercept

1.374

0.096

***

3.953

3.259

4.794

 

Chance for Success

0.014

0.013

 

1.014

0.987

1.042

 

K–12 Achievement

-0.024

0.013

 

0.976

0.951

1.002

 

The Teaching Profession

0.006

0.007

 

1.006

0.992

1.020

 

School Finance

0.026

0.008

**

1.026

1.009

1.043

 

Transitions and Alignment

-0.009

0.008

 

0.991

0.977

1.007

 

Accountability

0.006

0.008

 

1.006

0.989

1.022

 

 

 

 

 

 

 

 

***p < .001. **p < .01. *p < .05.

 

 

 

 

 

 



Only one classroom characteristic, classroom percentage of children with IEPs, was positively associated with both willingness to become a teacher again and intention to return. Teachers working in Title I schools had a higher probability of willingness to become a teacher again (odds ratio = 1.305) than teachers in other schools.


Several indicators of teacher experience were associated with intention to return to teaching and willingness to become a teacher again. Total years of teaching experience was positively associated with willingness to return, yet negatively associated with willingness to become a teacher again. New teachers were also more likely to express an intention to return to teaching (odds ratio = 1.817) than all other teachers.


Both alternatively certified teachers and special education teachers were less likely to report an intention to return than all other teachers. Pacific Islander teachers had a much higher probability of willingness to become a teacher again (odds ratio = 3.625) than White teachers. Hispanic teachers had a lower probability of willingness to become a teacher again.


Teachers with higher Classroom Control scores also tended to have a higher probability of intention to return and tended to have a higher probability of willingness to become a teacher again. Teachers in the Demands group had a much lower probability of intention to return (odds ratio = .605) and a much lower probability of willingness to become a teacher again (odds ratio = .484). Similarly, teachers in the Resourced group had a much higher probability of intention to return (odds ratio = 1.529) than teachers in the Balanced group and a much higher probability of willingness to become a teacher again (odds ratio = 2.164).


Only one state policy climate variable was statistically significantly associated with willingness to become a teacher again. States with higher Chance for Success scores tended to have higher statewide probabilities of willingness to become a teacher again. The pseudo R2 value for the Level 2 model was .460. Also, one state policy climate variable was statistically significantly associated with intention to return. States with higher School Finance scores tended to have higher statewide probabilities of intention to return. The pseudo R2 value for the Level 2 model was .487.


DISCUSSION AND IMPLICATIONS


Policy implications stemming from the transactional model of stress (Lazarus & Folkman,1984) have been limited because of insufficient operationalization of the central construct of appraisal. Addressing this limitation in the literature, this study supported the validity of assessing an elementary teacher’s risk for stress using an Appraisal Index, an approach supported in previous research (McCarthy et al., 2016) and extended to a large national data set in this study. Consistent with theory, elementary teachers classified as being at higher risk for stress were more likely in this study to report reduced occupational health across a variety of indices.


Perhaps most important, this study demonstrated that stress is not a “one size fits all” proposition for teachers: Understanding risk for stress through the lens of classroom appraisals could allow researchers, policy makers, and administrators to better understand factors that are associated with higher and lower levels of risk for stress among elementary teachers. Meaningful results obtained in this study are discussed in progressive order of the research questions. First, we review how teacher, classroom, school, and policy contexts contribute to teachers’ risk for stress. Then, we review how the aforementioned predictors and classification as at risk for stress correlate with burnout-like symptoms. Finally, we discuss how elementary teachers’ risk for stress classification, along with other teacher, classroom, school, and policy variables, is associated with professional commitment.


IMPLICATIONS FOR RESEARCH QUESTION 1: ASSOCIATIONS OF PROFESSIONAL CHARACTERISTICS, SCHOOL CONTEXT, AND STATE POLICY CLIMATE WITH RISK FOR OCCUPATIONAL STRESS


Results indicate that elementary teachers new to their school were less likely to be at risk for stress than their colleagues, which might signify that stress can accumulate the longer a teacher stays at a particular school. However, results also suggest that teachers’ risk for stress decreases as years of experience increases. These seemingly contradictory results illustrate an important nuance in elementary teachers’ workplace appraisals and subsequent risk for stress. When teachers are new to a school, they tend to appraise their environs more favorably (Goddard, O’Brien, & Goddard, 2006), perhaps because of their lack of familiarity with the new workplace environment. Therefore, a “change of scenery” may be conducive to lowering risk for stress for some teachers, but this effect may wane over time.


Conversely, elementary teachers’ years of experience is associated with a decrease in risk for stress. At one level, this finding points to a potential selection bias. Among more experienced teachers, those who left the profession at an earlier career stage because of lack of job satisfaction may no longer be present in this sample. However, these findings also suggest that if teachers remain in the profession, they become more adept at mitigating demands and bolstering resources. Research on the professional phases of teaching seems to support the claim that teachers learn how to adapt to workplace demands (Hargreaves, 2005; Katz, 1972). Administrators, therefore, might concentrate efforts to identify and support educators most likely to perceive their schools as demanding, including less experienced teachers who are not new to the school.


Supporting previous research (e.g., Billingsley, 2004), being a special education teacher or subject area specialist is associated with increased risk for stress, implying that these niche teachers appraise their environments differently from regular K–5 educators. Subject area specialists (i.e., physical education, art, and band) are less likely to have stable classroom environments. Students (or the teachers themselves) migrate in and out of these “specials,” creating workplace environments that have their own set of challenges. Specifically, both special education and subject area specialists are often viewed as ancillary teachers, lacking the supportive network of grade level colleagues and administrative recognition, which may contribute to these teachers being at higher risk for stress.


Findings also indicate that classroom and school contexts, including increased levels of IEP and LEP students per classroom, Title I status, and nonsuburban school indicators, are associated with risk for stress, aligning with previous research (McCarthy et al., 2016). Thus, working in environments with students who require the additional resources and instructional support can be demanding for elementary teachers. In elementary classrooms, where teachers are required to juggle multiple subjects and quite frequently teach in academically heterogeneous classrooms, addressing student accommodations is a daunting challenge. These additional demands contribute to teachers’ demanded appraisals. Results from this study suggest that principals and school leadership should weigh cautiously the assignment of students to classrooms. Overwhelming elementary teachers with students who require (and deserve) additional supports may adversely impact the occupational health of the teachers, which has negative consequences for instructional performance (Jennings, Frank, Snowberg, Coccia, & Greenberg, 2013).


An inspection of Tables 2 and 3 shows that alternatively licensed teachers are also more likely to be at risk for stress. Alternative licensed teachers often receive less traditional preservice education in lieu of on-the-job training to fill hard-to-staff positions (Darling-Hammond, Holtzman, Gaitlin, & Helig, 2005). Lacking sufficient preparation, coupled with working in challenging workplace environments, potentially exacerbates teachers’ risk for occupational stress. Furthermore, results from the study indicate that working in schools with higher levels of poverty (Title I) situated in urban environments is associated with increased levels of stress. Given the growing prevalence of alternative licensure programs (Darling-Hammond et al., 2005), greater attention should be placed on the level of preparation afforded these teachers. Future research should examine the extent to which alternatively certified teachers navigate their school environs compared with more traditionally certified programs. Though traditional teacher education now adopts many of the innovative policies found in alternative pathways, perhaps it would also be advantageous to investigate how conventional programs help teachers leverage existing support structures within schools.


Outside of teacher and classroom characteristics, Tables 2 and 3 indicate that elementary teachers reporting higher levels of classroom control inversely correlate with teachers’ risk for stress. Existing research suggests that teachers who have control over their workplace report greater workplace satisfaction and teacher self-efficacy (Pearson & Moomaw, 2005). Principals, as the education policy mediators within schools (Johnson, 2006), have the authority to grant teachers greater autonomy over their professional lives. Studies indicate that teachers empowered by principals indicate more satisfaction with their workplace (e.g., Johnson et al., 2014). Similarly, the current study suggests that teachers who perceive agency over day-to-day school operations perceive their environment as less demanded. Consequently, school leaders who support greater classroom-level independence among teachers might also help ameliorate teachers’ levels of occupational stress. Future research is needed to explore how teachers’ perceptions of classroom agency contribute to their risk for stress.


The inclusion of Quality Counts indicators in this study served the purpose of assessing whether policies aimed at tightening the loosely coupled system of public education are associated with teachers’ risk for stress. One rationale for imposing standardized assessments is to hold teachers accountable for students’ academic progress, purportedly to improve teacher quality and student learning (Darling-Hammond & Rustique-Forrester, 2005; Ingersoll, 2009). This study found that teachers working in states with higher rated Quality Counts accountability indices (which often constrain teachers’ work, time, and decision making) report increased levels of workplace stress, whereas teachers working in states with high marks on curriculum transition and alignment are less likely to be at risk for stress. These contradictory findings suggest that education policy, which potentially sanctions or incentivizes teachers, requires clarity and support structures to minimize the risk for occupational stress. Policy actors should consider the extent that education initiatives with strong accountability standards also provide clear alignment from one grade level to the next, thus helping to ensure that the aims of education better meet the capabilities of teachers and the students they serve.


IMPLICATIONS FOR RESEARCH QUESTION 2: ASSOCIATIONS OF PROFESSIONAL CHARACTERISTICS, SCHOOL CONTEXT, AND STATE POLICY CLIMATE WITH BURNOUT SYMPTOMS


Elementary teachers classified in the Demands group reported increased burnout symptoms compared with counterparts in the Balanced group, whereas Resourced teachers were significantly less likely to report burnout symptoms. This finding aligns with the considerable attention teacher burnout has received in the literature as a consequence of prolonged exposure to stress (Schaufeli & Enzmann, 1998). However, previous research has focused on burnout as a syndrome that develops over years and involves isolation from school colleagues, cynicism related to their workplace, and a sense of hopeless regarding job performance (Blase, 1986; Chang, 2009; Kyriacou, 1987). This study relied on burnout symptoms item available in the SASS, meaning that the full syndrome of burnout was not addressed. Although a limitation in terms of aligning this study’s findings with some previous research, examining associations between risk for stress and burnout symptoms, which could eventually result in full-fledged burnout, was an important goal of this study.


Results from the current study align with previous research (Jennings & Greenberg, 2009; Kyriacou, 1987; McCarthy et al., 2009) suggesting that risk for stress is associated with burnout symptoms. These findings offer compelling evidence that, using a classification system such the one presented in this study and in previous research (Lambert et al., 2015; McCarthy et al., 2016), school leaders might be able to determine which teachers are at risk for stress and therefore more likely to eventually experience burnout. Examining associations in this way uncovered unique complexities in the data. Interestingly, even after controlling for teachers’ appraisals of Demands and Resources (Appraisal Index), being a new teacher is associated with decreased burnout symptoms compared with more experienced educators. This finding aligns with previous research indicating that beginning teachers enter the classroom more idealistic and appraise their workplace more favorably than their more experienced colleagues (Goddard et al., 2006) and with research suggesting that burnout accumulates over time (Blase, 1986; Chang, 2009; Kyriacou, 1987). However, as school years go by, the newer teachers tend to replace their rose-tinted glasses with a more cynical view of the workplace (Hargreaves, 2005). Taken together, these results suggest that educational stakeholders could improve the occupational health of teaching staffs by identifying which teachers are most at risk for stress (i.e., the Demands group) but not yet reporting higher levels of burnout symptoms. It is here that preventive efforts could be fruitful, before the accumulation of stress results in the burnout syndrome that can be difficult to reverse (Freidman, 2006).


In addition to new teachers, findings indicated that African American teachers report increased burnout like symptoms compared with their colleagues. Research on teachers of color suggests that as a subgroup, they often perceive greater isolation, increased classroom management responsibility, and less collegiality compared with their White counterparts (Achinstein, Ogawa, Sexton, & Freitas, 2010; Kelly, 2007; Kohli, 2018). Interestingly, findings from the current study indicate that after controlling for Appraisal Index scores, African American teachers more likely to report greater burnout. This finding echoes what former U.S. Secretary of Education John King (2016) referred to as a “hidden tax” paid by African Americans—a burdensome load as role model, mentor, and disciplinarian. Previous studies note that African American teachers who perceive their school leadership as supportive report higher job satisfaction and remain in the classroom, perhaps countermanding perceived burdens (Achinstein et al., 2010). Interestingly, nascent studies analyzing the racial congruence of teachers and students suggest that African American teachers working with racially and ethnically congruent students are more likely to perceive their workplace positively (Fitchett, Hopper, Eyal, McCarthy, & Lambert, 2017; Villegas, Strom, & Lucas, 2012). This uniquely African American experience deserves further exploration, especially given the increased attrition rates of non-White teachers (Albert Shanker Institute, 2015).


Counterintuitively, teachers in states with higher ratings from Quality Counts on teacher professionalism (which includes measures tying teacher quality to student achievement) are more likely to report burnout symptoms. Although teacher professionalism remains an elusive goal in the eyes of many education advocates, the current evidence suggests that policies aimed at creating top-down mandates under the guise of professionalism might actually contribute to workplace stress and anxiety among practitioners. Constrained and controlling school environments are often referred to as bureaucratic intensification (Hargreaves, 1994). These intense, hierarchical policies have real consequences for teacher esprit de corps and workplace motivation. Creating such mandates, in some ways, might be antithetical to professionalizing teachers and deleterious to the well-being of classroom practitioners. Statutes designed to ramp up expectations of teacher quality using credentialing and student test scores might also adversely impact the overall occupational health of teachers who perceive these new policies as increasing demands on their occupational lives.


IMPLICATIONS FOR RESEARCH QUESTION 3: ASSOCIATIONS OF PROFESSIONAL CHARACTERISTICS WITH RISK FOR STRESS, AND STATE POLICY CLIMATE WITH OCCUPATIONAL COMMITMENT


Ultimately, elementary teachers’ dispositions and mental health influence their commitment to the classroom, which was assessed in this study using teachers’ responses to the items would be a teacher again (Table 5) and intent to remain in teaching (Table 6). Findings from the current study indicate that, when controlling for other teacher and classroom characteristics, Resourced teachers correlate with greater overall commitment to teaching compared with Balanced Teachers, whereas Demanded teachers appear less committed to teaching. Collectively, the analyses suggest that elementary teachers’ risk for stress is potentially associated with their (dis)connection with the profession. Teachers who self-appraise greater workplace demands report decreased obligation to the job. This lack of commitment also potentially contributes to increased teacher attrition (Ingersoll, 2001). Although turnover of inadequate and underperforming teachers is both justified and advantageous for schools (Adnot, Dee, Katz, & Wyckoff, 2016), a lack of professional commitment among stress-vulnerable teachers might potentially push out hardworking, yet overburdened, educators. Research suggests that unstable school staffing inversely affects school learning climates (Ronfeldt, Loeb, & Wyckoff, 2013). Classifying teachers based on their workplace appraisals offers one potential solution toward forecasting teachers’ risk for stress and increased likelihood for leaving the classroom. Education stakeholders can use appraisal models like the CARD or the items operationalized in this study to pinpoint high-performing, stress-vulnerable teachers; tailor supports designed toward resourcing these teachers; and counsel out of the profession others for whom resources are not feasible.


Results in Tables 5 and 6 illustrate numerous findings related to teacher characteristics. When controlling for classroom appraisal, the percentage of IEP students in the classroom is positively associated with professional intention. Accounting for teachers’ stress appraisal and the characteristics of the students, the finding demonstrates the unique value-added effect of a teacher who is committed to a particular subset of students. Furthermore, teachers’ years of experience is positively associated with professional intention to remain teaching, yet negatively correlates with would be a teacher again. When accounting for the appraisal model, results suggest that more experienced teachers are less inclined to leave teaching and find employment in other employment sectors. However, when considering their careers retrospectively, they are more likely to question their professional lives. Lynn (2002) referred to the “career wind-down” (p. 181) of teachers, a stage where teachers are more likely to be content in their work but also have conflicted feelings regarding their job performance. Collectively, results suggest that when controlling for teachers’ appraisals, unique attributes related to both professional longevity and teachers’ commitment to students manifest. To potentially improve professional commitment, education stakeholders in teacher education and school leadership should consider how educators are taught to cope with the inevitable classroom demands.


Furthermore, teachers working in states with a higher Chance for Success Quality Counts index are associated with an increased likelihood that they would choose teaching again. Moreover, increased School Finance indices are associated with an increased professional intention to remain in the teaching profession. Thus, public policy offering macro-level support of resources outside of schools (cultural capital, parental employment), along with in-school supports (per pupil expenditure), correlates with increased professional commitment among teachers.


LIMITATIONS


This study offered a unique analysis of teachers’ risk for stress and its association with burnout-like systems and professional commitment. Results from this study included building-level characteristics as contextual variables at the teacher level. It is possible that some of the variance attributed to teachers might be associated with unique school environments, such as more stressful schools. Future analysis should attempt to examine the specific contribution of schools on teachers’ risk for stress.


The data used to report these findings were self-reported. However, given that the overall purpose of the study hinged on teachers’ self-appraisals, and NCES secondary data are deidentified, we contend that the potential for social desirability bias is minimal. Moreover, this study only examined elementary-grade teachers. Because middle and high school teachers work in different contexts, it is feasible that their appraisals of the workplace would be different. Additional studies are needed to examine the unique perceptions of teachers working at other grade levels.


Finally, this study employed cross-sectional data. In making more accurate estimations of teachers’ professional commitment, future research would benefit from the use of longitudinal data to determine the extent to which teachers’ intentions to remain or leave translate into actually leaving the profession. Longitudinal research could also be employed to extend the cross-sectional associations found in this study into an examination of the extent to which stressors, across multiple cohorts of teachers, lead to the demanded condition and in turn to burnout and exiting the teaching profession.


CONCLUSION


Efforts to promote teachers’ occupational health are unlikely to be successful until we better understand the complex relationships that exist among risk for stress and teacher characteristics, school context, and policy climate. The results of this study suggested that risk for stress can be reliably assessed using a national data set. At the local level, these findings have important implications for teacher educators and school leadership. Although certain teacher factors, such as classroom control and alternative licensure, were consistently related to occupational health, other factors, such teacher experience, were more nuanced. Findings support efforts by teacher educators to equip beginning practitioners with strategies for navigating potential resources while coping with the various the workplace demands of the classroom. Moreover, principals who want to improve the occupational health of their faculty might consider supporting teachers’ classroom independence and using classification systems based on balanced theories, like those presented in this study and others, to identify vulnerable staff members and potentially protect against teacher stress. At the policy level, seemingly well-intentioned policies, such as efforts to improve teacher quality and professionalism, might actually burden teachers with increased demands. Conversely, providing increased clarity of curricular alignment, thereby making the work of teachers more transparent, perhaps alleviates teacher stress. Education stakeholders seeking to nurture the occupational health of the school labor force should be mindful of the climate and context in which public policies are enacted.


Notes


1. The Classroom Control scale was constructed from Questions 54a–54f of the SASS Teacher Questionnaire. These Likert scale items correspond to variables T0280–T0285. The focus of these items is the sense of control teachers feel they have over various aspects of their planning and teaching. Sample items include “Selecting textbooks and other instructional materials” (54a, T0280) and “Selecting teaching techniques” (54c, T0283).


2. The Burnout scale was constructed from Questions 57a–57g of the SASS Teacher Questionnaire. These Likert scale items correspond to variables T0313–T0319. The focus of these items is the emotional reactions teachers have to the working conditions in their school. Sample items include, “The stress and disappointments of teaching at this school really are not worth it” (57a, T0313), and “I think about staying home from school because I am just too tired to go” (57g, T0319).


3. It is not possible to calculate r2 in a multilevel modeling context. However, it is possible to calculate pseudo R2 statistics that estimate the proportion of the variance in the outcome variable that is associated with the explanatory variables based on the following formula (Kreft & de Leeuw, 1998; Singer, 1998)

Pseudo R2 = 1 – (restricted error variance / unrestricted error variance). These statistics can be calculated for each level in a multilevel model. The unrestricted error variance terms for each level are based on the null model (no explanatory variables) and the restricted variance terms are based on the residual variances at each level after the explanatory variables have been add to the model. By extending J. Cohen’s interpretation guidelines for using r as an effect size measure (J. Cohen, 1992), values .09 - .25 can be considered moderate and greater than .25 can be considered large.


References


Achinstein, B., Ogawa, R., Sexton, D., & Freitas, C. (2010). Retaining teachers of color: A pressing problem and a potential strategy for “hard-to-staff” schools. Review of Educational Research, 80(1), 71–107.


Adnot, M., Dee, T., Katz, V., & Wyckoff, J. (2016). Teacher turnover, teacher quality, and student achievement in DCPS. Educational Evaluation and Policy Analysis, 39(1), 54–76. doi:10.3102/0162373716663646


Albert Shanker Institute. (2015). The state of teacher diversity in American education. Retrieved from http://www.shankerinstitute.org/sites/shanker/files/The%20State%20of%20Teacher%20Diversity_0.pdf


Aloe, A. M., Shisler, S. M., Norris, B. D., Nickerson, A. B., & Rinker, T. W. (2014). A multivariate meta-analysis of student misbehavior and teacher burnout. Educational Research Review, 12, 30–44. https://doi.org/10.1016/j.edurev.2014.05.003


Apple, M. W. (2004). Controlling the work of teachers. In D. J. Flinders & S. J. Thornton (Eds.), The curriculum studies reader (pp. 183–198). New York, NY: Psychology Press.


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(2), 274–284. https://doi.org/10.1037/0022-0663.99.2.274


Billingsley, B. S. (2004). Special education teacher retention and attrition a critical analysis of the research literature. Journal of Special Education, 38(1), 39–55. https://doi.org/10.1177/00224669040380010401


Blase, J. J. (1986). A qualitative analysis of sources of teachers stress: Consequences for performance. American Educational Research Journal, 23(1), 13–40. https://doi.org/10.3102/00028312023001013


Boyd, D., Grossman, P., Ing, M., Lankford, H., Loeb, S., & Wyckoff, J. (2011). The influence of school administrators on teacher retention decisions. American Educational Research Journal, 48(2), 303–333.


Brayfield, A. H., & Rothe, H. F. (1951). An index of job satisfaction. Journal of Applied Psychology, 35(5), 307. https://doi.org/10.1037/h0055617


Chang, M. L. (2009). An appraisal perspective of teacher burnout: Examining the emotional work of teachers. Educational Psychology Review, 21(3), 193–218.


Cochran-Smith, M., & Lytle, S. (2006). Troubling images of teaching in No Child Left Behind. Harvard Educational Review, 76(4), 668–697. https://doi.org/10.17763/haer.76.4.56v8881368215714


Cohen, D. K., & Moffitt, S. L. (2009). The ordeal of equality: Did federal regulation fix the schools? Cambridge, MA: Harvard University Press.


Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155–159. doi:10.1037/0033-2909.112.1.155


Crocco, M. S., & Costigan, A. T. (2007). The narrowing of curriculum and pedagogy in the age of accountability: Urban educators speak out. Urban Education, 42(6), 512–535. https://doi.org/10.1177/0042085907304964


Darling-Hammond, L., Holtzman, D. J., Gatlin, S. J., & Heilig, J. V. (2005). Does teacher preparation matter? Evidence about teacher certification, teach for America, and teacher effectiveness. Education Policy Analysis Archives, 13(0), 42. https://doi.org/10.14507/epaa.v13n42.2005


Darling-Hammond, L., & Rustique-Forrester, E. (2005). The consequences of student testing for teaching and teacher quality. Yearbook of the National Society for the Study of Education, 104(2), 289–319. https://doi.org/10.1111/j.1744-7984.2005.00034.x


Farber, B. A. (1991). Crisis in education: Stress and burnout in the American teacher (Vol. 21). San Francisco, CA: Jossey-Bass.


Fitchett, P., Hopper, E. B., Eyal, M., McCarthy, C. J., & Lambert, R. G. (2017). Racial congruence, teacher stress and professional commitment among African American female teachers. In A. Farine-Wu, A. Allen, & C. W. Lewis (Eds.), Black female teachers: Diversifying the United States’ teacher workforce (pp. 135–158). Bingley, England: Emerald.


Folkman, S., Lazarus, R. S., Dunkel-Schetter, C., DeLongis, A., & Gruen, R. J. (1986). Dynamics of a stressful encounter: Cognitive appraisal, coping, and encounter outcomes. Journal of Personality and Social Psychology, 50, 992–1003.


Friedman, I. A. (2006). Classroom management and teacher stress and burnout. In C. M. Evertson & C. S. Weinstein (Eds.), Handbook of classroom management: Research, practice, and contemporary issues (pp. 925–944). Mahwah, NJ: Erlbaum.


Gilbert, R. B., Adesope, O. O., & Schroeder, N. L. (2014). Efficacy beliefs, job satisfaction, stress and their influence on the occupational commitment of English-medium content teachers in the Dominican Republic. Educational Psychology, 34(7), 876–899. https://doi.org/10.1080/01443410.2013.814193


Goddard, R., O’Brien, P., & Goddard, M. (2006). Work environment predictors of beginning teacher burnout. British Educational Research Journal, 32(6), 857–874.


Good, T. L., & Brophy, J. E. (2000). Looking in classrooms (8th ed.). New York, NY: Longman.


Goodlad, J. I. (2004). A place called school. New York, NY: McGraw-Hill.


Grissom, J. A., Nicholson-Crotty, S., & Harrington, J. R. (2014). Estimating the effects of No Child Left Behind on teachers’ work environments and job attitudes. Educational Evaluation and Policy Analysis, 36(4), 417–436. https://doi.org/10.3102/0162373714533817


Hanushek, E. A. (2008). Incentives for efficiency and equity in the school system. Perspektiven Der Wirtschaftspolitik, 9(s1), 5–27. https://doi.org/10.1111/j.1468-2516.2008.00272.x


Hargreaves, A. (1994). Changing teachers, changing times: Teachers’ work and culture in the postmodern age. New York, NY: Teachers College Press.


Hargreaves, A. (2005). Educational change takes ages: Life, career, and generational factors in teachers’ emotional responses to educational change. Teaching and Teacher Education, 21, 967–983.


Hobfoll, S. E. (1989). Conservation of resources: A new attempt at conceptualizing stress. American Psychologist, 44(3), 513. https://doi.org/10.1037//0003-066X.44.3.513


Ingersoll, R. M. (2001). Teacher turnover and teacher shortages: An organizational analysis. American Educational Research Journal, 38(3), 499–534.


Ingersoll, R. M. (2007). Short on power, long on responsibility. Educational Leadership, 65(1), 20–-25.


Ingersoll, R. M. (2009). Who controls teachers’ work? Power and accountability in America’s schools. Cambridge, MA: Harvard University Press.


Ingersoll, R. M., & Merrill, E. (2011). The status of teaching as a profession. In J. Ballantine & J. Spade (Eds.), Schools and society: A sociological approach to education (4th ed., pp. 185–189). Los Angeles, CA: Pine Forge Press/Sage Publications.


Ingersoll, R. M., & Smith, T. M. (2003). The wrong solution to the teacher shortage. Educational Leadership, 60(8), 30–33.


Jennings, P. A., Frank, J. L., Snowberg, K. E., Coccia, M. A., & Greenberg, M. T. (2013). Improving classroom learning environments by Cultivating Awareness and Resilience in Education (CARE): Results of a randomized controlled trial. School Psychology Quarterly, 28(4), 374–390. doi:10.1037/spq0000035


Jennings, P. A., & Greenberg, M. T. (2009). The prosocial classroom: Teacher social and emotional competence in relation to student classroom outcomes. Review of Educational Research, 79(1), 491–525. https://doi.org/10.3102/0034654308325693


Jepson, E., & Forrest, S. (2006). Individual contributory factors in teacher stress: The role of achievement striving and occupational commitment. British Journal of Educational Psychology, 76(1), 183–197. http://doi.org/10.1348/000709905X37299


Johnson, S. M. (2004). Finders and keepers: Helping new teachers survive and thrive in our schools. San Francisco, CA: Jossey-Bass.


Johnson, S. M. (2006). The workplace matters: Teacher quality, retention, and effectiveness (Working paper). Washington, DC: National Education Association Research Department. Retrieved from http://eric.ed.gov/?id=ED495822


Johnson, S. M., Reinhorn, S. K., Charner-Laird, M., Kraft, M. A., Ng, M., & Papay, J. P. (2014). Ready to lead, but how? Teachers’ experiences in high-poverty urban schools. Teachers College Record (1970), 116(10). Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4391200/


Katz, L. G., (1972). Developmental stages of preschool teachers. The Elementary School Journal, 73(1), 50–54.


Kelly, H. (2007). Racial tokenism in the school workplace: An exploratory study of Black teachers in overwhelmingly White schools. Educational Studies, 41(3), 230–254. doi:10.1080/00131940701325712


King, J. (2016, May 15). The invisible tax on teachers of color. Washington Post, 1–4. Retrieved from https://www.washingtonpost.com/opinions/the-invisible-tax-on-black-teachers/2016/05/15/6b7bea06-16f7-11e6-aa55-670cabef46e0_story.html


Klassen, R. M., & Chiu, M. M. (2011). The occupational commitment and intention to quit of practicing and pre-service teachers: Influence of self-efficacy, job stress, and teaching context. Contemporary Educational Psychology, 36(2), 114–129. http://doi.org/10.1016/j.cedpsych.2011.01.002


Kohli, R. (2018). Behind school doors: The impact of hostile racial climates on urban teachers of color. Urban Education, 53(3), 1–27. doi:10.1177/0042085916636653


Kreft, I., & De Leeuw, J. (1998). Introducing multilevel modeling. London, England: Sage.


Kyriacou, C. (1987). Teacher stress and burnout: An international review. Educational Research, 29(2), 146–152. doi:10.1080/0013188870290207


Ladd, H. F. (2011). Teachers’ perceptions of their working conditions: How predictive of planned and actual teacher movement? Educational Evaluation and Policy Analysis, 33(2), 235–261. http://doi.org/10.3102/01623711398128


Lambert, R. G., McCarthy, C., Fitchett, P. G., Lineback, S., & Reiser, J. (2015). Identification of elementary teachers’ risk for stress and vocational concerns using the national schools and staffing survey. Education Policy Analysis Archives, 23(43). http://dx.doi.org/10.14507/epaa.v23.1792


Lambert, R. G., McCarthy, C. J., O’ Donnell, M., & Wang, C. (2009). Measuring elementary teacher stress and coping in the classroom: Validity evidence for the classroom appraisal of resources and demands. Psychology in the Schools, 46, 973–988.


Lauen, D. L., & Tyson, K. (2009). Perspectives from the disciplines: Sociological contributions to education policy research and debates. In G. Sykes, B. Schneider, & D. N. Plank (Eds.), Handbook of education policy research (pp. 71–82). New York, NY: Routledge.


Lazarus, R. (2001). Relational meaning and discrete emotions. In K. R. Scherer, A. Schorr, & T. Johnstone (Eds.), Appraisal processes in emotion: Theory, methods, research (pp. 37–67). Series in Affective Science. New York, NY: Oxford University Press.


Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York, NY: Springer.


Liu, X. S., & Ramsey, J. (2008). Teachers’ job satisfaction: Analyses of the teacher follow-up survey in the United States for 2000–2001. Teaching and Teacher Education, 24(5), 1173–1184. http://doi.org/10.1016/j.tate.2006.11.010


Loeb, S., Darling-Hammond, L., & Luczak, J. (2005). How teaching conditions predict teacher turnover in California schools. Peabody Journal of Education, 80(3), 44–70. http://doi.org/10.1207/s15327930pje8003_4


Lynn, S. K. (2002). The winding path: Understanding the career cycle of teachers. The Clearing House: A Journal of Educational Strategies, Issues and Ideas, 75(4), 179–182. doi:10.1080/00098650209604926


Marvel, J., Lyter, D. M., Peltola, P., Strizek, G. A., Morton, B. A., & Rowland, R. (2007). Teacher attrition and mobility: Results from the 2004–05 Teacher Follow-Up Survey (NCES 2007-307). Washington, DC: National Center for Education Statistics.


Matheny, K. B., Aycock, D. W., Curlette, W. L., & Junker, G. N. (1993). The coping resources inventory for stress: A measure of perceived resourcefulness. Journal of Clinical Psychology, 49(6), 815–830. doi:10.1002/jclp.10219


McCarthy, C. J., Lambert, R. G., Crowe, R. W., & McCarthy, C. J. (2010). Coping, stress, and job satisfaction as predictors of Advanced Placement Statistics teachers’ intention to leave the field. NASSP Bulletin, 94, 306–326.


McCarthy, C. J., Lambert, R. G., Lineback, S., Fitchett, P., & Baddouh, P. (2016). Assessing teacher appraisals and stress in the classroom: Review of the classroom appraisal of resources and demands. Educational Psychology Review, 28(3), 577–603.  http://dx.doi.org/10.1007/s10648-015-9322-6 


McCarthy, C. J., Lambert, R. G., O’Donnell, M., & Melendres, L. (2009). Relationship of elementary teachers’ experience, perceived demands, and coping resources to burnout symptoms. Elementary School Journal, 109, 1–19.


McCarthy, C. J., Lambert, R. G., & Reiser, J. (2014). Vocational concerns of elementary teachers: Stress, job satisfaction, and occupational commitment. Journal of Employment Counseling, 51, 59–74.


Meurs, J. A., & Perrewé, P. L. (2011). Cognitive activation theory of stress: An integrative theoretical approach to work stress. Journal of Management, 37, 1043–1068.


Monk, D. H. (1989). The education production function: Its evolving role in policy analysis. Educational Evaluation and Policy Analysis, 11(1), 31–45. http://doi.org/10.2307/1163714


Pearson, L. C., & Moomaw, W. (2005). The relationship between teacher autonomy and stress, work satisfaction, empowerment, and professionalism. Educational Research Quarterly, 29(1), 37–53.


Plank, D. N., & Keesler, V. (2009). Education and the shrinking state. The handbook of education policy research. In G. Sykes, B. Schneider, & D. N. Plank (Eds.), Handbook of education policy research (pp. 694–704). New York, NY: Routledge.


Raymond, M. (2010). Quality counts and the Chance-For-Success index. Education Next, 10(2). Retrieved from http://search.proquest.com/openview/7fa3eeabaf262b34ffd0494935a3a88f/1?pq-origsite=gscholar&cbl=1766362


Ronfeldt, M., Loeb, S., & Wyckoff, J. (2013). How teacher turnover harms student achievement. American Educational Research Journal, 50(1), 4–36. doi:10.3102/0002831212463813


Schaufeli, W., & Enzmann, D. (1998). The burnout companion to study and practice: A critical analysis. London, England: Taylor & Francis.


Singer, J. (1998). Using SAS PROC MIXED to fit multilevel models, hierarchical models, and individual growth models. Journal of Education and Behavioral Statistics, 24(4), 323–355.


Smith, T. M. (2007). How do statelevel induction and standardsbased reform policies affect induction experiences and turnover among new teachers? American Journal of Education, 113(2), 273–309. https://doi.org/10.1086/510168


Smith, T. M., & Ingersoll, R. M. (2004). What are the effects of induction and mentoring on beginning teacher turnover? American Educational Research Journal, 41(3), 681–714. https://doi.org/10.3102/00028312041003681


Spilt, J. L., Koomen, H. M. Y., & Thijs, J. T. (2011). Teacher wellbeing: The importance of teacher–student relationships. Educational Psychology Review, 23(4), 457–477. http://doi.org/10.1007/s10648-011-9170-y


Tourkin, S., Thomas, T., Swaim, N., Cox, S., Parmer, R., Jackson, B., . . . Zhang, B. (2010). Documentation for the 2007-08 Schools and Staffing Survey (Publication number NCES 2010332). Washington, DC: National Center for Education Statistics.


van den Tooren, M., de Jonge, J., Vlerick, P., Daniels, K., & Van de Ven, B. (2011). Job resources and matching active coping styles as moderators of the longitudinal relation between job demands and job strain. International Journal of Behavioral Medicine, 18(4), 373–383. https://doi.org/10.1007/s12529-011-9148-7


Villegas, A. M., Strom, K., & Lucas, T. (2012). Closing the ethnic/racial gap between students of color and their teachers: An elusive goal. Equity & Excellence in Education, 45(2), 283–301. doi:10.1080/10665684.2012.656541


Weick, K. E. (1976). Educational organizations as loosely coupled systems. Administrative Science Quarterly, 21(1), 1–19. https://doi.org/10.2307/2391875


Yang, X., Ge, C., Hu, B., Chi, T., & Wang, L. (2009). Relationship between quality of life and occupational stress among teachers. Public Health, 123(11), 750–755. doi:10.1016/j.puhe.2009.09.018



APPENDIX


Statistical Models


Level-1 Model – Continuous Outcomes


  Continuous Outcomeij = β0j + β1j*(TOTYREXPij) + β2j*(CLSCNTij) + β3j*(IEPij) + β4j*(LEPij) + β5j*(NEWTOTCHij) + β6j*(NEWTOSCHij) + β7j*(MALEij) + β8j*(AFAMij) + β9j*(ASNij) + β10j*(PACIij) + β11j*(NTAMij) + β12j*(HISPij) + β13j*(ALTCERTij) + β14j*(NO_CERTij) + β15j*(TITLE1ij) + β16j*(SASij) + β17j*(SPEDij) + β18j*(URBANij) + β19j*(RURALij) + rij

 
 Level-2 Model – Continuous Outcomes

β0j = γ00 + γ01*(CFSj) + γ02*(ACHj) + γ03*(TTPj) + γ04*(SFj) + γ05*(TAAj) + γ06*(SAAj) + u0j


Level-1 Model – Dichotomous Outcomes

  Prob(Dichotomous Outcomeij=1|βj) = ϕij
 log[ϕij/(1 - ϕij)] = ηij
 ηij = β0j + β1j*(TOTYREXPij) + β2j*(CLSCNTij) + β3j*(IEPij) + β4j*(LEPij) + β5j*(NEWTOTCHij) + β6j*(NEWTOSCHij) + β7j*(MALEij) + β8j*(AFAMij) + β9j*(ASNij) + β10j*(PACIij) + β11j*(NTAMij) + β12j*(HISPij) + β13j*(ALTCERTij) + β14j*(NO_CERTij) + β15j*(TITLE1ij) + β16j*(SASij) +

β 17j*(SPEDij) + β18j*(URBANij) + β19j*(RURALij)

Level-2 Model – Dichotomous Outcomes

  β0j = γ00 + γ01*(CFSj) + γ02*(ACHj) + γ03*(TTPj) + γ04*(SFj) + γ05*(TAAj) + γ06*(SAAj) + u0j


Where:

Continuous Outcome = The Appraisal Index and Burnout scale scores. Both variables have been standardized (M = 0, SD = 1) so that coefficients are scaled in standard deviation units.

Categorical Outcome = Classification in the Demands group, Would become a teacher again, and Intention to remain in teaching for the next academic year.

TOTYREXP = Teacher years of teaching experience, group mean centered, and coefficients indicate the change in the outcome for every additional year of difference from the state average years of teaching experience.

CLSCNT = Classroom Control scale score, standardized (M = 0, SD = 1) and group mean centered, and coefficients indicate the change in the outcomes for every unit of difference from the state average Classroom Control scale score.

IEP = The percentage of each teacher’s classroom that is composed of children with an IEP, scaled 0%–100%, group mean centered, and coefficients indicate the change in the outcomes for every percentage point difference from the state average.

LEP = The percentage of each teacher’s classroom that is composed of children with limited English proficiency, scaled 0%–100%, group mean centered, and coefficients indicate the change in the outcomes for every percentage point difference from the state average.

 NEWTOTCH = 0 / 1, 1 = Teacher is new to teaching, less than 2 years of teaching experience, and coefficients indicate the mean difference in the outcomes between new teachers and all other teachers.

NEWTOSCH = 0 / 1, 1 = Teacher is new to the school, less than 2 years of experience in the school, and coefficients indicate the mean difference in the outcomes between teachers new to their school and all other teachers.

MALE = 0 / 1, 1 = Male teacher, coefficients indicate the mean difference in the outcomes between males and females.

AFAM = 0 / 1, 1 = African American teacher, coefficients indicate the mean difference in the outcomes between African American teachers and White teachers.

ASN 0 / 1, 1 = Asian teacher, coefficients indicate the mean difference in the outcomes between Asian teachers and White teachers.

PACI 0 / 1, 1 = Pacific Islander teacher, coefficients indicate the mean difference in the outcomes between Pacific Islander teachers and White teachers.

 NTAM 0 / 1, 1 = Native American teacher, coefficients indicate the mean difference in the outcomes between Native American teachers and White teachers.

HISP 0 / 1, 1 = Hispanic teacher, coefficients indicate the mean difference in the outcomes between Hispanic teachers and all other teachers.

ALTCERT 0 / 1, 1 = Alternatively certified teacher, coefficients indicate the mean difference in the outcomes between alternatively certified teachers and all other teachers.

NO_CERT 0 / 1, 1 = Teacher with no certification, coefficients indicate the mean difference in the outcomes between teachers with no certification and all other teachers.

TITLEI 0 / 1, 1 = Teacher works in a Title I school, coefficients indicate the mean difference in the outcomes between teachers working in Title I schools and all other teachers.

SAS 0 / 1, 1 = Teacher is a subject area specialist or enrichment teacher, coefficients indicate the mean difference in the outcomes between these teachers and all other teachers.

SPED 0 / 1, 1 = Teacher is a special education teacher, coefficients indicate the mean difference in the outcomes between these teachers and all other teachers.

URBAN 0 / 1, 1 = Teacher works in an urban school, coefficients indicate the mean difference in the outcomes between these teachers and teachers working in suburban schools.

RURAL 0 / 1, 1 = Teacher works in a rural school, coefficients indicate the mean difference in the outcomes between these teachers and teachers working in suburban schools.

CFS = Quality Counts “Chance for Success” total score, All Quality Counts predictors are scaled 0%–100%, grand mean centered, and coefficients indicate the change in the outcomes for every percentage point difference from the national average percentage.

ACH = Quality Counts “K-12 Achievement” total score.

TTP = Quality Counts “The Teaching Profession” total score.

SF = Quality Counts “School Finance” total score.

TAA = Quality Counts “Transitions and Alignment” total score.

SAA = Quality Counts “Standards, Assessment, and Accountability” total score




Cite This Article as: Teachers College Record Volume 120 Number 12, 2018, p. 1-42
https://www.tcrecord.org ID Number: 22503, Date Accessed: 6/12/2021 7:52:45 AM

Purchase Reprint Rights for this article or review
 
Article Tools
Related Articles

Related Discussion
 
Post a Comment | Read All

About the Author
  • Richard Lambert
    University of North Carolina at Charlotte
    E-mail Author
    RICHARD LAMBERT is a professor in the Department of Educational Leadership at the University of North Carolina at Charlotte, director of the Center for Educational Measurement and Evaluation, and Editor of NHSA Dialog: A Research-to-Practice Journal for the Early Intervention Field. He earned his Ph.D. in research, measurement, and statistics and Ed.S. in counseling psychology from Georgia State University. His research interests include formative assessment for young children, applied statistics, and teacher stress and coping. Recent publications: Lambert, R. G., Kim, D., Durham, S., & Burts, D. (2017). Differentiated rates of growth across preschool dual language learners. Bilingual Research Journal, 40(1), 81–101, doi:10.1080/15235882.2016.1275884; and Martin, C., Lambert, R. G., Polly, D., Wang, C., & Pugalee, D. K. (2016). The measurement properties of the Assessing Math Concepts’ assessments of primary students’ number sense skills. Journal of Applied Measurement, 17(3).
  • Christopher McCarthy
    University of Texas at Austin
    E-mail Author
    CHRISTOPHER J. MCCARTHY is a professor in the Department of Educational Psychology at the University of Texas at Austin. His research interests include stress and coping in educational contexts, group counseling, and career development. Dr. McCarthy’s primary scholarly focus is on researching factors that cause stress for K–12 teachers and developing interventions to promote teacher wellness. He is currently Editor of the Journal for Specialists in Group Work. Recent publications: McCarthy, C. J., Whittaker, T., Boyle, L., & Eyal, M. (2017). Quantitative approaches to group work: Suggestions for best practices. Journal for Specialists in Group Work, 42(1), 3–16; and McCarthy, C. J., Lambert, R. G., Lineback, S., Fitchett, P., & Baddouh, P. (2016). Assessing teacher appraisals and stress in the classroom: Review of the Classroom Appraisal of Resources and Demands. Educational Psychology Review, 28(3), 577–603, http://dx.doi.org/10.1007/s10648-015-9322-6
  • Paul Fitchett
    University of North Carolina at Charlotte
    E-mail Author
    PAUL G. FITCHETT is an associate professor in the Cato College of Education at the University of North Carolina at Charlotte. He studies the intersections between teacher working conditions, student learning outcomes, and educational policy. His previous research has been published in Education Policy Analysis Archives, Education Policy, and Teachers and Teaching Education.
  • Maytal Eyal
    University of Texas at Austin
    E-mail Author
    MAYTAL EYAL is a doctoral student in counseling psychology at the University of Texas, Austin. She examines stress and trauma within a variety of populations, particularly among educators and members of marginalized groups. A recent publication is: Whittaker, T., Boyle, L. & Eyal, M., & McCarthy, C. J. (in press). What really happens in group research? Results of a content analysis of recent quantitative research in JSGW. Journal for Specialists in Group Work.
 
Member Center
In Print
This Month's Issue

Submit
EMAIL

Twitter

RSS