Special Educators’ Working Conditions,Self-Efficacy,and Practices Use With Students With Emotional/Behavioral Disorders

Abstract

Students with emotional/behavioral disorders (EBD) in self-contained settings depend on special educators to deliver high-quality instruction and behavior management, and special educators depend on administrators to create supportive working environments. Yet, to date, no studies have examined how working conditions relate to special educators’ provision of effective instructional or behavior management practices for students with EBD in self-contained settings. To fill this crucial gap, we conducted a national survey of 171 special educators serving students with EBD in self-contained settings. Using structural equation modeling, we found special educators who experienced more supportive working conditions (i.e., stronger logistical resources and lower demands) reported more manageable workloads, experienced less emotional exhaustion and stress, felt greater self-efficacy for instruction, and reported using evidence-supported instructional practices more often with their students. Results have implications for future research and practice.

Keywords

emotional & behavioral disorders special education teacher/staff effectiveness staff development/inservice

Students with emotional/behavioral disorders (EBD) depend on their teachers to use high-quality academic and behavior management practices to address substantial academic and social-emotional challenges (Kauffman & Landrum, 2018). Students with EBD demonstrate low academic achievement and problematic behaviors that impede success; thus, they need effective (research-supported) instruction characterized by frequent opportunities to respond to academic prompts, evidence-based practices, and high rates of feedback (Common et al., 2020). Furthermore, they need highly effective behavior management to address problematic behaviors (e.g., behavior contracts) and provide feedback (e.g., praise, token economies) for positive behavior (Lewis et al., 2004; Simonsen et al., 2008). These practices are well researched, yet observational studies suggest teachers seldom enact them in the service of students with EBD (e.g., Maggin et al., 2011; McKenna & Ciullo, 2016). Even when placed in self-contained schools and classes—settings intended to provide effective, intensive services to students with the most significant behavioral needs (Rozalski et al., 2010)—researchers consistently find students experience limited opportunities to respond, academic feedback, and praise (Levy & Vaughn, 2002; McKenna & Ciullo, 2016), marking such experiences as potential mechanisms through which poor outcomes persist.

Working conditions may be implicated in these problems (Bettini et al., 2017; Leko et al., 2018). Working conditions are the contexts of teachers’ work, including the demands placed on them and the social and logistical resources they have for meeting those demands (Bettini et al., 2016; Bettini, Cumming et al., 2020; O’Brien et al., 2019). A growing body of research in educational leadership and policy indicates that working conditions may be powerfully related to the quality of instruction teachers provide (e.g., Johnson et al., 2012; Ronfeldt et al., 2015). For example, Johnson et al. found teachers’ ratings of administrative support and school culture explained a significant portion of students’ learning gains (measured using value added scores) over a year. Other studies have found a variety of working conditions (e.g., curricular resources, collegial interactions) contribute to improvements in teachers’ effectiveness over time (e.g., Jackson & Bruegmann, 2009; Kraft & Papay, 2014; Ronfeldt et al., 2015; Sun et al., 2017).

These studies indicate that leaders could leverage working conditions as a means to improve services for students with EBD; however, to date, no studies have examined how working conditions relate to special education teachers’ (SETs) provision of effective instructional or behavior management practices for students with EBD in self-contained settings (Bettini et al., 2017). In the United States, approximately 31.2% of K–12 students with EBD receive instruction in self-contained settings for >60% of the day (Office of Special Education and Rehabilitative Services, 2018). Given that students in these settings have the most significant behavioral needs and are the most at-risk population, necessitating best practices and intensive services from SETs, the lack of extant working conditions research in self-contained settings for students with EBD is concerning. Furthermore, understanding how working conditions relate to the use of effective practices in these settings is particularly important, given students’ high risk for persistent, negative long-term outcomes (e.g., unemployment, contact with the justice system; Wagner, 2014). Thus, our study aims to better understand the role of SETs’ working conditions in self-contained settings for students with EBD in their self-reported use of effective instructional and behavior management practices.

Theoretical Foundation

Our study is conceptually based in conservation of resources (COR) theory (Alarcon, 2011; Hobföll, 1989), which posits that individuals aim to balance resources and demands in their work environment. Specifically, individuals strive to build and maintain resources—including social supports (e.g., collegiality) and logistical supports (e.g., time, materials; Halbesleben et al., 2014)—to meet the demands of their job. When resources are insufficient to meet demands, employees experience persistent stress, resulting in emotional exhaustion (a component of burnout; Brunsting et al., 2014), reduced self-efficacy, and reduced investment in fulfilling job responsibilities effectively (Alarcon, 2011). Thus, when SETs experience high demands (e.g., more lessons to plan, more students to teach) and low resources (e.g., weak administrative support, insufficient planning time, and inadequate instructional resources), they may feel that they cannot manage the demands of their job, experience increased stress and emotional exhaustion, and reduced self-efficacy (Bettini et al., 2018; Bettini, Cumming et al., 2020). The potential result is weaker job performance, as evidenced by less consistent use of effective academic and behavior management practices. Researchers have begun using COR theory as a theoretical foundation to examine SETs’ working conditions (e.g., Bettini et al., 2018; O’Brien et al., 2019), yet no studies to date have used COR theory to explain variability in SETs’ self-efficacy or use of instructional practices.

Working Conditions

We focus on working conditions that prior studies indicate may shape SETs’ instruction: (a) demands (i.e., instructional groups, instructional responsibilities), (b) social resources (i.e., administrative support, school culture, and paraprofessional support), and (c) logistical resources (i.e., instructional resources, planning time; Bettini et al., 2016).

Demands

Demands are SETs’ work responsibilities, including their instructional groups (e.g., class size, student needs) and instructional responsibilities (e.g., number of grade levels and subjects to teach). In a systematic review of research on how working conditions relate to SETs’ instruction, Bettini et al. (2016) concluded that the strongest evidence base relates to the size and heterogeneity of the instructional groups SETs are assigned to teach; when assigned smaller instructional groups in which students have similar learning needs, SETs demonstrate better instructional skills and their students experience stronger achievement gains (e.g., Bishop et al., 2010; Russ et al., 2001; Wanzek & Vaughn, 2007). By contrast, no research has examined how instructional responsibilities (i.e., number of grades and subjects to plan and teach) relate to instructional quality or effectiveness (Bettini et al., 2016). Yet, SETs report that planning and delivering instruction across subjects and grades poses substantial challenges, as it requires them to be knowledgeable about many standards and curricula, and also to manage the logistical complexities of simultaneously planning and providing instruction on different content (Bettini et al., 2019). Furthermore, in a recent study (Bettini, Cumming et al., 2020), we found that SETs in self-contained settings for students with EBD, who are responsible for more subjects and grades, rate their planning time as less adequate, report their workloads are less manageable, experience greater emotional exhaustion, and are consequently more likely to intend to leave. More research is needed to understand how these demands relate to their use of effective practices.

Social resources

The social context of SETs’ work (i.e., administrative support, school culture, and paraprofessional support) likely plays an important role in supporting the use of instructional and behavior management practices (Bettini et al., 2016). Schools are social environments, in which teachers develop knowledge of and capacity to use effective practices through collaboration (Grossman & Thompson, 2004; Youngs et al., 2012). Furthermore, administrators support teachers’ knowledge and skills by fostering a school culture that supports teacher learning and by providing conditions (e.g., time and materials) necessary to enact effective practices (Bettini et al., 2016; Billingsley et al., 2017).

Despite a strong research base with general educators (e.g., Ronfeldt et al., 2015), Bettini et al.’s (2016) literature review found that studies examining how social resources relate to SETs’ use of effective practices is exceptionally limited. Yet, the handful of existing qualitative and mixed-methods studies consistently find that schools and districts that promote high achievement for students with disabilities have cultures of collaboration and shared responsibility for students (e.g., McLeskey et al., 2014). In addition, although researchers have not examined how SETs’ social resources in self-contained settings for students with EBD relate to their use of instruction and behavior management practices (Bettini et al., 2017), studies have shown that these resources matter for these teachers’ stress and emotional exhaustion (Bettini, Cumming et al., 2020). Other research has demonstrated that burnout is associated with weaker student gains on Individualized Education Program (IEP) goals, decreased instructional quality, and lower student engagement (Ruble & McGrew, 2013; Wong et al., 2017). Thus, by decreasing burnout, social resources could plausibly relate to increased use of effective instruction and behavior management practices.

Logistical resources

Logistical resources are practical supports that teachers need to fulfill their responsibilities, including curricular resources and planning time (O’Brien et al., 2019). A strong body of research with general educators, including studies permitting causal inferences, indicates curricular resources can powerfully shape the quality and effectiveness of teachers’ instructional practices (e.g., Jackson & Makarin, 2018; Jimenez et al., 2014). A small body of qualitative and mixed-methods research with SETs affirms studies of general educators, suggesting SETs use stronger instructional practices when they have structured curricula that support them in planning and providing instruction relevant to their students’ needs (Bishop et al., 2010; Brownell et al., 2014; Siuty et al., 2018). In contrast, teachers without strong curricula tend to use an ad hoc array of resources to plan and provide instruction, resulting in less relevant, lower quality instruction (Siuty et al., 2018).

Less research has examined planning time, but extant studies indicate time to plan for instruction is essential (Bettini et al., 2016). For example, Allinder (1996) found that SETs’ perceptions of adequacy of planning time were related to their use of effective practices learned in professional development, whereas Vannest et al. (2010) found that when SETs had insufficient planning time, they used instructional time to complete other responsibilities (e.g., paperwork), thereby reducing students’ instructional time. Furthermore, studies have found that ratings of adequacy of planning time are related to emotional exhaustion (Albrecht et al., 2009; Bettini, Cumming et al., 2020), which is, in turn, associated with the use of effective instructional practices (Ruble & McGrew, 2013; Wong et al., 2017). Thus, by reducing available instructional time and contributing to emotional exhaustion, planning time may be related to SETs’ use of effective practices.

Affective Outcomes

COR theory posits that demands and resources shape employees’ feelings and attitudes about their work (i.e., affective outcomes). For example, the degree to which employees feel that demands can be completed within the time allotted (i.e., workload manageability) determines the extent to which they feel stress, burnout, and self-efficacy for their work (e.g., Alarcon, 2011; Halbesleben et al., 2014). Based on a growing body of educational research, we focus on specific affective outcomes that likely play a role in SETs’ use of instructional and behavior management practices—workload manageability, emotional exhaustion, stress, and self-efficacy.

Workload manageability

Consistent with Bettini et al. (2018), we define workload manageability (sometimes referred to as role overload) as “teachers’ subjective perceptions of the degree to which responsibilities can be completed adequately within time allotted” (p. 113). An emerging body of research indicates that SETs’ perceptions of their workload manageability serves to mediate relationships between their working conditions (i.e., demands and resources) and perceived emotional exhaustion (a component of burnout) and stress (Bettini et al., 2018; Bettini, Cumming et al., 2020).

Emotional exhaustion and stress

In response to persistent stress (Brunsting et al., 2014), emotional exhaustion is when “emotional resources are depleted” and employees “feel they are no longer able to give of themselves” emotionally (Maslach & Jackson, 1981, p. 99). Prior studies indicate that SETs who perceive their workloads as less manageable experience higher stress and emotional exhaustion (Bettini et al., 2018; Bettini, Cumming et al., 2020), which likely relate to the quality of their work and intent to stay in the field. Indeed, researchers have found that the relationships between working conditions, workload manageability, and emotional exhaustion and stress serve to predict SETs’ intent to stay in the profession (Bettini, Cumming et al., 2020). Additional insight is needed into how these relationships shape SETs’ use of effective practices in self-contained settings for students with EBD.

Self-efficacy

Tschannen-Moran and Hoy (2001) defined self-efficacy as “a [teacher’s] judgment of his or her capabilities to bring about desired outcomes of student engagement and learning” (p. 783), which can vary by task domain (e.g., instruction, classroom management). Because employees feel less efficacious when demands are greater and resources are weaker (e.g., Benight et al., 1999; Chen et al., 2009), self-efficacy is likely sensitive to changes in SETs’ perceptions of workload manageability. In addition, a robust body of research indicates that, when individuals feel more efficacious, they are more likely to perform well in a task (e.g., Stajkovic et al., 2018). Thus, self-efficacy is a potential mediator between working conditions and job performance. Indeed, in a literature review of research with teachers, Zee and Koomen (2016) found evidence that highly efficacious educators tend to use more effective instructional and behavior management practices, and that teacher self-efficacy is associated with students’ academic adjustment. Although studies on SETs’ self-efficacy are more limited, researchers have shown that SETs’ self-efficacy is related to burnout (e.g., Ruble et al., 2011), which in turn is associated with decreased instructional quality (Ruble & McGrew, 2013; Wong et al., 2017). Given that SETs’ self-efficacy may be linked to both their working conditions and their use of effective practices, it is important to include self-efficacy when examining the relationships among working conditions and effective use of instructional and behavior management practices in self-contained settings for students with EBD.

Purpose of the Study

In this study, we aim to better understand how SETs’ working conditions relate to their workload manageability, emotional exhaustion, stress, self-efficacy, and use of effective instructional and behavior management practices. We focus on working conditions experienced by SETs in self-contained settings for students with EBD, given that these settings are designed for those students with the most significant needs and intended to provide the most intensive and effective services to meet these needs. We hypothesized as follows:

Hypothesis 1: Workload manageability, emotional exhaustion, and stress will mediate relationships between working conditions and self-efficacy; SETs who report stronger resources and fewer demands will perceive workloads as more manageable and experience less emotional exhaustion and stress, and will thus feel more efficacious for instruction and for behavior management.

Hypothesis 2: Self-efficacy will mediate relationships between working conditions, workload manageability, stress, emotional exhaustion, and SETs’ self-reported use of instructional and behavior management practices; SETs who report fewer demands and more resources will rate workloads as more manageable, experience less emotional exhaustion and stress, greater self-efficacy, and use effective instructional and behavior management practices more often.

Method

Sampling

Our current sample is drawn from a national survey study of SETs serving students with EBD in self-contained settings (Bettini, Cumming et al., 2020; O’Brien et al., 2019). To obtain a national sample, we built a sampling frame from the U.S. Department of Education’s list of all school districts, inclusive of all district types (e.g., public, charter). We stratified districts by student population to ensure representation of districts of various sizes, based on cut points used in the National Center for Education Statistics’ national surveys: very large (>33,552 students), large (10,474–33,552), medium (3,523–10,473), and small (25–3,522). We randomly selected 25 districts from each stratum (100 total). We recruited district special education administrators to provide contact information for relevant SETs in their districts. We replaced districts who declined or had no eligible participants with the next district in the same stratum, to reach 25 districts per stratum. District recruitment lasted from Summer 2017 to Spring 2018.

Of the 224 districts we attempted to recruit, 44 (19.64%) agreed to participate, but three violated study protocols, resulting in a sample of 41 districts (18.14%): 12 very large (29.27%), 10 large (24.39%), nine medium (21.95%), and 10 small (24.39%). Districts represented all U.S. regions: Northeast (seven, 17.07%), Midwest (five, 12.20%), South (17, 41.86%), and West (12, 29.27%). We compared demographic indicators for all participating and nonparticipating districts; with the exception of serving significantly fewer American Indian students (see Supplemental Table S1), districts were comparable. Participating districts’ enrollment (7,079) was larger than declining districts (2,648), yet the difference was statistically insignificant.

Survey Administration Procedures

We received approval for the study from the institutional review boards of Florida International University, George Mason University, Wake Forest University, and Boston University. District administrators provided contact information for potential SETs. Before survey administration, district administrators emailed eligible SETs to inform them of the district’s approval for the study. We administered the survey (see supplemental Table S2) at two time points: (a) Fall 2017 with districts who had agreed to participate at that time (n = 20), and (b) Spring 2018 with districts recruited after the first administration (n = 21). Thus, SETs completed the survey at only one time point. We directly emailed SETs a link to the Qualtrics survey, which included the consent form, and embedded skip logic to eliminate participants who were not members of the target population. Following Dillman’s (2007) recommendations to increase response rates, we sent a reminder email after 1 week to SETs who had not yet responded. We then mailed a paper survey, a self-addressed and stamped envelope, and US$2 to the remaining nonresponders. We sent the survey to 459 SETs, of whom 235 (51.20%) responded. Of these, 171 met inclusion criteria (i.e., teaching students with EBD in a self-contained setting). Of the 171 SETs, 72.73% had certification in education and 63.64% had certification in special education. On average, SETs had 12.74 years of teaching experience and were employed at their current school for 5.45 years. See O’Brien et al. (2019) for additional demographics.

Instrumentation

The survey included items measuring working conditions (i.e., social resources, logistical resources, and demands), affective outcomes (workload manageability, emotional exhaustion, stress, and self-efficacy), and self-reported use of effective instructional and behavior management practices. Survey development included an iterative revision in response to an expert review panel and a series of cognitive interviews with members of the target population to refine items; it is described in more detail in other articles (Bettini, Cumming et al., 2020; O’Brien et al., 2019).

Except for observed indicators (lessons to plan, hours planning outside the school day, number of students, and number of paraprofessionals), we coded items so higher scores correspond to positive experiences (e.g., a higher planning time score indicates an SET reported stronger planning time, while a higher emotional exhaustion score signifies an SET reported less emotional exhaustion). Except for self-efficacy and use of instructional and behavior management practices, we tested all scales in a prior study, using confirmatory factor analysis (CFA); they demonstrated good model fit and strong item loadings (see Supplemental Table S3; Bettini, Cumming et al., 2020).

We used CFA to investigate scales that were not part of our prior analyses (see Table 1): self-efficacy for instruction, self-efficacy for behavior, reported use of effective instructional practices, and reported use of effective behavior management practices. Using Mplus (Muthén & Muthén, 2018), we set factors’ variance to 1 to address scale indeterminacy. We evaluated goodness of fit using nonsignificant chi-square, root mean square error of approximation (RMSEA; <.10), Bentler’s comparative fit index (CFI; >.90), and Tucker–Lewis index (TLI; >.90; Kline, 2016).

Table 1.

Confirmatory Factor Analysis Results for Measurement Models.

Variable	Model fit	Items	Loading
Self-efficacy for instruction	χ² = .037, p = .98; RMSEA = 0.00; CFI = 1.00; TLI = 1.06	I have opportunities to provide the kind of instruction I want to provide.	.475***
		Use a variety of assessment strategies.	.738***
		Provide an alternative explanation or example when students are confused.	.670***
		Craft good questions for your students.	.626***
Self-efficacy for behavior	χ² = 4.00, p = .41; RMSEA = 0.00; CFI = 1.00; TLI = 1.00	Promote positive social interactions among students.	.815***
		Promote students’ social-emotional skills.	.893***
		Promote a positive classroom culture. Develop effective behavior plans. Implement effective behavior plans.	.809* .663* .689***
Instructional practices	χ² = 8.67; p = .12; RMSEA = 0.07; CFI = 0.98; TLI = 0.95	I provide a high rate of opportunities to respond to during my instruction.	.686***
		I engage my students in observable ways during teacher directed instruction (i.e., using response cards, choral responding, and other methods).	.832***
		I use evidence-based methods (i.e., methods shown through research to be effective) to deliver my instruction I praise students for academic behaviors (e.g., on task behaviors). I correct students’ academic errors.	.711* .493* .318***
Behavior management practices	χ² = 0.01; p = .93; RMSEA = 0.00; CFI = 1.00; TLI =1.07	I praise students for social behaviors (e.g., working well with peers).	.663***
		I use other systems to acknowledge appropriate behavior (group contingencies, behavior contracts, or token economies).	.678***
		I correct students’ social errors	.479***
		I correct students’ behavioral errors.	.360***

Note. Loadings are standardized. All items are coded so higher numbers indicate more positive experiences. RMSEA = root mean square error of approximation; CFI = comparative fit index; TLI = Tucker–Lewis index.

***p < .001.

Self-efficacy

We used eight items adapted from Tschannen-Moran and Hoy’s (2001) well-validated Teachers’ Self-Efficacy Scale (TSES). Expert reviewers expressed concern that these items would take too much cognitive energy, within the extensive survey, because they did not share a common stem (e.g., “How well can you . . . To what extent can you . . .”); thus, we created a stem (i.e., “Please indicate the extent to which you feel you are able to do the following . . .”) and adapted existing items to fit the stem (e.g., “Craft good questions for your students”). Based on feedback from an expert panel, we also added five items to reflect unique aspects of SETs’ self-efficacy for behavior management in self-contained classes for students with EBD (e.g., “Develop effective behavior plans”), for a total of 12 items.

Consistent with their factor structure in other studies (e.g., Klassen et al., 2009; Tschannen-Moran & Hoy, 2001), we posited these items would load onto three factors: self-efficacy for behavior management, self-efficacy for instruction, and self-efficacy for engagement. For the purposes of the present analysis, we excluded self-efficacy for engagement.

Self-efficacy for instruction

Initially made up of three items, we added one item about use of effective instructional practices (“I have opportunities to provide the kind of instruction I want to provide”), which was phrased to better capture SETs’ belief in their capacity. The model fit the data exactly, χ²(2) = 0.04, p = .98, with most items demonstrating strong loadings.

Self-efficacy for behavior management

The five-item behavior management scale initially did not fit the data well. Based on modification indices and theoretical overlap, we allowed “Develop effective behavior plans” and “Implement effective behavior plans” to correlate. The model fit the data exactly, χ²(4) = 4.00, p = .41, with item loadings above .60.

Effective practices

We used Simonsen et al.’s (2008) systematic review of the literature on evidence-based practices and their proposed self-assessment tool, as well as Lewis et al.’s (2004) review on research-supported practices for students with EBD to develop 21 items related to SETs’ reported use of effective practices. Of the 21 items, six captured practices for academic instruction and four were related to addressing student behavior. The remaining items assessed other classroom practices (e.g., communicating expectations). In the present analysis, we included items related to effective instructional and behavior management practices only.

Effective instructional practices

We used five items to capture the extent to which SETs reported using effective instructional practices. The scale demonstrated good model fit, χ²(5) = 8.67, p = .12, with all items loading above .30.

Effective behavior management practices

The four-item behavior management scale fit the data exactly, χ²(1) = 0.01, p = .93, with items demonstrating adequate loadings. Although the model fit exactly, we found very low item variance; the vast majority of SETs rated themselves so highly on this scale that variability in their ratings of their behavior management practices could not be detected. This led us to drop the scale from the study.

Analysis

To determine the pattern of missing data, we conducted Little’s Missing Completely at Random (MCAR) test (Little, 1988) and separate variance t tests using Missing Value Analysis with SPSS 25. Overall, 7.1% of the data were missing from the data set, with the majority of missing data related to SETs’ reported years of teaching (19.3%), the number of paraprofessionals in their class (14.6%), their race (12.6%), and gender (11.1%). A nonsignificant Little’s MCAR test, χ²(2179) = 2169.67, p =.552, revealed that the data were MCAR. However, significant t-test results indicated data were likely missing at random (MAR). Thus, we used structural equation modeling (SEM) to examine hypothesized relationships, using full information maximum likelihood estimation with robust standard errors (estimator = MLR) to address non-normality and missing data. Mplus provides maximum likelihood estimation under MCAR and MAR for continuous and categorical variables, and combinations of such variables (Muthén & Muthén, 2018).

To determine the potential effect of clustering (i.e., teachers within districts), we calculated the intraclass correlation coefficient (ICC) of district on the outcome of interest, namely, instructional practices. The ICC was 2.11%, well below the 5% to 10% point at which multilevel modeling is recommended to account for clustering (LeBreton & Senter, 2008). Furthermore, because the majority of the participating schools had fewer than three participating teachers, clustering at the school level was not advisable. Thus, we proceeded without clustered analyses.

The sample size (N = 171) was too small to test all relationships simultaneously in a single model in SEM (Kline, 2016). Thus, we first analyzed data using bootstrapping (1,000) that offers nonsymmetrical confidence intervals (CIs), which is particularly relevant for small samples (Nevitt & Hancock, 2001). Yet, we found poor model fit with the full model. Thus, as done in previous research (Bettini, Cumming et al., 2020) with small sample sizes, we used an iterative process using a series of structural equation models to test complex relationships among variables. Building on prior analysis of the same data set in which we investigated how working conditions related to workload manageability, emotional exhaustion and stress, and intent to leave (Bettini, Cumming et al., 2020), we followed an iterative process to test hypothesized relationships with self-efficacy and use of instructional and behavioral management practices (see Supplemental Figures S1–S6, and Figures 1 -3). We chose to do separate analyses of the current data set for several reasons. Although we anticipated that similar relationships would be discovered between working conditions and teachers’ reported affective outcomes, we expected that specific working conditions (i.e., demands, logistical resources, and social resources) would have unique effects on teachers’ reported use of behavioral and instructional practices in their classrooms that would vary from those related to their intent to stay in teaching; such findings would have important implications for how specific working conditions could be leveraged to improve teachers’ practice use with their students with EBD in self-contained settings.

Figure 1.

Full model examining the relationships between working conditions, affective outcomes, and use of instructional practices.

Figure 2.

Planning time as a mediator between resources, demands, and workload manageability.

Figure 3.

Planning time as a mediator, without social resources, post hoc model.

First, we examined hypothesized relationships among workload manageability, emotional exhaustion, stress, and self-efficacy for behavior and instruction in two separate models (Model 1a and 1b; see Supplemental Figures S1 and S2). Next, we added instructional practices to examine hypothesized direct and indirect relationships (Model 2; see Supplemental Figure S3). In the remaining models, we added each set of working conditions to examine hypotheses related to social resources (Model 3; see Supplemental Figure S4), logistical resources (Model 4; see Supplemental Figure S5), and demands (Model 5; see Supplemental Figure S6). For the full model, we combined all significant pathways from Models 1 through 5 and added SET experience as a control (Figure 1). In a previous study on how working conditions relate to intent to leave (Bettini, Cumming et al., 2020), we found planning time mediated relationships among other working conditions and workload manageability. Thus, we added planning time as a mediator (see Figure 2). Based on findings, we tested a final post hoc model without social resources (see Figure 3).

Results

Structural Equation Models

Iterative model development

In Model 1a and 1b (see Supplemental Figures S1 and S2), we tested whether workload manageability predicted emotional exhaustion and stress, and whether these mediated relationships between workload manageability and SETs’ self-efficacy for behavioral and instructional practices. Model 1a, χ²(146) = 190.13, p = .008; RMSEA = 0.04; CI = [0.02, 0.06]; CFI = 0.97; TLI = 0.97, and 1b, χ²(113) = 149.64, p = .012; RMSEA = 0.04; CI = [0.02, 0.06]; CFI = 0.97; TLI = 0.96, both fit closely. Unexpectedly, in Model 1a, emotional exhaustion and stress had nonsignificant relationships with self-efficacy for behavioral practices. We found SETs reported very strong self-efficacy for behavioral practices, resulting in low variance and nonsignificant findings. As such, we did not test additional models with behavioral self-efficacy. However, in Model 1b, we found emotional exhaustion predicted self-efficacy for instruction, and stress indirectly predicted self-efficacy for instruction through emotional exhaustion; SETs who reported less emotional exhaustion indicated greater self-efficacy for instruction. Thus, all remaining models include self-efficacy for instruction. With the addition of instructional practices to the model (Model 2; see Supplemental Figure S3), we found the model fit closely, χ²(203) = 301.97, p = .000; RMSEA = 0.06, CI = [0.04, 0.07]; CFI = 0.94; TLI = 0.93, with relationships in the expected directions. In Model 3 (see Supplemental Figure S4), in which we added social resources to the model, we found the model fit the data poorly, χ²(1103) = 1685.17, p = .000; RMSEA = 0.06, CI = [0.05, 0.07]; CFI = 0.86; TLI = 0.85. As expected, administrative support, paraprofessional training, and paraprofessional trust positively and directly predicted workload manageability. Yet, as in a prior analysis with intent to leave as a terminal outcome (Bettini, Cumming et al., 2020), several insignificant paths (collective responsibility, collaborative culture, and paraprofessional trust) and a negative relationship between workload manageability and number of paraprofessionals were found. In Model 4 (see Supplemental Figure S5), we tested logistical resources, and found acceptable model fit, χ²(394) =594.54, p = .000; RMSEA = 0.06, CI = [0.05, 0.06]; CFI = 0.91; TLI = 0.90, with findings in expected directions. In Model 5 (see Supplemental Figure S6), we examined demands and found adequate model fit, χ²(292) = 446.01, p =.000; RMSEA = 0.06, CI = [0.05, 0.07]; CFI = 0.91; TLI = 0.90. As in a previous analysis with intent to leave as a terminal outcome (Bettini, Cumming et al., 2020), the number of lessons SETs needed to plan negatively predicted workload manageability, whereas having instructional groups in which students had similar needs directly predicted more manageable workloads. Neither lessons to plan nor instructional grouping had direct relationships with self-efficacy for instruction and instructional practices, yet lessons to plan did demonstrate indirect effects through workload manageability and emotional exhaustion and stress. In the full model (see Figure 1), we tested all significant paths from Models 1 to 5, adding years of experience to the model. The final model fit poorly, χ²(1063) = 1659.21, p = .000; RMSEA = 0.06, CI = [0.06, 0.07]; CFI = 0.82; TLI = 0.81, and several previously significant pathways no longer predicted workload manageability; lessons to plan, curricular resources, instructional grouping, paraprofessional training, and administrative support were all unrelated to workload manageability in this model.

Mediation and post hoc models

Based on results of a prior analysis with intent to leave as the outcome (Bettini, Cumming et al., 2020), we added planning time as a mediator of relationships between other working conditions and intent. Model fit slightly improved, χ²(1061) = 1610.73, p = .000; RMSEA = 0.06, CI = [0.05, 0.07]; CFI = 0.84; TLI = 0.83, yet still fit poorly (see Figure 2). Because Model 3, which included social resources, was the weakest, we tested one additional post hoc model excluding social resources (see Figure 3). Model fit greatly improved, χ²(514) = 752.65, p =.000; RMSEA = 0.05, CI = [0.04, 0.06]; CFI = 0.90; TLI = 0.89. This model explained 45% of variance in planning time, 22% of variance in curricular resources, 53% of variance in workload manageability, 64% of variance in emotional exhaustion, 25% of variance in stress, 7% of variance in self-efficacy for instruction, and 48% of variance in use of instructional practices. As expected, workload manageability predicted SETs’ emotional exhaustion (β .62, p < .001) and stress (β .51, p < .001), such that SETs who reported more manageable workloads were more likely to indicate being less emotionally exhausted and stressed. Emotional exhaustion directly predicted self-efficacy for instruction (β .26, p = .005); SETs who reported less emotional exhaustion indicated greater self-efficacy for instruction. In turn, self-efficacy positively predicted SETs’ reported use of effective instructional practices (β .69, p < .001); SETs with stronger self-efficacy for instruction were more likely to report using effective instructional practices. Stress also had a direct positive relationship with emotional exhaustion (β .28, p = .005), which in turn predicted SETs’ self-efficacy for instruction, mediated through emotional exhaustion. As anticipated, workload manageability was indirectly related to SETs’ reported use of instructional practices, mediated by emotional exhaustion and self-efficacy for instruction (β .17, p = .006) and, to a lesser extent, through stress (β .04, p = .039).

Some logistical resources and demands had direct relationships with workload manageability. Curricular resources (β .24, p = .003) and planning time (β .54, p <.001) positively predicted workload manageability; SETs who indicated that they had access to curricular resources and sufficient planning time were more likely to rate their workload as manageable. Conversely, hours spent planning had a direct negative effect on workload manageability (β –.164, p = .001), such that SETs who reported spending more hours outside the school day planning were more likely to find their workload less manageable. As hypothesized, workload manageability mediated indirect relationships between working conditions (curricular resources, β .05, p = .051; hours spent planning, β –.04, p = .018; instructional grouping, β .06, p = .020) and use of effective instructional practices through emotional exhaustion, stress, and self-efficacy for instruction. Unexpectedly, instructional grouping and number of lessons to plan did not predict workload manageability in the model.

In our post hoc model (see Figure 3), we had hypothesized that planning time would mediate relationships between other working conditions and workload manageability. First, we found instructional grouping (β .22, p <.001) and access to curricular resources (β .50, p = .019) positively predicted planning time, such that SETs who indicated that they taught more homogenous groups of students and had access to curricular resources were more likely to indicate they had sufficient time to plan. As with workload manageability, hours spent planning outside the contractual school day had a direct and negative effect on planning time (β –.25, p =.011) and number of lessons to plan did not predict planning time. Curricular resources (β .12, p = .014) and hours planning (β –.13, p = .017) both indirectly predicted workload manageability through planning time. In turn, planning time indirectly predicted self-efficacy and reported use of instructional practices through workload manageability and emotional exhaustion and stress (β .08; p = .015).

Discussion

A small but growing body of research has highlighted the importance of working conditions in supporting SETs serving students with EBD in self-contained settings (Bettini et al., 2017). Although previous studies have shown links between working conditions and teachers’ self-efficacy (e.g., Kiely et al., 2014; Ross et al., 2012), as well as between self-efficacy and teachers’ use of effective practices (e.g., Zee & Koomen, 2016), none have examined the complex relationships among these variables and emotional exhaustion and stress, and none have focused on SETs serving students with EBD in self-contained settings. Thus, the purpose of our study was to investigate how working conditions related to SETs’ affective outcomes (workload manageability, emotional exhaustion, stress, and self-efficacy) and reported use of effective instructional and behavior management practices. In partial support of our hypotheses, our study is the first to find that SETs who experienced more supportive working conditions (i.e., stronger logistical resources, lower demands) rated their workloads as more manageable, experienced less emotional exhaustion and stress, and thus felt more efficacious in using effective instructional practices; furthermore, these factors mediated relationships between their working conditions and their reported use of effective instructional practices. However, we were unable to detect predictors of self-efficacy for reported use of behavior management practices, and social resources did not demonstrate good model fit when included. In the ensuing sections, we discuss results, describe limitations, and highlight implications for research and practice.

Role of Working Conditions and Affective Outcomes in Self-Efficacy and Instructional Practice Use

We found certain working conditions and specific affective outcomes (workload manageability, emotional exhaustion, and stress) played important roles in SETs’ self-efficacy for instruction and reported use of effective instructional practices. Specifically, access to curricular resources, adequate planning time, and hours spent planning outside of the school day indirectly predicted SETs’ reported use of effective instructional practices, mediated through self-efficacy for instruction. Working conditions have been shown previously to predict SETs’ intent to stay in self-contained settings for students with EBD (Albrecht et al., 2009; Bettini, Cumming et al., 2020), but ours is the first study to demonstrate relationships between these working conditions and self-efficacy. Prior research has also demonstrated a relationship between self-efficacy and other important outcomes (e.g., burnout; Ruble et al., 2011), which aligns with our findings related to workload manageability, emotional exhaustion, and stress as significant mediators. Thus, for SETs in self-contained settings for students with EBD, providing access to adequate curricular resources, protecting planning time (especially for SETs with less homogeneous instructional groups), and reducing the hours they plan outside of the school day are ways school leaders may promote instructional self-efficacy and use of effective instructional practices.

Contrary to our hypotheses, instructional grouping and number of lessons to plan, which we conceptualized as demands, did not directly predict workload manageability. Instructional grouping, however, did have an indirect relationship with workload manageability through planning time; SETs who had more homogeneous instructional groups tended to view planning time as sufficient, their workload as more manageable, and reported greater self-efficacy for instruction. Thus, logistical resources may be more important than certain demands for these SETs’ instructional self-efficacy (Bettini, Cumming et al., 2020).

Logistical resources may be more important than social resources as, contrary to our expectations, the weakest tested models included social resources, which had to be dropped from our final model. Prior studies have found that social resources can predict teacher self-efficacy. For example, Goddard and Goddard (2001) found teachers had stronger self-efficacy in schools characterized by stronger collective efficacy (i.e., a shared belief that all teachers in a school can positively influence student outcomes). For novice teachers, sources of social support, such as collegial and administrative support, seem to be especially important, and have been found to predict self-efficacy (e.g., Tschannen-Moran & Hoy, 2007). Thus, insignificant results could indicate this relationship is different for SETs in self-contained settings. Indeed, prior studies have consistently found these teachers are often quite isolated, engaging in very limited instructional interactions with colleagues and administrators; instead, their social supports tend to be related to managing challenging student behavior (O’Brien et al., 2019). For example, in a qualitative analysis of SETs’ experiences teaching in these settings, Bettini et al. (2019) found that although teachers generally felt a sense of goodwill from their colleagues and administrators, this goodwill did not translate into meaningful, active support for their instruction. Rather, SETs described having only superficial instructional interactions with colleagues, while administrators’ support largely focused on their students’ behavior. In addition, these SETs often are the only teachers in their school serving students with EBD (O’Brien et al., 2019); thus, it is possible that social supports, even when they are present, may not support self-efficacy for instruction or use of effective instructional practices. In addition, it is possible that social resources were less important for instructional self-efficacy and practice use in this sample, given that the average amount of teaching experience was almost 13 years.

Self-Efficacy and Use of Behavior Management Practices

Contrary to our expectations, SETs consistently reported very high self-efficacy for behavior management, resulting in little factor variance. Similarly, their self-reported use of effective behavior management practices was so uniformly high that we were unable to test models predicting this as an outcome. These findings are in direct contradiction to ample observational research indicating that effective behavior management practices are observed at low frequency in these settings (e.g., Levy & Vaughn, 2002; McKenna & Ciullo, 2016). There are several possible explanations for this discrepancy. First, it is possible that SETs have little foundation on which to form accurate self-evaluations of their use of these practices (e.g., limited feedback, limited opportunities to observe others teaching students with EBD) and thus overestimate the extent to which they use these practices; another possibility is that they are using these practices at the rates reported, but with poor quality and lack of fidelity. Finally, it is possible that our participants, who reported an average of almost 13 years of teaching, had accurate self-evaluations, and were generally stronger at using effective behavior management practices than teachers in prior studies.

Of note, SETs’ self-efficacy for instruction and reported use of effective instructional practices showed substantially more variability. There are several possible explanations for why this differed from their self-reported behavior management practices. First, this could potentially indicate that SETs have access to more information about the quality of their instructional practices, leading to more variability in their self-appraisal. Another possibility is that, because these items are self-reported, SETs of students with challenging behavior may have stronger social desirability bias when reporting their behavior management practices as they may view these practices as more central to their roles. Alternatively, responses could reflect stronger preparation and support for managing behavior, as SETs in these settings tend to report feeling more supported to manage student behavior than to provide strong instruction (e.g., Bettini et al., 2019).

Limitations

First, we used SETs’ self-reported use of instructional and behavior management practices as our dependent variable, which has potential for self-report bias. We chose self-report due to greater feasibility, in a national study, than observation. Prior research has indicated self-reported use of effective practices does (a) provide some indication of teachers’ actual practice use, and (b) correlate with observational data (Porter et al., 2007). As such, self-report has been used in seminal research on how professional development shapes instruction (e.g., Desimone et al., 2002; Garet et al., 2001; Leko et al., 2018). Although these items may be subject to social desirability bias, prior research indicates that, in low-stakes situations, self-report items that are nonjudgmentally phrased and which focus on behaviors as opposed to quality (as ours were) are less subject to social desirability bias (Porter et al., 2007). Yet a clear limitation is that, even if SETs provide honest assessments, those assessments are only as reliable as their capacity to effectively self-evaluate, which may vary based on their knowledge and skill. As such, future research using observations of instruction is needed to confirm findings and to determine how self-reported instructional practices relate to actual use of these practices among these SETs.

The sampling strategy introduces several other limitations. First, we surveyed SETs in self-contained settings for students with EBD, and results cannot be generalized to other populations. Second, regional variation in working conditions could bias findings, as some regions were better represented than others. Bias could also potentially be introduced by collecting data at two time points (i.e., October/November and February/March), as SETs’ perceptions of working conditions may change across the school year. Third, we did not control for which grade levels SETs taught, despite some evidence that controlling for elementary versus secondary may be important (Bettini, Gilmour et al., 2020). Fourth, our sample size was relatively small for SEM, and we may have lacked power to detect some relationships. However, given the number of relevant variables plus the sample size, alternative approaches (e.g., Bayesian) may have lacked power to detect the unique effects of types of working conditions (e.g., logistical resources, social resources)—using an iterative SEM process allowed us to investigate such effects, as well as to parse out which specific working conditions contributed to poor fit in the initial model. Thus, because we were unable to test a complete model in one structural equation, necessitated by sample size, we encourage researchers to extend our findings with a larger sample.

Finally, models that included social resources did not fit closely, and coefficients may not be trustworthy; those from Models 1, 2, 4, 5, and post hoc are trustworthy, as they fit closely or exactly, but as these excluded some factors, they may be susceptible to omitted variable bias.

Implications for Future Research

Despite limitations, this is the first study to provide insight into how working conditions relate to SETs’ self-efficacy and use of effective practices in self-contained settings for students with EBD, and it has important implications for research. We focused on self-contained settings as they are designed for students with the most significant needs; however, because the majority of students with disabilities are served in inclusive settings, additional research is needed to examine how working conditions shape SETs’ instruction in other educational settings (e.g., general education, resource) along the continuum of placements. It is possible that the working conditions most salient to SETs’ instruction may vary depending on their service delivery model, such that SETs in inclusive settings may depend more on social resources (e.g., general educators) than SETs in self-contained settings. Future research is also needed to confirm these findings with SETs serving other populations of students (e.g., learning disabilities, autism). Such research would benefit from using observational data on instructional practice and student achievement gains, as findings could indicate which working conditions shape instruction in sufficiently powerful ways to contribute to student outcomes. Given the limited prior research on working conditions and instructional practices (Bettini et al., 2016), our findings suggest that continuing to add to this growing literature base will be crucial for finding ways to improve instruction and outcomes for students with disabilities by leveraging working conditions.

Future studies should consider including a wider range of predictors in the model. We included relatively few personal characteristics in the present analysis (e.g., knowledge, beliefs about students, and preparation). Research incorporating these characteristics could provide important insights into which working conditions matter most for which populations of SETs. For example, it is possible that SETs with weaker preparation may need more curricular resources and planning time. Parsing these kinds of nuances could help leaders identify which supports to prioritize for which SETs. Furthermore, although this study provides initial validity evidence for our scales, scholars should consider further validating these scales by (a) examining additional psychometric properties (e.g., determining the extent to which reported curricular resources reflect provided curricular resources), and (b) validating with other populations of SETs (e.g., teaching students with learning disabilities in inclusive settings).

Implications for Policy and Practice

Our findings indicate that, when SETs feel their workloads are less manageable, they are less likely to report using effective instructional practices, which can have important implications for the instruction provided to students with EBD. Leaders should examine ways to better support SETs in self-contained settings for students with EBD, particularly by protecting their planning time and ensuring they have access to adequate curricular resources to meet their students’ learning needs. We also encourage school leaders to provide instructional supports (e.g., training in use of resources) to SETs in these settings, in addition to behavioral supports. Finally, leaders should consider how working conditions shape the work of their SETs serving students with EBD under other service delivery models (e.g., general education, resource). Given that students with EBD have both significant social-emotional and behavioral and academic challenges (Kauffman & Landrum, 2018), it will be crucial for leaders to examine the adequacy of supports provided to SETs to ensure that students with EBD are receiving targeted, evidence-supported practices for both their behavioral and instructional needs.

Conclusion

Students with EBD not only experience emotional and behavioral challenges, but also exhibit academic deficits. Effective SETs are key to addressing these needs, and our findings demonstrate that working conditions may contribute to their use of effective instructional practices, by predicting their self-efficacy, workload manageability, stress, and emotional exhaustion. It is crucial, therefore, that school leaders address SETs’ working conditions both to improve their experiences, and to ensure students with EBD receive the kinds of effective practices necessary to improve their outcomes.

Supplemental Material

Figure_S1 – Supplemental material for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders

Supplemental material, Figure_S1 for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders by Michelle M. Cumming, Kristen Merrill O’Brien, Nelson C. Brunsting and Elizabeth Bettini in Remedial and Special Education

Supplemental Material

Figure_S2 – Supplemental material for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders

Supplemental material, Figure_S2 for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders by Michelle M. Cumming, Kristen Merrill O’Brien, Nelson C. Brunsting and Elizabeth Bettini in Remedial and Special Education

Supplemental Material

Figure_S3 – Supplemental material for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders

Supplemental material, Figure_S3 for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders by Michelle M. Cumming, Kristen Merrill O’Brien, Nelson C. Brunsting and Elizabeth Bettini in Remedial and Special Education

Supplemental Material

Figure_S4 – Supplemental material for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders

Supplemental material, Figure_S4 for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders by Michelle M. Cumming, Kristen Merrill O’Brien, Nelson C. Brunsting and Elizabeth Bettini in Remedial and Special Education

Supplemental Material

Figure_S5 – Supplemental material for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders

Supplemental material, Figure_S5 for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders by Michelle M. Cumming, Kristen Merrill O’Brien, Nelson C. Brunsting and Elizabeth Bettini in Remedial and Special Education

Supplemental Material

Figure_S6 – Supplemental material for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders

Supplemental material, Figure_S6 for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders by Michelle M. Cumming, Kristen Merrill O’Brien, Nelson C. Brunsting and Elizabeth Bettini in Remedial and Special Education

Supplemental Material

Table_S1 – Supplemental material for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders

Supplemental material, Table_S1 for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders by Michelle M. Cumming, Kristen Merrill O’Brien, Nelson C. Brunsting and Elizabeth Bettini in Remedial and Special Education

Supplemental Material

Table_S2 – Supplemental material for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders

Supplemental material, Table_S2 for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders by Michelle M. Cumming, Kristen Merrill O’Brien, Nelson C. Brunsting and Elizabeth Bettini in Remedial and Special Education

Supplemental Material

Table_S3 – Supplemental material for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders

Supplemental material, Table_S3 for Special Educators’ Working Conditions, Self-Efficacy, and Practices Use With Students With Emotional/Behavioral Disorders by Michelle M. Cumming, Kristen Merrill O’Brien, Nelson C. Brunsting and Elizabeth Bettini in Remedial and Special Education

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by a grant (No. 201800060) from the Spencer Foundation.

ORCID iD

Michelle M. Cumming

Supplemental Material

Supplemental material for this article is available on the Remedial and Special Education website along with the online version of this article.

References

Alarcon

G. M.

(2011). A meta-analysis of burnout with job demands, resources, and attitudes. Journal of Vocational Behavior, 79, 549–562. https://doi.org/10.1016/j.jvb.2011.03.007

Albrecht

S. F.

Johns

B. H.

Mountstevens

Olorunda

(2009). Working conditions as risk or resiliency factors for teachers of students with emotional and behavioral disabilities. Psychology in the Schools, 10, 1006–1022. https://doi.org/10.1002/pits.20440

Allinder

R. M.

(1996). When some is not better than none: Effects of differential implementation of curriculum-based measurement. Exceptional Children, 62, 525–535. https://doi.org/10.1177/001440299606200604

Bettini

Cumming

O’Brien

K.M.

Brunsting

Ragunanthan

Sutton

Chopra

(2020). Predicting special educators’ intent to continue teaching students with emotional/behavioral disorders in self-contained classes. Exceptional Children, 86, 209–228. https://doi.org/10.1177/0014402919870464

Bettini

Gilmoure

A. F.

Williams

T.O.

Billingsley

B.S.

(2020). Predicting special and general eductors’ intent to continue teaching using conservation of resources theory. Exceptional Children, 86, 310–329. https://doi.org/10.1177/0014402919870464

Benight

C. C.

Swift

Sanger

Smith

Zeppelin

(1999). Coping self-efficacy as a mediator of distress following a natural disaster. Journal of Applied Social Psychology, 29, 2443–2464.

Bettini

E. A.

Crockett

Brownell

M. T.

Merrill

(2016). Relationships between working conditions and special educators’ instruction. The Journal of Special Education, 50, 178–190. https://doi.org/10.1177/0022466916644425

Bettini

E. A.

Cumming

M. M.

Merrill

K. L.

Brunsting

N. C.

Liaupsin

C. J.

(2017). Working conditions in self-contained settings for students with emotional disturbance. The Journal of Special Education, 51, 83–94. https://doi.org/10.1177/0022466916674195

Bettini

E. A.

Jones

N. D.

Brownell

M. T.

Conroy

M. A.

Leite

W. L.

(2018). Relationships between novice teachers’ social resources and workload manageability. The Journal of Special Education, 52, 113–126. https://doi.org/10.1177/0022466918775432

10.

Bettini

E. A.

Wang

Cumming

Kimerling

Schutz

(2019). Special educators’ experiences of roles and responsibilities in self-contained classes for students with emotional/behavioral disorders. Remedial and Special Education, 40, 177–191. https://doi.org/10.1177/0741932518762470

11.

Billingsley

McLeskey

Crockett

J. B.

(2017). Principal leadership: Moving toward inclusive and high-achieving schools for students with disabilities (Document No. IC-8). Collaboration for Effective Educator Development, Accountability, and Reform Center, University of Florida.

12.

Bishop

A. G.

Brownell

M. T.

Klingner

J. K.

Leko

M. M.

Galman

S. A.

(2010). Differences in beginning special education teachers: The influence of personal attributes, preparation, and school environment on classroom reading practices. Learning Disability Quarterly, 33, 75–92. https://doi.org/10.1177/073194871003300202

13.

Brownell

M. T.

Lauterbach

A. A.

Dingle

M. P.

Boardman

A. G.

Urbach

J. E.

Leko

M. M.

. . . Park

(2014). Individual and contextual factors influencing special education teacher learning in literacy learning cohorts. Learning Disability Quarterly, 37, 31–44. https://doi.org/10.1177/0731948713487179

14.

Brunsting

N. C.

Sreckovic

M. A.

Lane

K. L.

(2014). Special education teacher burnout: A synthesis of research from 1979 to 2013. Education and Treatment of Children, 37, 681–711. https://doi.org/10.1353/etc.2014.0032

15.

Chen

Westman

Eden

(2009). Impact of enhanced resources on anticipatory stress and adjustment to new information technology: A field experimental test of conservation of resources theory. Journal of Occupational Health Psychology, 14, 219–230.

16.

Common

E. A.

Lane

K. L.

Cantwell

E. D.

Brunsting

N. C.

Oakes

W. P.

. . . Bross

L. A.

(2020). Teacher-delivered strategies to increase students’ opportunities to respond: A systematic methodological review. Behavioral Disorders, 45, 67–84. https://doi.org/10.1177/0198742919828310

17.

Desimone

Porter

Garet

Yoon

K. S.

Birman

(2002). Effects of professional development on teachers’ instruction: Results from a three-year longitudinal study. Educational Evaluation and Policy Analysis, 24, 81–112. https://doi.org/10.3102/01623737024002081

18.

Dillman

D. A.

(2007). Mail and Internet surveys: The tailored design method. John Wiley.

19.

Garet

M. S.

Porter

A. C.

Desimone

Birman

B. F.

Yoon

K. S.

(2001). What makes professional development effective? Results from a national sample of teachers. American Educational Research Journal, 38, 915–945. https://doi.org/10.3102/00028312038004915

20.

Goddard

R. D.

Goddard

Y. L.

(2001). A multilevel analysis of the relationship between teacher and collective efficacy in urban Schools. Teaching and Teacher Education, 17, 807–818. https://doi.org/10.1016/S0742-051X(01)00032-4

21.

Grossman

Thompson

(2004). District policy and beginning teachers: A lens on teacher learning. Educational Evaluation and Policy Analysis, 26, 281–301.

22.

Halbesleben

J. R. B.

Neveu

J.-P.

Paustian-Underdahl

S. C.

Westman

(2014). Getting to the “COR”: Understanding the role of resources in conservation of resources theory. Journal of Management, 40, 1334–1364. https://doi.org/10.1177/0149206314527130

23.

Hobföll

S. E.

(1989). Conservation of resources: A new attempt at conceptualizing stress. American Psychologist, 44, 513–524. https://doi.org/10.1037/0003-066X.44.3.513

24.

Jackson

Bruegmann

(2009). Teaching students and teaching each other: The importance of peer learning for teachers. American Economic Journal: Applied Economics, 1(4), 85–108. https://doi.org/10.1257/app.1.4.85

25.

Jackson

Makarin

(2018). Can online off-the-shelf lessons improve student outcomes? Evidence from a field experiment. American Economic Journal: Economic Policy, 10(3), 226–254. https://doi.org/10.1257/pol.20170211

26.

Jimenez

B. A.

Saunders

A. F.

(2014). The additive effects of scripted lessons plus guided notes on science quiz scores of students with intellectual disability and autism. The Journal of Special Education, 47, 231–244. https://doi.org/10.1177/0022466912437937

27.

Johnson

S. M.

Kraft

M. A.

Papay

J. P.

(2012). How context matters in high-need schools: The effects of teachers’ working conditions on their professional satisfaction and their students’ achievement. Teachers College Record, 114(10), 1–39.

28.

Kauffman

J. M.

Landrum

T. J.

(2018). Characteristics of emotional and behavioral disorders of children and youth (11th ed.). Pearson.

29.

Kiely

M. T.

Brownell

M. T.

Lauterbach

A. A.

Benedict

A. E.

(2014). Teachers’ beliefs about students with special needs and inclusion. In Fives

Gill

M. G.

(Eds.), Handbook of research on teacher beliefs (pp. 475–491). Routledge.

30.

Klassen

R. M.

Bong

Usher

E. L.

Chong

W. H.

Huan

V. S.

Wong

I. Y.

Georgiou

(2009). Exploring the validity of a Teachers’ Self-Efficacy Scale in five countries. Contemporary Educational Psychology, 34(1), 67–76. https://doi.org/10.1016/j.cedpsych.2008.08.001

31.

Kline

R. B.

(2016). Principles and practice of structural equation modeling. Guilford Press.

32.

Kraft

M. A.

Papay

J. P.

(2014). Can professional environments in schools promote teacher development? Explaining heterogeneity in returns to teaching experience. Educational Evaluation and Policy Analysis, 36, 476–500. https://doi.org/10.3102/0162373713519496

33.

LeBreton

J. M.

Senter

J. L.

(2008). Answers to 20 questions about interrater reliability and interrater agreement. Organizational Research Methods, 11(4), 815–852. https://doi.org/10.1177/1094428106296642

34.

Leko

M. M.

Chiu

M. M.

Roberts

C. A.

(2018). Individual and contextual factors related to secondary special education teachers’ reading instructional practices. Journal of Special Education, 51, 236–250. https://doi.org/10.1177/0022466917727514

35.

Levy

Vaughn

(2002). An observational study of teachers’ reading instruction of students with emotional or behavioral disorders. Behavioral Disorders, 27, 215–235. https://doi.org/10.1177/019874290202700303

36.

Lewis

T. J.

Hudson

Richter

Johnson

(2004). Scientifically supported practices in emotional and behavioral disorders: A proposed approach and brief review of current practices. Behavioral Disorders, 29, 247–259. https://doi.org/10.1177/019874290402900306

37.

Little

R. J. A.

(1988). A test of missing completely at random for multivariate data with missing values. Journal of the American Statistical Association, 83, 1198–1202.

38.

Maggin

D. M.

Wehby

J. H.

Partin

T. C. M.

Robertson

Oliver

R. M.

(2011). A comparison of the instructional context for students with behavioral issues enrolled in self-contained and general education classrooms. Behavioral Disorders, 36, 84–99. https://doi.org/0.1177/019874291103600201

39.

Maslach

Jackson

S. E.

(1981). The measurement of experienced burnout. Journal of Organizational Behavior, 2, 99–113. https://doi.org/10.1002/job.4030020205

40.

McKenna

J. W.

Ciullo

(2016). Typical reading instructional practices provided to students with emotional and behavioral disorders in a residential and day treatment setting: A mixed methods study. Residential Treatment for Children & Youth, 33, 225–246. https://doi.org/10.1080/0886571X.2016.1207217

41.

McLeskey

Waldron

Redd

(2014). A case study of a highly effective inclusive elementary school. Journal of Special Education, 48, 59–70. https://doi.org/10.1177/0022466912440455

42.

Muthén

L. K.

Muthén

B. O.

(2018). Mplus user’s guide (8th ed.).

43.

Nevitt

Hancock

G. R.

(2001). Performance of bootstrapping approaches to model test statistics and parameter standard error estimation in structural equation modeling. Structural Equation Modeling, 8, 353–377.

44.

O’Brien

K. M.

Brunsting

N. C.

Bettini

Cumming

M. M.

Ragunathan

Sutton

(2019). Special educators’ working conditions in self-contained settings for students with emotional or behavioral disorders: A descriptive analysis. Exceptional Children, 86, 40–57. doi:10.1177/0014402919868946.

45.

Office of Special Education and Rehabilitative Services. (2018). 40th Annual Report to Congress on the Implementation of the Individuals with Disabilities Education Act. https://www2.ed.gov/about/reports/annual/osep/2018/parts-b-c/40th-arc-for-idea.pdf

46.

Porter

A. C.

Smithson

Blank

Zeidner

(2007). Alignment as a teacher variable. Applied Measurement in Education, 20, 27–51. https://doi.org/10.1080/08957340709336729

47.

Ronfeldt

Farmer

S. O.

McQueen

Grissom

J. A.

(2015). Teacher collaboration in instructional teams and student achievement. American Educational Research Journal, 52, 475–514. https://doi.org/10.3102/0002831215585562

48.

Ross

S. W.

Romer

Horner

R. H.

(2012). Teacher well-being and the implementation of school-wide positive behavior interventions and supports. Journal of Positive Behavior Interventions, 14, 118–128. https://doi.org/10.1177/1098300711413820

49.

Rozalski

Stewart

Miller

(2010). How to determine the least restrictive environment for students with disabilities. Exceptionality, 18, 151–163. https://doi.org/10.1080/09362835.2010.491991

50.

Ruble

L. A.

McGrew

J. H.

(2013). Teacher and child predictors of achieving IEP goals of children with autism. Journal of Autism and Developmental Disorders, 43, 2748–2763. https://doi.org/10.1007/s10803-013-1884-x

51.

Ruble

L. A.

Usher

E. L.

McGrew

J. H.

(2011). Preliminary investigation of the sources of self-efficacy among teachers of students with autism. Focus on Autism and Other Developmental Disabilities, 26, 67–74. https://doi.org/10.1177/1088357610397345

52.

Russ

Chiang

Rylance

B. J.

Bongers

(2001). Caseload in special education: An integration of research findings. Exceptional Children, 67, 161–172. https://doi.org/10.1177/001440290106700202

53.

Simonsen

Fairbanks

Briesch

Myers

Sugai

(2008). Evidence-based practices in classroom management: Considerations for research to practice. Education and Treatment of Children, 31, 351–380. https://doi.org/10.1353/etc.0.0007

54.

Siuty

M. B.

Leko

M. M.

Knackstedt

K. M.

(2018). Unraveling the role of curriculum in teacher decision making. Teacher Education and Special Education, 41, 39–57. https://doi.org/10.1177/0888406416683230

55.

Stajkovic

A. D.

Bandura

Locke

E. A.

Lee

Sergent

(2018). Conceptual models of influence of the big five personality traits and self-efficacy on academic performance: A meta-analytic path-analysis. Personality and Individual Differences, 120, 238–245.

56.

Sun

Loeb

Grissom

J. A.

(2017). Building teacher teams: Evidence of positive spillovers from more effective colleagues. Educational Evaluation and Policy Analysis, 39, 104–125. https://doi.org/10.3102/0162373716665698

57.

Tschannen-Moran

Hoy

A. W.

(2001). Teacher efficacy: Capturing an elusive construct. Teaching and Teacher Education, 17, 783–805. https://doi.org/10.1016/S0742-051X(01)00036-1

58.

Tschannen-Moran

Hoy

A. W.

(2007). The differential antecedents of self-efficacy beliefs of novice and experienced teachers. Teaching and Teacher Education, 23, 944–956. https://doi.org/10.1016/j.tate.2006.05.003

59.

Vannest

K. J.

Soares

D. A.

Harrison

J. R.

Brown

Parker

R. I.

(2010). Changing teacher time. Preventing School Failure, 54, 86–98. https://doi.org/10.1080/10459880903217739

60.

Wagner

(2014). Longitudinal outcomes and post-high school status of students with emotional or behavioral disorders. In Walker

H. M.

Gresham

F. M.

(Eds.), Handbook of evidence-based practices for emotional and behavioral disorders: Applications in schools (pp. 86–103). Guilford Press.

61.

Wanzek

Vaughn

(2007). Research-based implications from extensive early reading interventions. School Psychology Review, 36, 541–561.

62.

Wong

V. W.

Ruble

L. A.

McGrew

J. H.

(2017). Too stressed to teach? Teaching quality, student engagement, and IEP outcomes. Exceptional Children, 83, 412–427. https://doi.org/10.1177/0014402917690729

63.

Youngs

Frank

K. A.

Thum

Y. M.

Low

(2012). The motivation of teachers to produce human capital and conform to their social contexts. In Smith

Desimone

Porter

A. C.

(Eds.), Yearbook of the National Society for the Study of Education: Vol. 110. Organization and effectiveness of high-intensity induction programs for new teachers (pp. 248–272). Blackwell.

64.

Zee

Koomen

H. M. Y.

(2016). Teacher self-efficacy and its effect on classroom processes, student academic adjustment, and teacher well-being: A synthesis of 40 years of research. Review of Educational Research, 86, 981–1015. https://doi.org/10.3102/0034654315626801

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB

0.06 MB

0.05 MB

0.06 MB

0.02 MB

0.08 MB

0.03 MB