Early Childhood Research Quarterly 42 (2018) 280–290

Contents lists available at ScienceDirect

Early Childhood Research Quarterly

Does professional development reduce the influence of teacher stress

on teacher–child interactions in pre-kindergarten classrooms?夽
Lia E. Sandilos a,∗ , Priscilla Goble b , Sara E. Rimm-Kaufman c , Robert C. Pianta d
a
College of Education, Temple University, 263 Ritter Annex, 1301 Cecil B. Moore Avenue, Philadelphia, PA 19122, United States
b
School of Family & Consumer Sciences, Texas State University, 601 University Drive, San Marcos, TX 78666, United States
c
Curry School of Education, University of Virginia, 238 Ruffner Hall, P.O. Box 800784, Charlottesville, VA 22904, United States
d
Curry School of Education, University of Virginia, 131D Bavaro Hall, P.O. Box 400260, Charlottesville, VA 22904, United States

a r t i c l e i n f o a b s t r a c t

Article history: The present study examines the extent to which participation in a 14-week professional development
Received 15 July 2016 course designed to improve teacher–child interactions in the classroom moderated the relation between
Received in revised form 9 September 2017 teacher-reported job stress and gains in observed teacher–child interaction quality from the beginning
Accepted 13 October 2017
to the end of the intervention. Participants were preschool teachers (N = 427; M age = 42) with an average
Available online 2 November 2017
of 11 years of experience teaching. Teachers reported how intensely they experienced different sources
of stress at pre-test only (i.e., prior to being randomized into the treatment condition [course or control]).
Keywords:
Teacher–child interactions were measured through classroom observations at pre and post intervention.
CLASS Pre-K
Teacher stress
Results demonstrated that control teachers reporting higher professional investment stress showed fewer
Professional development gains in observed emotional support relative to control teachers experiencing less professional invest-
Preschool ment stress. These findings were not evident for teachers in the course condition. Interestingly, teachers
with higher professional investment stress showed fewer gains in instructional support in the control
condition and greater gains in the course condition, relative to teachers in their respective treatment
groups who reported lower levels of professional investment stress. Findings suggest that participation
in the professional development intervention had a buffering effect on the negative association between
professional investment stress and emotional support. With regard to instructional support, it is possible
that teachers’ heightened awareness and anxiety over their need to develop professionally may have
made them more responsive to an intervention designed to improve practice.
© 2017 Elsevier Inc. All rights reserved.

1. Introduction reduce the quality of these important interactions with students

(Li-Grining et al., 2010; Yoon, 2002). As a result, many educational
Children’s experiences in early learning environments are crit- researchers have expended great effort to try to better understand
ical for developmental outcomes. A growing body of research teacher stress and identify ways in which teachers can be sup-
on preschool settings indicates that high quality interactions ported so that stress does not take a toll on their instruction and
between teachers and students foster children’s academic and interactions (e.g., Chang, 2009; Kyriacou, 2001; Lambert, O’Donnell,
social-emotional readiness skills (Burchinal, Zaslow, & Tarullo, Kusherman, & McCarthy, 2006; Zhai, Raver, & Li-Grining, 2011).
2016; Raver et al., 2011; Yoshikawa et al., 2013). At the same This study draws from Jennings and Greenberg’s (2009) model
time, literature reports that teachers’ experience with stress can of a prosocial classroom, which advances the importance of teacher
emotional well-being. The model describes the impact of stress as
a “burnout cascade” (p. 492). Stressed teachers without the emo-
夽 We express our appreciation to the generous programs and teachers who par- tional resources to meet the challenges of teaching have poorer
ticipated in this study. The research reported herein was supported by the Institute quality interactions with their students. Consequently, students
of Education Sciences, U.S. Department of Education, through Grants R305B130013 spend more time off task and show more problem behaviors, which
and R305A060021 to the University of Virginia. The opinions expressed are those of in turn, produce a more stressful classroom climate. The stressful
the authors and do not represent views of the U.S. Department of Education.
∗ Corresponding author. classroom climate leads to more challenge, stress, and exhaustion
E-mail addresses: lia.sandilos@temple.edu (L.E. Sandilos), goble@txstate.edu for the teacher and the cascade continues. Jennings and Green-
(P. Goble), serk@virginia.edu (S.E. Rimm-Kaufman), rcp4p@virginia.edu berg indicate that in-service professional development may be one
(R.C. Pianta).

https://doi.org/10.1016/j.ecresq.2017.10.009
0885-2006/© 2017 Elsevier Inc. All rights reserved.
 L.E. Sandilos et al. / Early Childhood Research Quarterly 42 (2018) 280–290 281

way to ameliorate feelings of job stress and boost the quality of Breckler, 2000; Greenberg, Brown, & Abenavoli, 2016). Stress is
teacher–child relationships. Yet, professional development oppor- also not unidimensional. Teachers’ experience with job stress stems
tunities take time, increase workload, and could even result in more from various sources–both personal and environmental influences.
feelings of stress and burnout, rather than less (Ozer & Beycioglu, Some teacher stress reflects investment in their professional career.
2010). Therefore, the current study bridges these two lines of Other stress stems from behavioral problems and low motivation
research: teacher stress and professional development. Given on the part of their students. Yet another source of stress emanates
that existing empirical and theoretical work indicates teacher from work issues such as high workload, unusually high numbers
stress is negatively associated with the quality of teacher–child of students in their class, or too many professional responsibili-
interactions, the purpose of this study was to examine whether par- ties (Fimian & Fastenau, 1990; von der Embse, Kilguss, Solomon,
ticipation in a high quality professional development intervention Bowler, & Curtiss, 2015).
targeting teachers’ interactions and relationships with students Prior research has described professional investment stress as
moderates the relation between teachers’ self-reported job stress the most prevalent source of stress among teachers. Teachers’
and their observed teacher–child interactions. feelings of stress related to professional investment included frus-
trations with lack of control over job-related decisions and limited
1.1. Teacher–child interactions access to professional growth opportunities, as well as a feeling
of low intellectual or emotional stimulation (Fimian & Fastenau,
Types of interactions that are beneficial to young students relate 1990). Indeed, more recent data suggest that ongoing professional
to the emotional and instructional supports provided by teach- learning opportunities for teachers in the U.S. are limited com-
ers, as well as teachers’ management of the classroom (Pianta & pared to other high-achieving countries (Darling-Hammond, Wei,
Hamre, 2009). Teachers facilitate the development of students’ & Andree, 2010). Additionally, for preschool teachers, perceived
social-emotional competence by creating warm and emotionally lack of control over their job is regarded as a major contributor to
supportive environments infused with mutual respect and posi- job stress and has been shown to be detrimental to classroom prac-
tive communication, providing opportunities for autonomy, and tice (Curbow et al., 2000; Hagekull & Hammarberg, 2004; Raver,
demonstrating sensitivity to students’ emotions (Birch & Ladd, 2004).
1997; Brock & Curby, 2014; Denham, Bassett, & Zinsser, 2012; Managing student behavior is another prominent and fre-
Pianta & Steinberg, 1992). Teachers establish productive and orga- quently studied source of stress for teachers, regardless of the grade
nized classrooms that support academic instruction and student level taught (Friedman-Krauss, Raver, Neuspiel, & Kinsel, 2014;
learning behaviors by instituting consistent behavioral expecta- Grayson & Alvarez, 2008). As one example, a study of teacher stress
tions and classroom routines, as well as maximizing instructional in the primary through secondary grades indicated that stressors
learning time (Choi et al., 2016; McLeod, Fisher, & Hoover, 2003). related to student behavior had more detrimental effects on feel-
Relatedly, teachers provide high quality instruction by fostering a ings of teaching efficacy than stress associated with overall teaching
deeper understanding of academic concepts through open-ended workload (Klassen & Chiu, 2010). Moreover, the concurrent rela-
questions, problem solving, and real-world application, and by tion between high teacher stress levels and increased behavioral
encouraging language development through conversation, repeti- issues or teacher-student conflict has been highlighted in the lit-
tion, and elaboration (Hamre & Pianta, 2005; Torgesen, 2002). erature regarding teacher well-being (e.g., Jennings & Greenberg,
National trends in teacher–child interactions indicate that the 2009; Whitaker et al., 2015).
quality of interactions often declines over the course of the year Structural features of early childhood education, such as large
(National Center on Quality Teaching and Learning, 2013). To bet- class size and limited planning time, have long been considered
ter understand how teachers can maintain high quality interactions stressful aspects of the occupation (Raver, 2004). More recently,
with students throughout the school year, researchers need to con- increased workload associated with documentation and paperwork
sider the psychological and professional factors that influence the has accompanied accountability reforms (Gooze, 2014; Stipek,
quality of teacher–child interactions (Rimm-Kaufman & Hamre, 2006). In many early childhood settings these high work-related
2010). Teachers’ emotional well-being contributes to their func- demands do not occur in tandem with adequate job supports, and
tioning in the classroom setting (Hamre & Pianta, 2004; Jennings & for that reason, teachers experience high rates of burnout and
Greenberg, 2009; Sandilos et al., 2015). Teachers who experience turnover in the early childhood workforce (6; Maslach, Schaufeli, &
high levels of stress face difficulty achieving and sustaining high Leiter, 2001; Whitebook et al., 2014).
quality interactions with their students (Li-Grining et al., 2010). Existing literature shows that teacher stress is complex and
multifaceted. Not only is it important to consider the overall job
1.2. Teacher stress in early childhood settings stress a teacher is experiencing, but it is also valuable to exam-
ine sources of stress separately to disentangle key stressors that
Early childhood teachers are under immense pressure to ensure teachers face in their profession (Chang, 2009; Curbow et al., 2000;
that their young students are prepared for future success in school. Klassen, Foster, Sajani, & Bowman, 2009). Furthermore, exploring
These pressures have been compounded by the reality that early the ways in which different sources of stress influence teachers’
childhood professionals still tend to be underappreciated by society interactions with students has the potential to provide more spe-
as evidenced by lack of mobility within the career track and dis- cific levers for improving school-based professional development
parities in pay between preschool teachers and teachers in grades and other interventions targeting work-based stress.
K-12 (National Survey of Early Care and Education, 2013; USDH,
2016; Whitebook, Philips, & Howes, 2014). The high professional 1.3. Teacher stress and interaction quality
demands and low compensation make early childhood educators
particularly vulnerable to stress (Gooze, 2014). Teacher stress has consequences for the quality of interactions
Teacher stress is broadly defined as a negative emotional expe- in the preschool classroom (Hamre & Pianta, 2004), which in turn
rience associated with an individual’s ability to cope with job influences the social, emotional, and academic skills of the young
stressors (Kyriacou, 2001). Teachers’ experience with stress is children (Raver, 2004). For instance, preschool teachers who report
related to a variety of negative outcomes, such as poor health, high levels of stress tend to be in classrooms rated lower in emo-
lower quality instruction, burnout, absenteeism, and turnover (e.g., tional support, classroom management, and instructional support
Alkon, Ramler, & MacLennan, 2003; Curbow, Spratt, Ungaretti, & (Collmann, 2012; Li-Grining et al., 2010). Conversely, reductions
282 L.E. Sandilos et al. / Early Childhood Research Quarterly 42 (2018) 280–290

in stress are linked to teachers’ provision of higher quality envi- 1.5. Present study
ronments and improved outcomes for students (Zinsser, Bailey,
Curby, Denham, & Bassett, 2013). As one example, lower lev- The present study examines the extent to which a 14-week pro-
els of overall teacher job stress and higher ratings of classroom fessional development course designed to improve teacher–child
management were related to increased ratings of learning moti- interactions in the classroom moderated the relation between
vation in preschoolers (Pakarinen et al., 2010). Although existing sources of teacher-reported job stress and gains in observed qual-
work establishes the link between higher stress and lower qual- ity of teacher–child interactions from the beginning to the end of
ity teacher–child interactions in early childhood classrooms, less is the intervention. The following research question was explored: Is
known about different sources of job stressors and their implica- the relation between teachers’ ratings of different sources of stress
tions for preschool teachers’ interactions with their students, which and the quality of their interactions with preschool students mod-
is an issue addressed in the present study. erated by participation in a PD intervention focused on improving
teacher–child interactions?
The current study extends previous research by examining how
1.4. Professional development different sources of job stress relate to preschool teachers’ interac-
tion quality. Three sources of stress (i.e., professional investment,
One potential way to buffer against the effects of teacher job discipline and motivation, and work-related stressors) and three
stress is to provide teachers with additional supports and skills types of interactional quality (i.e., emotional support, classroom
through professional development (Jennings & Greenberg, 2009). organization, and instructional support) were examined separately
High quality professional development (PD) can result in positive to better understand how various sources of stress might be dif-
changes in preschool teachers’ instruction and improved outcomes ferentially associated with gains in different types of interactional
for young children (Kinzie et al., 2014; Pianta, Mashburn, Downer, quality. Moderation was tested by examining whether or not the
Hamre, & Justice, 2008). However, the PD experiences provided to relations between source of stress and change in interaction qual-
teachers in schools and center-based settings vary widely in struc- ity depended on assignment to the intervention (course) or control
ture and quality (Sheridan, Edwards, Marvin, & Knoche, 2009). The condition.
variability and inconsistent quality of teacher PD spawned a large
body of research exploring effective and systematic methods for 2. Method
providing additional training to teachers. Concurrently, the mount-
ing evidence that teacher–child interactions are a vital aspect of The data for this study were drawn from a large multi-site, ran-
early childhood programming has resulted in the creation of PD domized controlled trial (i.e., National Center for Research on Early
interventions that target interaction quality in preschools (e.g., Childhood Education [NCRECE] Professional Development Study;
Domitrovich et al., 2009; Pianta, Mashburn et al., 2008). Initial Downer et al., 2014; Hamre et al., 2012). The larger study evaluated
outcomes from PD interventions aimed to improve teacher–child two forms of professional development aimed to improve teach-
relationships indicate that teachers can change the way they inter- ers’ interactions with children over three phases of intervention.
act with students through the use of observations and feedback The first phase of the intervention (Phase 1) consisted of a 14-week
(Hamre, Downer, Jamil, & Pianta, 2012; Zan & Donegan-Ritter, college-level course, as described below. Data for this study were
2014). drawn from Phase 1 only.1
Despite the strong research on PD and teacher–child interac-
tions in early childhood, somewhat less is known about the relation 2.1. Participants
between preschool teachers’ experience with stress and PD expe-
riences. One possibility is that additional PD is not sufficient to Participants were preschool teachers recruited from large com-
influence the relation between stress and interaction quality. Yet munity preschools and Head Start programs across ten sites (i.e.,
another possibility is that participation in PD could provide teach- cities) in the United States (New York, NY; Hartford, CT; Chicago,
ers with new skills and strategies to utilize in the classroom, which IL (2 sites); Stockton, CA; Dayton, OH; Columbus, OH; Memphis,
may in turn render teachers’ feelings of stress less influential on TN; Charlotte, NC; Providence, RI). Participants were eligible for
their interactions with students. International work on this topic NCRECE based on four criteria: (a) they were the lead teacher in a
has shown that teachers exhibit positive attitudes toward PD when publicly-funded classroom in which the majority of children were
they perceive a link between their accomplishments and the PD eligible for kindergarten the following school year, (b) the majority
experience, and demonstrate negative attitudes toward PD when of children did not have an IEP at the start of the current school
they feel the experience increased their feelings of burnout (Ozer year, (c) classroom instruction was primarily in English, and (d)
& Beycioglu, 2010). Though not directly examined in this study, it high-speed internet access was available for teacher use at the pro-
is important to note that there are a variety of potential mech- gram site. The research team extended invitations to the program
anisms influencing the relation between PD and teacher stress. administrators, established program agreements, and facilitated
For example, when administered effectively, PD experiences can IRB approval at specific sites. Teacher recruitment was conducted
increase teachers’ feelings of self-efficacy in the targeted area of PD after administrators confirmed that their site would participate in
instruction (Goddard, Hoy, & Woolfolk Hoy, 2004; Rimm-Kaufman the study. After agreeing to take part in the study, teachers were
& Sawyer, 2004; Tschannen-Moran & McMaster, 2009), which, in then randomized into the course or control condition within site
turn, may mitigate any negative feelings of stress associated with for the first phase of the study (Hamre et al., 2012). The number
that aspect of their job (Schwarzer & Hallum, 2008). Moreover, of teachers assigned to the course condition ranged from 37% to
a characteristic of high quality PD is providing teachers with the 56% across the ten sites. Incentives for participation in the study
opportunity to discuss new concepts or skills and to talk about included the receipt of stipends as data collection materials were
problems that arise in the classroom (Garet, Porter, Desimone, submitted as well as monthly books. Teachers’ names were also
Birman, & Yoon, 2001), the act of which may be stress-relieving.
Thus, participation in PD targeted at improving teacher–child inter-
actions may buffer the negative association between teachers’ 1
Phase 2 included coaching and Phase 3 encompassed a post-intervention period
feelings of stress and interaction quality between teachers and of data collection. For more information regarding Phases 2 and 3, please see related
students. NCRECE publications (e.g., Downer et al., 2014; Pianta et al., 2014).
 L.E. Sandilos et al. / Early Childhood Research Quarterly 42 (2018) 280–290 283

Table 1 2.2. Intervention description

Items for teacher stress inventory composites.

Items Of the 427 teachers, 218 (51%) were randomly assigned to

Work-Related Stress the professional development intervention; a 14-week course
There is little time to prepare lessons/responsibilities. designed to increase teachers’ knowledge about the vital role
I have too much work to do. that teacher–child interactions play in learning and skill acquisi-
The pace of the day is too fast. tion. The PD was constructed to build specific skills for observing
My personal priorities are being shortchanged due to time demands.
teacher–child interactions that contribute to language and liter-
There is too much administrative paperwork in my job.
My class is too big. acy skills. The intervention targeted language and literacy because
observational evidence suggests that preschool teachers focus
Professional Investment Stress
I lack opportunities for professional improvement.
more on these skills than other academic content areas (Early et al.,
My personal opinions are not sufficiently aired. 2010).
I am not emotionally/intellectually stimulated by my job. The course provided very specific knowledge about effective
I lack control over classroom decisions/matters. interactions and used the Classroom Assessment Scoring System
Discipline and Motivation Stress (CLASS; Pianta, La Paro et al., 2008), a validated observational mea-
I feel frustrated because some students would do better if they tried harder. sure, as the framework for this knowledge. Teachers were taught
I feel frustrated because of discipline problems in my classroom. to make explicit links between teachers’ behavioral actions and
I feel frustrated when my authority is rejected by pupils/administration.
I feel frustrated having to monitor pupil behavior.
intended consequences for children. For example, when learning
I feel frustrated attempting to teach students who are poorly motivated. about behavior management, teachers were encouraged to watch
I feel frustrated because of inadequate/poorly defined discipline practices. and analyze videos that highlighted the ways in which specific
Note. Items are scored on a range of 1 (not stressful) to 5 (highly stressful). teacher actions led to more or less positive behaviors among stu-
dents in the classrooms. The course also targeted teachers’ skills in
detecting effective teacher–child interactions through video anal-
ysis.
The course was delivered in 14, 3-h-long sessions through col-
laborations with local colleges and universities in each site. On
entered into periodic drawings for gift certificates. At many of the average, teachers had good attendance rates (M = 82%, SD = 0.25).
sites, teachers received three college-level credits for their partici- There were a range of 5–15 teachers in each course section. The
pation in the course. first three sessions provided teachers with information on the
Of the 427 teachers (M age = 42) who enrolled in Phase 1 of the framework for the course and covered materials such as why
study, 239 teachers had complete data at the end of the intervention preschool experiences are important for long-term development,
year. The most common reason for missing data was that teachers the importance of teacher–child interactions and relationships for
dropped out of the study due to competing time commitments. In promoting children’s development, and introduction to the three
addition, several teachers were missing CLASS observational data broad domains of the CLASS: Emotional Support, Classroom Organi-
because they took the course in the summer and they were not zation, and Instructional Support. Following this introduction, 1–2
teaching at that time (Hamre et al., 2012). Little’s MCAR test (con- sessions focused more deeply on each of the CLASS domains. Each
ducted using SPSS Version 23) demonstrated that missing values session introduced teachers to types of effective interactions and
for the key study variables were missing at random (MAR), X2 presented videos in which teachers analyzed the extent to which
(6) = 10.22, p = 0.12. Accordingly, full information maximum like- these interactions were present or absent. Homework included
lihood estimator (FIML) in Mplus7 (Muthén & Muthén, 1998–2012) readings and analyzing videos online. The next series of sessions
was used to handle missing data for the full sample (N = 427). focused on language and literacy development and instruction. In
On average, teachers were approximately 42 years old, had the final session, teachers were asked to film themselves delivering
11 years of experience teaching, and made $30,000 to $34,999 a language and literacy activity and to share their video with fellow
a year. Teachers also had an average of 15 years of education teachers while highlighting examples of effective (or ineffective)
and were diverse in terms of their educational background (high interactions throughout.
school or vocational degree = 10%; associate’s degree = 22%; bach- The course was delivered by instructors who were required
elor’s degree = 47%; master’s degree or higher = 21%). Half of the to have a master’s degree or higher in early childhood education
teachers had a degree in child development (5%) or early child- or a related field, experience in early childhood education and
hood education (45%). About one third had an education degree of in teaching college-level students from diverse backgrounds, and
some type (33%; e.g., elementary education, special education) and proficiency with technology. Each course section was taught in
the remaining teachers (17%) did not have an education-related person by an individual instructor. To prepare for the course, the
degree. Most of the teachers were African American (44%) or Euro- instructors attended a week-long training and received on-going
pean American (30%). Relatively fewer participants were Latino implementation support from NCRECE staff (e.g., weekly phone
American/Hispanic (16%) or Asian and other ethnicities (7%). calls from course developers). To monitor the fidelity of implemen-
The majority (63%) of teachers worked in Head Start programs tation, course sessions were videotaped and coded by the NCRECE
and the remaining teachers were employed in either public schools staff and instructors received feedback on their implementation.
or private centers. Almost half of the children in teachers’ class- Videotaped coding as well as reports provided by the participat-
rooms were girls (48%). On average, classrooms were characterized ing teachers and course instructors indicated that the course was
by mostly African American (47%) and Latino American/Hispanic administered with high levels of fidelity with instructors covering
(32%) children. About 18% of students had limited English profi- 95% of course material and using high quality instructional methods
ciency and 9% had IEPs. There were approximately 18 children in (Hamre et al., 2012; LoCasale-Crouch et al., 2011).
each classroom. Descriptive statistics are presented in Table 1 by Teachers in the control condition received business-as-usual
treatment group. There were no significant differences in the demo- supports and were not exposed to the intervention coursework,
graphic or work characteristics between teachers in the course and though they may have been taking other courses at the time (Hamre
control groups. There were also no differences in the demographic et al., 2012).
make-up of their classrooms.
284 L.E. Sandilos et al. / Early Childhood Research Quarterly 42 (2018) 280–290

2.3. Procedures and measures motivation (e.g., student behavior problems, student motivation;
␣ = 0.80).
Teachers completed an online questionnaire prior to beginning
the course that included demographic information and a self-report 2.3.3. Covariates
measure of job stress. The self-report measure of job stress was A number of covariates were included either because they were
collected only at the start of Phase 1 of the study. Teacher use determined to be conceptually important in this sample or because
of effective teacher–child interactions was coded from videotapes they had appeared in prior studies of teacher–child interactions
teachers submitted to the research team. All teachers were pro- (Cabell et al., 2013; Hamre et al., 2012; Mashburn et al., 2008;
vided with a digital video camera and were asked to submit four Pianta, Mashburn et al., 2008). The full list of covariates included:
30-min videos during the course phase of the study. Teachers were teacher race/ethnicity, annual income, years of education, years
provided with specific instructions regarding the types of lessons of teaching experience, teaching efficacy/beliefs,2 and Head Start
to tape. Each videotape was required to meet the following criteria: affiliation (e.g., teaching in a Head Start center). Additionally, to
(a) at least 30 consecutive minutes of instruction, (b) a language or examine change in teacher–child interactions using residualized
literacy lesson/activity is taking place that is reflective of a typical change scores, the time 1 (fall) CLASS domain was entered into the
day in the classroom, (c) the teacher is interacting with students, model as a predictor and the corresponding time 2 (spring) CLASS
and (d) the teacher is visible throughout the taping. Two 15-min domain was used as the outcome in each model.
segments were CLASS coded from each 30-min video. The segments
selected for this study were gathered at two time points: a) time 2.4. Analytic approach
1, corresponding to the time period before the start of the course
through the first two weeks of the course, and b) time 2, corre- Correlation coefficients showed associations among the stress
sponding to the time period between the mid-point of the course subscales (work-related stressors, professional investment, and
to within 2 weeks after the last day of class. discipline and motivation) with coefficients ranging from 0.56 to
Coding was randomly assigned to raters at the segment level. 0.67. Including all three subscales in a single model raised multi-
Each segment was double-coded. Coders attended a 2-day CLASS collinearity concerns. However, analyzing each separately meant
training and had to pass the CLASS reliability that findings in models for each source of stress also contain
test, which requires scoring five segments and demonstrating information about the portion of variance shared across sub-
consistency with master codes (80% of codes within 1 point of mas- scales. Resolution involved a two-part approach: first, analyzing
ter code). Average reliability for the Reliability I test was 84%, with a each source of stress alone in separate models (keeping all the
range of 60% to 100%. Coders who did not pass this initial test were other covariates the same), and second, computing models with
provided with feedback and given a second test. Coders who did all three subscales entered simultaneously. Results revealed sig-
not pass this second test were assigned to other responsibilities. nificant moderation regardless of the analytic approach. In models
Throughout the coding period, all coders attended weekly meet- examining each source of stress separately, simple slopes for spe-
ings that focused on assessing progress and reliability, as well as cific sources of stress were significant; whereas, in the model with
addressing issues of potential drift. During these meetings 89% of all subscales entered simultaneously, the simple slopes of the mod-
codes were within 1 point of the master code. eration analyses exhibited similar trends but were non-significant
(for both groups). Careful examination of the results from both
models speaks to the potential suppression effects at play in the
2.3.1. Teacher–child interactions model containing the three correlated predictors. Thus, results are
Teacher–child interactions were measured using the Class- reported from models that included each source of stress sepa-
room Assessment Scoring System – Pre-Kindergarten (CLASS Pre-K; rately and findings are interpreted in ways that acknowledge this
Pianta, La Paro et al., 2008). The CLASS Pre-K includes10 dimen- decision.
sions assessing teacher–child interactions. Each dimension is rated The current study had 10 data collection sites. Programs/centers
on a 7-point scale with behavioral indicators and anchor point within sites were not recorded as part of the data collection proce-
descriptions provided for low (1–2), medium (3–5), and high (6–7) dure. To correct for the non-independence of teachers within site,
levels of that dimension. The dimensions are aggregated to create we used the Mplus TYPE = COMPLEX option, which provides scaled
three primary domains. Emotional support consisted of positive cli- standard errors robust to non-independence and non-normality.
mate, negative climate (reversed), regard for student perspectives,
and teacher sensitivity. Classroom organization consisted of behav- 3. Results
ior management, instructional learning formats, and productivity.
Instructional support consisted of concept development, quality of Preliminary analyses were conducted using SPSS Version 23 to
feedback, and language modeling. examine the descriptive statistics of all study variables. The means
and standard deviations of all variables used in the final models are
listed by condition in Table 2. T-tests were conducted to examine
2.3.2. Teacher job stress differences in all variables due to treatment condition (course vs.
Teachers’ reported job stress was collected only at the pre- control; see Table 2). Results indicated significant differences on
test time point using the Teacher Stress Inventory (TSI; Fimian & all three teacher–child interaction outcome variables (emotional
Fastenau, 1990). In this inventory, teachers reported how intensely support, classroom organization, instructional support) at time 2.
they experienced stress, by responding to a series of statements Teachers in the course condition outperformed teachers in the con-
(e.g., my class is too big) using a Likert-type response scale rang- trol condition on all teacher–child interaction outcomes at time
ing from 1 (not stressful) to 5 (highly stressful). A full list of items 2. However, there were no differences between teachers on these
is presented in Table 1. Sixteen items make up the three sub- variables at time 1. There were no differences by condition for
scales of the inventory (subscales had adequate reliability [␣] teacher race/ethnicity, annual income, years of education, years
in the current sample): Work-related stressors (e.g., workload,
size of classes, professional responsibilities; ␣ = 0.80); profes-
sional investment (e.g., opportunities for professional improvement, 2
Details regarding Teaching Efficacy/Beliefs (Tschannen-Moran & Woolfolk Hoy,
emotional/intellectual stimulation; ␣ = 0.75); and discipline and 2001) can be obtained from the primary author.
 L.E. Sandilos et al. / Early Childhood Research Quarterly 42 (2018) 280–290 285

Table 2
Descriptive statistics for covariates by treatment condition.

Condition

Course (N = 218) Control (N = 209)

% M SD % M SD

Covariates
Teacher Characteristics
Race (African-American) 43 46
Ethnicity (Hispanic) 17 15
Annual Income $32,531 $10,870 $33,174 $11,775
Years of Education 15.51 1.54 15.71 1.65
Years of Experience 13.80 8.64 14.39 9.48
Teaching Efficacy/Beliefs 7.44 1.01 7.59 0.97
Classroom Characteristics
Head Start affiliation 61 35
Teacher-child Interactions
Emotional Support (pre) 5.30 0.48 5.26 0.50
Classroom Organization (pre) 2.50 0.68 2.36 0.69
Intructional Support (pre) 5.17 0.64 5.22 0.65

Teacher Stress Inventory

Work-related Stressors 2.11 0.67 2.14 0.79
Professional Investment 1.61 0.65 1.56 0.58
Discipline and Motivation 1.77 0.62 1.71 0.63

Teacher-child Interactions
Emotional Support (post) 5.42 a 0.49 5.17 a 0.58
Classroom Organization (post) 2.96 b 0.64 2.51 b 0.65
Intructional Support (post) 5.46 c 0.59 5.27 c 0.65

Note. Pre = pre-midterm. Post = post-midterm. Items with the same subscript significantly differ between course and control condition at p < 0.05. df for t-tests ranged from
237 to 345.

of teaching experience, teaching efficacy/beliefs, and Head Start ized change in interactional quality, interaction effects were tested
affiliation. Correlations between the study controls, predictors, and in a second set of regression models (Model 2, Table 4). Signif-
outcomes are presented by condition in Table 3. icant moderation was found for the course condition (receipt of
Moderation analyses were run in Mplus7 (Muthén & Muthén, PD course) in the relations between two types of teacher-reported
1998–2012) because the program supports the use of a full infor- job stress, professional investment and discipline and motivation,
mation maximum likelihood estimator (FIML) to handle missing and residualized change in all three types of interactional qual-
data and appropriately adjusts standard errors when clustering is ity. Specifically, the interaction between treatment condition and
present. All main effects were tested in an initial set of regression teacher-reported stress in professional investment was signifi-
models (Model 1, Table 4). As can be seen in Table 4, teachers who cantly related to change in emotional support (ˇ = 0.27, p < 0.01),
participated in the PD intervention made significantly greater gains classroom organization (ˇ = 0.15, p < 0.05), and instructional sup-
in the quality of their interactions with children across all domains port (ˇ = 0.26, p < 0.001). Additionally, the interaction between
(i.e., emotional support, classroom organization, instructional sup- treatment condition and teacher-reported stress in discipline and
port) than teachers in the control condition. motivation was significantly related to change in instructional sup-
To explore the moderation effect of participation in the PD port (ˇ = 0.16, p < 0.05).
intervention on the relation between sources of stress and residual-

Table 3
Correlations between study variables and spring teacher-child interactions by treatment condition

Condition

Course Control

ES CO IS ES CO IS

Covariates
Teacher Race (Black) 0.02 0.17 0.12 −0.10 0.07 −0.03
Teacher Ethnicity (Hispanic) 0.05 0.10 0.16 −0.03 −0.18 −0.08
Teacher Annual Income 0.07 0.19 * −0.06 0.00 0.12 0.25 **
Teacher Years of Education −0.04 0.10 0.05 0.09 0.18 * 0.10
Teacher Years of Experience 0.12 −0.15 0.19 * 0.05 0.02 −0.01
Teaching Efficacy/Beliefs 0.11 0.16 0.05 0.10 0.23 * 9.00
Head Start Affiliation −0.02 0.00 −0.16 0.08 −0.03 0.02
Teacher-child Interactions (pre) 0.19 * 0.21 * 0.28 *** 0.37 *** 0.22 * 0.49 ***

Teacher Stress Inventory

Teacher-child InteractionsTeacher-child InteractionsWork-related Stressors −0.13 0.00 −0.23 ** 0.04 −0.06 −0.02
Professional Investment 0.01 0.07 −0.07 −0.24 ** −0.23 ** −0.14
Discipline and Motivation −0.13 −0.01 −0.16 −0.16 −0.19 * −0.21 *

Note. ES = Emotional Support. CO = Classroom Organization. IS = Instructional Support. Teacher-child Interactions (pre) = corresponding Teacher-child Interactions (Emotional
Support, Classroom Organization, Instructional Support) observed pre-midterm. * p < 0.05, ** p < 0.01, *** p < 0.001.
286 L.E. Sandilos et al. / Early Childhood Research Quarterly 42 (2018) 280–290

Table 4
Main effects of sources of stress and moderation effects of intervention to predict change in teacher-child interaction quality.

Emotional Support Classroom Organization Instructional Support

Model 1 Model 2 Model 1 Model 2 Model 1 Model 2

Teacher Race (Black) −0.05 (0.06) 0.06 (0.08) 0.11 (0.08)

Teacher Ethnicity (Hispanic) 0.01 (0.06) 0.06 (0.07) 0.02 (0.07)
Teacher Annual Income 0.01 (0.06) 0.09 (0.06) 0.03 (0.10)
Teacher Years of Education 0.05 (0.14) 0.00 (0.09) 0.08 (0.12)
Teacher Years of Experience 0.08 (0.06) 0.05 (0.05) −0.05 (0.05)
Teaching Efficacy/Beliefs 0.06 (0.09) 0.15 (0.09) 0.05 (0.06)
Head Start Affiliation 0.07 (0.07) 0.00 (0.08) 0.00 (0.05)
Teacher-child Interactions (pre) 0.31 (0.06)*** 0.38 (0.09)*** 0.28 (0.05)***

Set 1 R2 = 0.16* R2 = 0.16* R2 = 0.23*** R2 = 0.23*** R2 = 0.20*** R2 = 0.21***

Intervention Condition (Course) 0.23 (0.06)*** 0.23 (0.06)*** 0.20 (0.06)** 0.20 (0.07)** 0.31 (0.07)*** 0.32 (0.07)***
Work-related Stressors 0.00 (0.07) 0.00 (0.09) 0.00 (0.09) 0.00 (0.10) 0.02 (0.06) −0.04 (0.11)
Work-related Stressors x Condition −0.01 (0.09) −0.01 (0.08) 0.08 (0.09)

Set 2 R2 = 0.16* R2 = 0.21** R2 = 0.24*** R2 = 0.26*** R2 = 0.20*** R2 = 0.25***

Intervention Condition (Course) 0.24 (0.06)*** 0.25 (0.05)*** 0.20 (0.07)** 0.21 (0.06)*** 0.32 (0.07)*** 0.33 (0.06)***
Professional Investment −0.06 (0.07) −0.30 (0.10)** −0.04 (0.09) −0.17 (0.13) −0.01 (0.05) −0.24 (0.10)*
Professional Investment x Condition 0.27 (0.06)** 0.15 (0.07)* 0.26 (0.06)***

Set 3 R2 = 0.17** R2 = 0.17* R2 = 0.24*** R2 = 0.24*** R2 = 0.20*** R2 = 0.22***

Intervention Condition (Course) 0.24 (0.05)*** 0.24 (0.05)*** 0.20 (0.06)*** 0.20 (0.06)*** 0.32 (0.07)*** 0.32 (0.06)***
Discipline and Motivation −0.10 (0.06) −0.15 (0.10) −0.07 (0.07) −0.12 (0.08) −0.04 (0.08) −0.17 (0.11)
Discipline and Motivation x Condition 0.06 (0.08) 0.05 (0.07) 0.16 (0.07)*

Note. Model 1 = Main effects model. Model 2 = Moderation model. Teacher-child Interactions (pre) = corresponding Teacher-child Interactions (Emotional Support, Classroom
Organization, Instructional Support) observed pre-midterm. Covariate estimates are presented from a model excluding all treatment condition, teacher stress, and interaction
predictors.Standardized estimates are reported with standard errors in parenthesis. *p < 0.05. **p < 0.01. ***p < 0.001.

Fig. 2. Interaction between Professional Investment Stress and Treatment Condition

to Predict Change in Instructional Support.
*p < 0.05. **p < 0.01. *p < 0.001.

Fig. 1. Interaction between Professional Investment Stress and Treatment Condition

to Predict Change in Emotional Support. teachers with higher professional investment stress showed fewer
*p < 0.05. **p < 0.01. *p < 0.001.
gains in instructional support in the control condition and greater
gains in the course condition, relative to teachers in their respec-
Simple slopes were then examined to determine the nature tive treatment groups who reported lower levels of professional
of the interactions. Results revealed a negative relation between investment stress.
reports of professional investment stress and observed emotional Additionally, simple slopes were not significant for either group
support for teachers in the control condition (ˇ = −0.30, p < 0.01), (course or control condition) for two of the other significant
but not for teachers in the course condition (ˇ = 0.07, p > 0.05; Fig. 1). two-way interactions: a) the interaction between treatment con-
This finding confirms a moderating effect of the treatment con- dition and teacher-reported stress in professional investment on
dition (course vs. control) on the relation between professional change in classroom organization and b) the interaction between
investment and change in teachers’ emotional support across the treatment condition and teacher-reported stress in discipline and
year. That is, teachers in the control condition who reported higher motivation on change in instructional support. Although the trends
professional investment stress made fewer gains in emotional sup- were similar, non-significant simple slopes suggest that while the
port relative to control teachers experiencing less professional relation between teacher-reported stress and change in teacher-
investment stress. This association was not present for teachers child interaction quality significantly differed across the course and
in the course condition. control conditions, the main effect of stress did not reach statistical
Similar to the findings for emotional support, teachers’ reports significance in either group.
of stress in professional investment related negatively to change
in their observed instructional support for the control condition 3.1. Robustness check
(ˇ = −0.24, p < 0.05). In contrast, for the teachers in the course con-
dition, the relation between stress and change in instructional Given the high inter-correlations among the three sources of
support was positive (ˇ = 0.11, p < 0.01; Fig. 2). In other words, stress (i.e., 0.56–.67), moderation analyses were also run using
 L.E. Sandilos et al. / Early Childhood Research Quarterly 42 (2018) 280–290 287

structural equation modeling (SEM) in order to verify the robust- to the start of the intervention, it was not possible to directly test
ness of the regression findings. Specifically, observed stress items whether or not the course actually reduced professional investment
were used to create the three latent stress factors, which were stress for teachers in the course condition. Therefore, an important
then used to predict the latent CLASS outcomes with treatment next step in examining professional development interventions,
condition as the moderator. The SEM results were consistent with such as the model used in this study, is to explore the intervention’s
the regression analyses, with significant moderation occurring direct impact on sources of stress at post-test.
between professional investment stress and treatment condition Professional investment stress at the start of the school year
for emotional support (b = 0.94, p < 0.05) and instructional support was associated with growth in instructional support in the course
(b = 1.04, p < 0.05).3 However, because interactions using latent pre- group. One aspect of professional investment stress reflected teach-
dictors in Mplus must be run with the ‘type = random’ function, the ers’ concerns about their own professional growth; it is possible
SEM models were unable to account for clustering by site. Thus, the that teachers who were high in professional investment stress prior
regression results were selected as the main findings given that to the start of the intervention had greater uptake of the inter-
those analyses were consistent with SEM and also robust to the vention. This finding was not anticipated. However, it is plausible
effects of clustering. that teachers’ heightened awareness and anxiety over their need to
develop professionally may have made them more responsive to an
4. Discussion intervention designed to improve practice. Still, it is not clear why
teachers in the course condition showed a significant positive asso-
The present study explored the moderating effect of partici- ciation between professional investment stress and instructional
pation in PD on the relation between job stress and change in support, but not for emotional support. One potential explanation
teacher–child interactions. Prior research indicated that teachers relates to the intervention’s focus on improving interactions that
who participated in the course (intervention) condition exhibited contribute to language and literacy skills. It is possible that teach-
improvements in emotional support, classroom organization, and ers elected to participate in the larger study if they needed more
instructional support (Hamre et al., 2012). Extending those find- assistance with content-specific instructional supports, which may
ings, the current results demonstrated that participation in the help to explain why there was a significant positive association
PD course significantly moderated the relation between one spe- between professional investment stress and instructional support
cific source of job stress and change in two domains of observed in the course condition. To unpack this further, more information
teacher–child interaction quality. is needed regarding teachers’ motivations for participating in the
An examination of interactions and simple slopes between larger study.
stress and treatment condition revealed that professional invest- Although trends suggested that discipline and motivation stress
ment was the only source of stress significantly associated with had a more deleterious influence on teacher–child interactions in
teacher–child interaction quality. Professional investment stress the control group than in the course group, the main effect did
refers to teachers’ general displeasure with their career, such as not yield statistical significance. Existing literature reveals that
not feeling emotionally or intellectually stimulated by their job, student behavioral issues (e.g., behavioral problems, attentional
lacking sufficient opportunities for professional development, and issues, low motivation) can degrade teacher-student relationships
perceiving low levels of control over job-related decisions. For the due to ongoing conflict in the classroom (Friedman-Krauss et al.,
teachers in the control condition, high professional investment 2014; Grayson & Alvarez, 2008; Kyriacou, 2001). However, in this
stress at the start of the school year related negatively to growth in sample, we see no statistically detectable relation between teach-
emotional and instructional support over the course of the year. ers’ frustration with discipline issues and declines in the quality of
The mechanisms behind this finding align with the burnout lit- teacher–child interactions.
erature. Perceptions of limited job growth and lack of supports Work-related stress, such as reporting that there is too much
for career development can lead to a reduced sense of personal work to complete or that the pace of the school day is too fast,
accomplishment, which has long been considered an important did not relate to change in teacher–child interactions in either
aspect of career burnout (Maslach & Jackson, 1981). As teachers’ the course or control condition. Although factors such as work-
sense of accomplishment or growth in their career diminishes, overload and time pressures have been linked to job stress and
they may also experience waning enthusiasm for their job and eventual burnout (Lambert et al., 2006; Maslach et al., 2001), it may
feelings of emotional burnout. This experience of stress and emo- be that these job characteristics are more closely tied to general
tional burnout will not only influence teachers’ ability to support career attitudes rather than day-to-day interactions with students.
others emotionally, but could impair their own motivation and cog- These findings suggest that teachers in this sample who experi-
nition, consequently weakening their ability to provide high quality enced high work-related stress did not interact differently with
instructional supports to students over the course of a school year students when compared to teachers reporting lower levels of
(Jennings & Greenberg, 2009). work-related stress.
In contrast, for the teachers receiving PD, professional invest- Yet another surprising finding was that none of the sources
ment stress did not relate to change in the quality of emotional of stress were significantly associated with changes in levels of
support. These findings suggest that PD participation buffered the classroom organization, which assesses teachers’ ability to manage
relation between professional investment stress and the teachers’ behavior, facilitate a productive classroom that maximizes learn-
ability to foster an emotionally supportive classroom. Given that ing time, and provide varied materials for learning. This finding
the course focused directly on improving interactions and provided is unexpected given that existing literature has identified links
outlets for discussing classroom concerns, it is possible that teach- between teacher stress and classroom management (Friedman-
ers strengthened their skills in emotional support to such an extent Krauss et al., 2014; Klassen & Chiu 2010) and has described this
that their interactions were less vulnerable to personal factors such relation in the context of a “burnout cascade” (e.g., Jennings &
as emotional state. Since teacher stress was measured only prior Greenberg, 2009). It is possible that the lack of association stems
from key differences in the domains being assessed. The dimen-
sions of classroom organization tap into aspects of teaching and
3
SEM output using the ‘type = random’ function in Mplus does not provide stan-
instruction that may already be established as classroom routines
dardized beta weights. Therefore, only unstandardized betas are reported in the (e.g., behavioral expectations, rapid transitions, classroom mate-
robustness check section. rials; Pianta, La Paro et al., 2008). In contrast, emotional support
288 L.E. Sandilos et al. / Early Childhood Research Quarterly 42 (2018) 280–290

involves warmth and sensitivity in interpersonal interactions and extend existing work by including normative measures of emo-
instructional support requires cognitive energy (e.g., concentra- tional health in an effort to generate comparisons with the general
tion, memory) to cultivate higher-order thinking, all of which tend population and by administering stress measures at pre and post
to be affected by stress (American Psychiatric Association, 2013; intervention.
Sapolsky, 1996). Consequently, it may be that some of the more rou-
tinized aspects of teaching measured in classroom organization are
4.2. Implications for research and practice
less sensitive to an individual’s day-to-day emotional functioning
than the emotionally- or cognitively-demanding aspects of teach-
The findings from this study have both empirical and prac-
ing. Moreover, the duration and latency of the measurement of
tical implications. Research has indicated that teacher–child
CLASS scores may not have been sufficiently long enough to detect
interactions are beneficial to students’ academic and emotional
changes or a “cascading effect” in these more routinized aspects of
development and that high levels of teacher stress may be detri-
teaching. Continued work exploring the association between stress
mental to classroom quality (e.g., Li-Grining et al., 2010). The
and instructional effectiveness is needed to further understand
present findings reveal that professional investment stress may
these relationships (or lack thereof).
be particularly salient to teacher–child interaction quality. Con-
These results are important for administrators and policy-
tinued research exploring the differential influence of sources of
makers who face challenging decisions on whether or not to adopt
stress on teacher functioning is needed to better understand the
professional development for teachers. Decision-makers should
ways in which teachers can be more effectively supported in their
weigh the advantage of professional development against the dis-
profession. Future studies of various early childhood PD models
advantage of adding a time-consuming activity to the lives of
should also consider teachers’ emotional well-being as an outcome
already burdened and busy teachers. Though specific to the NCRECE
of interest in addition to changes in teachers’ instructional practice.
professional development, these findings can inform decision-
In a practical context, this study has implications for train-
making. Engagement in this professional development appears to
ing within the early childhood education workforce. For example,
mitigate the effect of stress related to professional investment
rather than simply utilizing a quality rating system to evaluate
on teachers’ emotionally supportive interactions. The presence of
preschools, the rating tools could be used diagnostically to target
professional investment stress also may be a proxy of teachers’
professional development efforts. In light of the findings related
readiness to engage in PD aimed at improving instructional prac-
to professional investment stress, early childhood centers would
tice.
be well-served to provide more training and continuing educa-
tion opportunities to preschool teachers, particularly in areas of
4.1. Limitations and future directions
instruction that are causing them concern or stress. In particular,
preschools should consider PD and other experiences that allow
There are several limitations that warrant mention. The first
for conversations and feedback regarding classroom interactions
two limitations relate to teacher selection and the nature of the
(Garet et al., 2001) given that the opportunity to develop in this
intervention (Hamre et al., 2012). First, the teachers at each site
area appeared to positively affect teachers’ classroom practices and
volunteered to participate in this study. Given the voluntary nature
emotional functioning. In addition, preschool centers could bene-
of participation, we cannot determine the extent to which these
fit from surveying teachers at regular intervals to gauge aspects
volunteer teachers differed from non-participating teachers at the
of the work environment that may be causing stress. This type
different sites or from the larger population of pre-kindergarten
of school-based data collection would enable administrators to be
teachers in the United States. Additionally, the fact that teachers in
more responsive to instructional and contextual issues that arise in
the course condition were voluntarily taking the course may have
their centers, and it would also convey a message to teachers that
resulted in improved uptake of course material. Consequently, the
their emotional well-being is valued.
findings have limited generalizability beyond the current sample.
In conclusion, teachers’ emotional well-being demands atten-
Second, the control teachers were placed in a business-as-usual
tion as we explore ways to improve early childhood educational
condition and did not receive any alternative coursework as part
experiences. Continued consideration and exploration of the piv-
of the study. Future work is needed to examine whether or not
otal role that various sources of stress can play in teacher behavior is
coursework in a different area, such as programming focused
an important issue as we identify ways to improve developmental
on stress-reduction (e.g., mindfulness, wellness, social-emotional
outcomes for young children. Further, the potential buffering effect
learning; Greenberg et al., 2016; Roeser, Skinner, Beers, & Jennings,
that high quality PD can have on the relation between stress and
2012), still moderates the influence of stress on teachers’ interac-
teacher–child relationships may have great benefit for preschool
tions with students or if the coursework needs to be targeted to
education.
the outcome of interest (i.e., teacher–child interactions). Third, the
CLASS data collection timeline was such that there was variation
in teachers’ exposure to the course when they submitted CLASS References
videos (e.g., teachers may have submitted their first of four post-
test videos as early as the mid-point of the course). To ensure that Alkon, A., Ramler, M., & MacLennan, K. (2003). Evaluation of mental health
consultation in child care centers. Early Childhood Education Journal, 31, 91–99.
all teachers have roughly equivalent exposure to a PD intervention http://dx.doi.org/10.1023/B:ECEJ.0000005307.00142.3c
when examining outcomes of interest, future studies on this topic American Psychiatric Association. (2013). Diagnostic and statistical manual of
should strive to collect all pre- and post-data exclusively before and mental disorders: DSM-5. Washington, D.C: American Psychiatric Association.
Birch, S. H., & Ladd, G. W. (1997). The teacher-child relationship and children’s
after the PD has taken place. Lastly, the final limitations relate to
early school adjustment. Journal of School Psychology, 35, 61–80. http://dx.doi.
the measurement of stress. Teachers reported on stress through a org/10.1016/S0022-4405(96)00029-5
rating scale rather than a diagnostic instrument. Thus, we do not Brock, L. L., & Curby, T. W. (2014). Emotional support consistency and
teacher–child relationships forecast social competence and problem behaviors
know whether this sample of preschool teachers was more or less
in prekindergarten and kindergarten. Early Education and Development, 25,
stressed than the general population of preschool teachers or other 661–680. http://dx.doi.org/10.1080/10409289.2014.866020
working adults. In addition, stress was only measured at one time Burchinal, M., Zaslow, M., & Tarullo, L. (2016). Quality thresholds, features, and
point (pre-test) during this study. As a result, the main effect of the dosage in early care and education: Secondary data analyses of child outcomes.
Monographs of the Society for Research in Child Development, 81, 1–120.
intervention on stress reduction could not be examined and poten- Cabell, S. Q., DeCoster, J., LoCasale-Crouch, J., Hamre, B. K., & Pianta, R. C. (2013).
tial meditational processes could not be tested. Future studies can Variation in the effectiveness of instructional interactions across preschool
 L.E. Sandilos et al. / Early Childhood Research Quarterly 42 (2018) 280–290 289

classroom settings and learning activities. Early Childhood Research Quarterly, teacher supports: Associations with children’s mathematics and science
28, 820–830. http://dx.doi.org/10.1016/j.ecresq.2013.07.007 learning. Early Childhood Research Quarterly, 29, 586–599. http://dx.doi.org/10.
Chang, M. (2009). An appraisal perspective of teacher burnout: Examining the 1016/j.ecresq.2014.06.007
emotional work of teachers. Educational Psychology Review, 21, 193–218. Klassen, R. M., & Chiu, M. M. (2010). Effects on teachers’ self-efficacy and job
http://dx.doi.org/10.1007/s10648-009-9106-y satisfaction: Teacher gender, years of experience, and job stress. Journal of
Choi, J. Y., Castle, S., Williamson, A. C., Young, E., Worley, L., Long, M., et al. (2016). Educational Psychology, 102(3), 741. http://dx.doi.org/10.1037/a0019237
Teacher–child interactions and the development of executive function in Klassen, R. M., Foster, R. Y., Rajani, S., & Bowman, C. (2009). Teaching in the Yukon:
preschool-age children attending Head Start. Early Education and Development, Exploring teachers’ efficacy beliefs, stress, and job satisfaction in a remote
27, 1–19. http://dx.doi.org/10.1080/10409289.2016.1129864 setting. International Journal of Educational Research, 48, 381–394. http://dx.doi.
Collmann, J. (2012). Teacher stress and teacher?child interactions: an investigation of org/10.1016/j.ijer.2010.04.002
demographic, professional, and self-efficacy characteristics on the relationship Kyriacou, C. (2001). Teacher stress: Directions for future research. Educational
between teacher stress and teacher?child interactions. ProQuest Dissertations & Review, 53, 27–35. http://dx.doi.org/10.1080/0013191012003362 8
Theses Global. (1002360316). Retrieved from http://proxy.its.virginia.edu/ Lambert, R. G., O’Donnell, M., Kusherman, J., & McCarthy, C. (2006). Teacher stress
login?url=http://search.proquest.com/docview/1002360316?accountid=14678 and classroom structural characteristics in preschool settings. In R. G. Lambert,
Curbow, B., Spratt, K., Ungaretti, A., McDonnell, & Breckler, S. (2000). Development & C. J. McCarthy (Eds.), Understanding teacher stress in an age of accountability
of the child care worker job stress inventory. Early Childhood Research (pp. 105–120). Greenwich, CT: Information Age Publishing.
Quarterly, 15, 515–536. http://dx.doi.org/10.1016/S0885-2006(01)00068-0 Li-Grining, C., Raver, C. C., Champion, K., Sardin, L., Metzger, M., & Jones, S. M.
Darling-Hammond, L., Wei, R. C., & Andree, A. (2010). How high-achieving countries (2010). Understanding and improving classroom emotional climate and
develop great teachers. stanford center for opportunity policy in education – behavior management in the real world: The role of Head Start teachers’
research brief. pp. 1–8. [Retrieved from http://www.oup.hu/howhigh doug.pdf]. psychosocial stressors. Early Education and Development, 21, 65–94. http://dx.
Denham, S. A., Bassett, H. H., & Zinsser, K. (2012). Early childhood teachers as doi.org/10.1080/10409280902783509
socializers of young children’s emotional competence. Early Childhood LoCasale-Crouch, J., Kraft-Sayre, M., Pianta, R. C., Hamre, B. K., Downer, J. T., Leach,
Education Journal, 40, 137–143. http://dx.doi.org/10.1007/s10643-012-0504-2 A., et al. (2011). Implementing an early childhood professional development
Domitrovich, C. E., Gest, S. D., Gill, S., Bierman, K. L., Welsh, J. A., & Jones, D. (2009). course across 10 sites and 15 sections: Lessons learned. NHSA Dialog, 14,
Fostering high-quality teaching with an enriched curriculum and professional 275–292.
development support: The Head Start REDI program. The American Education Mashburn, A. J., Pianta, R. C., Hamre, B. K., Downer, J. T., Barbarin, O. A., Bryant, D.,
Research Journal, 46, 567–597. http://dx.doi.org/10.3102/0002831208328089 et al. (2008). Measures of classroom quality in prekindergarten and children’s
Downer, J. T., Pianta, R. C., Burchinal, M., Field, S., Hamre, B. K., LoCasale-Crouch, J., development of academic, language, and social skills. Child Development, 79,
Scott-Little, C. (2014). Coaching and coursework focused on teacher–child 732–749. http://dx.doi.org/10.1111/j.1467-8624.2008.01154.x
interactions during language/literacy instruction: Effects on teacher outcomes Maslach, C., & Jackson, S. E. (1981). The measurement of experienced burnout.
and children’s classroom engagement. Manuscript submitted for publication. Journal of Occupational Behaviour, 2, 99–113.
Early, D., Iruka, I. U., Ritchie, S., Barbarin, O. A., Winn, D.-M. C., Craw-ford, G. M., Maslach, C., Schaufeli, W. B., & Leiter, M. P. (2001). Job burnout. Annual Review of
et al. (2010). How do pre-kindergarteners spend their time? Gender, ethnicity, Psychology, 52, 397–422.
and income as predictors of experiences in pre-kindergarten classrooms. Early McLeod, J., Fisher, J., & Hoover, G. (2003). The key elements of classroom
Childhood Research Quarterly, 25, 177–193. http://dx.doi.org/10.1016/j.ecresq. management: Managing time and space, student behavior, and instructional
2009.10.003 strategies. Alexandria, VA: Association for Supervision and Curriculum
Fimian, M. J., & Fastenau, P. S. (1990). The validity and reliability of the Teacher Development (ASCD).
Stress Inventory: A re-analysis of aggregate data. Journal of Organizational Muthén, L. K., & Muthén, B. O. (1998). Mplus user’s guide (7th ed.). Los Angeles, CA:
Behavior, 11, 151–157. Muthén & Muthén., 1998–2012.
Friedman-Krauss, A. H., Raver, C. C., Neuspiel, J. M., & Kinsel, J. (2014). Child National Center on Quality Teaching and Learning. (2013). Improving teacher–child
behavior problems, teacher executive functions, and teacher stress in Head interactions: Using the CLASS in Head Start preschool programs. U.S. Department
Start classrooms. Early Education and Development, 25, 681–702. http://dx.doi. of Health and Human Services, Administration for Children and Families, Office
org/10.1080/10409289.2013.825190 of Head Start (Grant # 90HC0002). Retrieved from eclkc.ohs.acf.hhs.gov/hslc/
Garet, M. S., Porter, A. C., Desimone, L., Birman, B. F., & Yoon, K. S. (2001). What tta-system/teaching/docs/using-the-class.pdf
makes professional development effective? Results from a national sample of National Survey of Early Care, & Education. (2013). Number and characteristics of
teachers. American Educational Research Journal, 38, 915–945. early care and education (ECE) teachers and caregivers: Initial findings from the
Goddard, R. D., Hoy, W. K., & Hoy, A. W. (2004). Collective efficacy beliefs: National Survey of Early Care and Education (NSECE). OPRE Report # 2013-39..
Theoretical developments, empirical evidence: and future directions. Retrieved from. https://www.acf.hhs.gov/opre/resource/number-and-
Educational Researcher, 33, 3–13. characteristics-of-early-care-and-education-ece-teachers-and
Gooze, R. (2014). The health and well-being of early childhood educators: a need for Ozer, N., & Beycioglu, K. (2010). The relationship between teacher professional
compassion and commitment.. Retrieved from http://www.childtrends.org/the- development and burnout. Procedia-Social and Behavioral Sciences, 2(2),
health-and-well-being-of-early-childhood-educators-a-need-for-compassion- 4928–4932.
and-commitment/#sthash.XqpiXnFi.dpuf Pakarinen, E., Kiuru, N., Lerkannen, M., Poikkeus, Z., Siekkinen, M., & Nurmi, J.
Grayson, J. L., & Alvarez, H. K. (2008). School climate factors relating to teacher (2010). Classroom organization and teacher stress predict learning motivation
burnout: A mediator model. Teaching and Teacher Education, 24, 1349–1363. in kindergarten children. European Journal of Psychology of Education, 25,
http://dx.doi.org/10.1016/j.tate.2007.06.005 281–300. http://dx.doi.org/10.1007/s10212-010-0025-6
Greenberg, M. T., Brown, J. L., Abenavoli, & RM. (2016). Teacher stress and health Pianta, R. C., & Hamre, B. K. (2009). Conceptualization, measurement, and
effects on teachers, students, and schools. Edna Bennett Pierce Prevention improvement of classroom processes: Standardized observation can leverage
Research Center, Pennsylvania State University. Retrieved from. http://www. capacity. Educational Researcher, 38, 109–119. http://dx.doi.org/10.3102/
rwjf.org/en/library/research/2016/07/teacher-stress-and-health.html 0013189x09332374
Hagekull, B., & Hammarberg, A. (2004). The role of teachers’ perceived control and Pianta, R. C., & Steinberg, M. (1992). Teacher-child relationships and the process of
children’s characteristics in interactions between 6-year-olds and their adjusting to school. New Directions for Child and Adolescent Development, 1992,
teachers. Scandinavian Journal of Psychology, 45, 301–312. http://dx.doi.org/10. 61–80. http://dx.doi.org/10.1002/cd.23219925706
1111/j.1467-9450.2004.00409.x Pianta, R. C., DeCoster, J., Cabell, S., Burchinal, M., Hamre, B. K., Downer, J., et al.
Hamre, B. K., & Pianta, R. C. (2004). Self-reported depression in non-familial (2014). Dose–response relations between preschool teachers’ exposure to
caregivers: Prevalence and associations with caregiver behavior in child-care components of professional development and increases in quality of their
settings. Early Childhood Research Quarterly, 19, 297–318. http://dx.doi.org/10. interactions with children. Early Childhood Research Quarterly, 29, 499–508.
1016/j.ecresq.2004.04.006 http://dx.doi.org/10.1016/j.ecresq.2014.06.001
Hamre, B. K., & Pianta, R. C. (2005). Can instructional and emotional support in the Pianta, R. C., La Paro, K. M., & Hamre, B. K. (2008). Classroom assessment scoring
first-grade classroom make a difference for children at risk of school failure? system manual, K-3. Baltimore: Brookes Publishing Co.
Child Development, 76, 949–967. http://dx.doi.org/10.1111/j.1467-8624.2005. Pianta, R. C., Mashburn, A. J., Downer, J. T., Hamre, B. K., & Justice, L. (2008). Effects
00889.x of web-mediated professional development resources on teacher–child
Hamre, B. K., Pianta, R. C., Burchinal, M., Field, S., LoCasale-Crouch, J., Downer, J. T., interactions in pre-kindergarten classrooms. Early Childhood Research
et al. (2012). A course on effective teacher–child interactions: Effects on Quarterly, 23, 431–451. http://dx.doi.org/10.1016/j.ecresq.2008.02.001
teacher beliefs, knowledge, and observed practice. American Education Raver, C. C., Jones, S. M., Li-Grining, C., Zhai, F., Bub, K., & Pressler, E. (2011). CSRP’s
Research Journal, 49, 88–123. http://dx.doi.org/10.3102/0002831211434596 impact on low-income preschoolers’ preacademic skills: Self-regulation as a
Hamre, B., Downer, J. T., Jamil, F., & Pianta, R. C. (2012). Enhancing teachers’ mediating mechanism. Child Development, 82, 362–378. http://dx.doi.org/10.
intentional use of effective interactions with children: Designing and testing 1111/j.1467-8624.2010.01561.x
professional development interventions. Handbook of Early Education, Raver, C. (2004). Child care as a work support, a child-focused intervention, and a
507–532. job. In A. C. Crouter, & A. Booth (Eds.), Work-family challenges for low-income
Jennings, P. A., & Greenberg, M. T. (2009). The prosocial classroom: Teacher social families and their children (pp. 179–191). New York, NY: Psychology Press.
and emotional competence in relation to student and classroom outcomes. Rimm-Kaufman, S., & Hamre, B. (2010). The role of psychological and
Review of Educational Research, 79, 491–525. http://dx.doi.org/10.3102/ developmental science in efforts to improve teacher quality. Teachers College
0034654308325693 Record, 112, 2988–3023.
Kinzie, M. B., Whittaker, J. V., Williford, A. P., DeCoster, J., McGuire, P., Lee, Y., et al. Rimm-Kaufman, S. E., & Sawyer, B. E. (2004). Primary-grade teachers’ self-efficacy
(2014). MyTeacherPartner-Math/Science pre-kindergarten curricula and beliefs, attitudes toward teaching, and discipline and teaching practice
290 L.E. Sandilos et al. / Early Childhood Research Quarterly 42 (2018) 280–290

priorities in relation to the Responsive Classroom approach. The Elementary from. http://www2.ed.gov/about/inits/ed/earlylearning/files/ece-low-
School Journal, 104, 321–341. http://dx.doi.org/10.1086/499756 compensation-undermines-quality-report-2016.pdf
Roeser, R. W., Skinner, E., Beers, J., & Jennings, P. A. (2012). Mindfulness training von der Embse, N., Kilgus, S. P., Solomon, H., Bowler, M., & Curtiss, C. (2015). Initial
and teachers’ professional development: An emerging area of research and development and factor structure of the Educator Test Stress Inventory. Journal
practice. Child Development Perspectives, 6, 167–173. http://dx.doi.org/10.1111/ of Psychoeducational Assessment, http://dx.doi.org/10.1177/
j.1750-8606.2012.00238.x 0734282914548329
Sandilos, L. E., Cycyk, L. M., Hammer, C. S., Sawyer, B. E., Lopez, L. M., & Blair, C. Whitaker, R. C., Dearth-Wesley, T., & Gooze, R. A. (2015). Workplace stress and the
(2015). Depression, control, and climate: An examination of factors impacting quality of teacher–children relationships in Head Start. Early Childhood
teaching quality in preschool classrooms. Early Education and Development, 26, Research Quarterly, 30, 57–69.
1111–1127. http://dx.doi.org/10.1080/10409289.2015.1027624 Whitebook, M., Phillips, D., & Howes, C. (2014). Worthy work, STILL unlivable wages:
Sapolsky, R. M. (1996). Why stress is bad for your brain. Science, 273, 749–750. The early childhood workforce 25 years after the national child care staffing study.
http://dx.doi.org/10.1126/science.273.5276.749 Berkeley, CA: Center for the Study of Child Care Employment, University of
Schwarzer, R., & Hallum, S. (2008). Perceived teacher self-efficacy as a predictor of California. Retrieved from. http://cscce.berkeley.edu/files/2014/ReportFINAL.
job stress and burnout: Mediation analyses. Applied Psychology, 57, 152–171. pdf
http://dx.doi.org/10.1111/j.1464-0597.2008.00359.x Yoon, J. (2002). Teacher characteristics as predictors of teacher-student
Sheridan, S. M., Edwards, C. P., Marvin, C. A., & Knoche, L. L. (2009). Professional relationships: Stress, negative affect, and self-efficacy. Social Behavior and
development in early childhood programs: process issues and research needs. Personality: an international journal, 30, 485–493.
Early Education and Development, 20, 377–401. http://dx.doi.org/10.1080/ Yoshikawa, H., Weiland, C., Brooks-Gunn, J., Burchinal, M. R., Espinosa, L. M.,
10409280802582795 Gormley, W. T., et al. (2013). Investing in our future: The evidence base on
Stipek, D. (2006). No child left behind comes to preschool. The Elementary School preschool education. Foundation for Child Development. Retrieved from. http://
Journal, 106, 455–466. http://dx.doi.org/10.1086/505440 fcd-us.org/resources/evidence-base-preschool
Torgesen, J. K. (2002). The prevention of reading difficulties. Journal of School Zan, B., & Donegan-Ritter, M. (2014). Reflecting, coaching and mentoring to
Psychology, 40, 7–26. http://dx.doi.org/10.1016/S0022-4405(01)00092-9 enhance teacher–child interactions in Head Start classrooms. Early Childhood
Tschannen-Moran, M., & McMaster, P. (2009). Sources of self-efficacy: Four Education Journal, 42(2), 93–104. http://dx.doi.org/10.1007/s10643-013-0592-
professional development formats and their relationship to self-efficacy and 7
implementation of a new teaching strategy. The Elementary School Journal, 110, Zhai, F., Raver, C., & Li-Grining, C. (2011). Classroom-based interventions and
228–245. http://dx.doi.org/10.1086/605771 teachers’ perceived job stressors and confidence: Evidence from a randomized
Tschannen-Moran, M., & Woolfolk Hoy, A. (2001). Teacher efficacy: Capturing an trial in Head Start settings. Early Childhood Research Quarterly, 26, 442–452.
elusive construct. Teaching and Teacher Education, 17, 783–805. http://dx.doi. http://dx.doi.org/10.1016/j.ecresq.2011.03.003
org/10.1016/S0742-051X(01)00036-1 Zinsser, K. M., Bailey, C. S., Curby, T. W., Denham, S. A., & Bassett, H. H. (2013).
U.S. Department of Health and Human Services, Human Services, & U.S. Exploring the predictable classroom: Preschool teacher stress, emotional
Department of Education. (2016). High-quality early learning settings depend on supportiveness, and students’ social-emotional behavior in private and Head
a high-quality workforce: Low compensation undermines quality.. Retrieved Start classrooms. NHSA Dialog, 16, 90–108.