Sabol, Et Al (2018)

Child Development, March/April 2018, Volume 89, Number 2, Pages 556–576
Low-Income Ethnically Diverse Children’s Engagement as a Predictor of

School Readiness Above Preschool Classroom Quality
Terri J. Sabol Natalie L. Bohlmann
Northwestern University Montana State University
Jason T. Downer
University of Virginia
This study examined whether children’s observed individual engagement with teachers, peers, and tasks
related to their school readiness after controlling for observed preschool classroom quality and children’s base-
line skills. The sample included 211 predominately low-income, racially/ethnically diverse 4-year-old children
in 49 preschool classrooms in one medium-sized U.S. city. Results indicated that children’s positive engage-
ment with (a) teachers related to improved literacy skills; (b) peers related to improved language and self-
regulatory skills; and (c) tasks related to closer relationships with teachers. Children’s negative engagement
was associated with lower language, literacy, and self-regulatory skills, and more conflict and closeness with
teachers. Effect sizes were small to medium in magnitude, and some expected relations between positive
engagement and school readiness were not found.
Children’s readiness for school at kindergarten identifying an additional target for interventions to
entry forms the foundation for later success in improve the outcomes of low-income children.
school and beyond (Boivin & Bierman, 2013). Yet in There are a host of studies suggesting that once
the United States, there are large disparities in children start out behind, it is difficult for them to
school readiness based on children’s family back- catch up (Reardon, 2011). This “skills beget skills”
grounds. At school entry, low-income children are approach to understanding children’s development
almost a year behind in their language and literacy indicates that children’s early skills or behaviors,
skills compared with their high-income peers (Bern- such as literacy or self-regulation skills, are strongly
stein, West, Newsham, & Reid, 2014). There is related to later skills or outcomes (Duncan et al.,
strong evidence that high-quality preschool class- 2007; Knudsen, Heckman, Cameron, & Shonkoff,
rooms can help reduce income-based achievement 2006). Measuring children’s early skills in these
gaps (Barnett, 2011). However, children have very areas provides important screening information for
different experiences within these settings (Willi- children who are more or less likely to be ready for
ford, Maier, Downer, Pianta, & Howes, 2013). This school.
study explores the extent that low-income chil- The most popular and rigorously studied
dren’s individual engagement within preschool set- approach to promote low-income children’s early
tings relates to their development, potentially skills and later life chances are early childhood edu-
cation programs. Estimates suggest that about 60%
This article was supported, in part, by the National Institute of of 4-year-old children in families below the poverty
Child Health and Human Development and the Interagency line are enrolled in preschool, up from about 30%
Consortium on Measurement of School Readiness through Grant two decades ago (Barnett & Yarosz, 2007). Increas-
R01 HD051498, as well as by the Institute of Education Sciences,
U.S. Department of Education through Grant R305A060021 (both ingly, attention has turned to the elements of pre-
to the University of Virginia). Dr. Jason Downer is the author of school contexts that contribute to children’s
the inCLASS measure that is used in this study but does not development. Most research in this area primarily
financially profit from this intellectual property. The opinions
expressed are those of the authors and do not represent views of focuses on the contribution of the average
the National Institute of Child Health and Human Development
or the U.S. Department of Education.
Correspondence concerning this article should be addressed to © 2017 The Authors
Terri Sabol; Northwestern University; 2120 Campus Drive, Evan- Child Development © 2017 Society for Research in Child Development, Inc.
ston, IL 60208 Electronic mail may be sent to terri.sabol@north- All rights reserved. 0009-3920/2018/8902-0016
western.edu. DOI: 10.1111/cdev.12832
14678624, 2018, 2, Downloaded from https://srcd.onlinelibrary.wiley.com/doi/10.1111/cdev.12832 by Nat Prov Indonesia, Wiley Online Library on [17/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Children’s Engagement 557
classroom experiences in a preschool setting on primary mechanisms through which classrooms

child outcomes (Camilli, Vargas, Ryan, & Barnett, afford children opportunities to learn vocabulary
2010). This body of work consistently demonstrates and emergent literacy skills, become engaged in
that classrooms that are emotionally supportive, academics, develop social skills, and, ultimately,
well-organized, and cognitively stimulating are become ready for school (Eccles & Roeser, 1999).
associated with greater gains for children on aver- Drawing from this theory, researchers have focused
age in the preschool year compared with lower on the ways in which teachers promote learning for
quality classrooms (Mashburn et al., 2008; Sabol, children within early learning environments
Hong, Pianta, & Burchinal, 2013). (Downer, Sabol, & Hamre, 2010). Compelling evi-
Characterizing children’s average classroom dence suggests that classrooms with high-quality
experiences can inform broad, teacher-level inter- teacher–child interactions can play a critical role in
ventions to increase classroom quality. However, it promoting positive life trajectories for low-income
may miss the fact that individual children have dif- children (Hamre & Pianta, 2005; Mashburn et al.,
ferent experiences than the average student. In fact, 2008). The direct exchanges between children and
there is considerable variability in children’s teachers in the preschool setting are typically
engagement with teachers, peers, and activities in a described as proximal processes and argued to be
classroom setting, which may play a unique and the main features of programs and aspects of pro-
important role in individual development (Downer, gram quality that help to explain the effect of pre-
Booren, Lima, Luckner, & Pianta, 2010). Moreover, school participation on child outcomes (NICHD
although measuring children’s early skills on key Early Child Care Research Network, 2002).
domains (e.g., reading, math) using direct assess- On average, children acquire more knowledge,
ments or teacher report provides important infor- develop better social skills, and more consistently
mation on early performance, it does little to pay attention and regulate their behavior when
explain how children then engage in the preschool they are in classrooms where the teacher is warm,
classroom on a day-to-day basis (Curby, Downer, sensitive, and attuned to the needs of children; pro-
& Booren, 2014). vides a productive classroom with clear behavioral
Most previous research either examines the effect expectations; and promotes engagement in the con-
of classroom interactions or the role of individual text of cognitively stimulating, language-rich tasks
children’s engagement in the classroom on chil- (Williford et al., 2013). Children from economically
dren’s outcomes. Very few studies focus on interac- disadvantaged backgrounds particularly appear to
tions in the classroom at multiple levels (i.e., benefit from high-quality classrooms, where low-
interactions at the classroom level and the individ- income children make greater gains in high-quality
ual level) and relations to children’s development versus low-quality classrooms (Hamre & Pianta,
(for an exception, see Chien et al., 2010; Williford 2005). Moreover, targeting class-wide teacher–child
et al., 2013). In this study, we capitalize on recent interactions has been shown to improve children’s
advances in observational methods to characterize broader learning experiences and subsequent child
children’s individual engagement in the classroom outcomes (Raver et al., 2008).
context with teachers, peers, and tasks as well as
observed classroom quality. We explore whether
Children’s Individual Engagement and Relations to
individual children’s observed experiences are
School Readiness in Preschool
related to children’s school readiness skills beyond
children’s previous skills and observed classroom- According to development systems theory, class-
level processes. room-level interactions are inherently dynamic and
may, in part, depend on the needs or preferences of
an individual child (Pianta et al., 2011). For exam-
Classroom-Level Contributions to School Readiness in
ple, a teacher may present a rhyming game to pro-
Preschool
mote children’s phonological skills. The extent to
Developmental science highlights the importance which the child advances her skills depends on a
of temporal interactions among children and the host of classroom-level factors, such as whether the
key individuals in their lives for fostering learning teacher is encouraging of children and creates a
and development (Bronfenbrenner & Morris, 1998). warm, comfortable climate and whether the teacher
Based on the years of observational research in provides feedback to children if they are struggling
classroom environments, it is generally agreed that with the concept. Yet, it also depends on the child’s
children’s interactions with adults and peers are the own engagement in the activity, including whether
558 Sabol, Bohlmann, and Downer
the child is attuned to the teacher, stays focused on that the developmental-ecological approach of
the activity, and refrains from conflict with her observing children’s individual engagement helps to
peers (Downer, Booren, et al., 2010). Thus, a more advance the understanding of children’s behavior in
comprehensive theoretical framework on the ecol- routine, everyday classroom contexts where learning
ogy of school environments emphasizes the impor- occurs. Children’s individual engagement is a multi-
tance of the context in which child engagement dimensional construct that refers to children’s inter-
occurs and individual engagement within those action with three main components of the classroom
contexts (Bronfenbrenner & Morris, 1998; Chien environment: teachers, peers, and tasks (Fredricks,
et al., 2010). That is, beyond mere access, what Blumenfeld, & Paris, 2004; Williford et al., 2013).
young children actually do in preschool in terms of Most past research considers each of these domains
their engagement with classroom resources may separately in association with children’s school readi-
function as a key mechanism through which the ness skills and often does not account for other
preschool environment can establish a strong foun- domains’ relation with outcomes. In this study, we
dation for learning. Under this developmental-eco- include multiple domains of children’s engagement
logical framework, classroom and individual in the same model—positive engagement with teach-
interactions are related but are theorized to repre- ers, peers, and task, and negative engagement—and
sent distinct constructs. examine whether there is an association with four
Most past work on interactions in preschool set- school readiness domains: language skills, literacy
tings focus primarily on classroom-level interac- skills, self-regulation, and teacher–child relation-
tions, and very little is understood about the role of ships. Overall, we expect that engagement across all
children’s individual engagement and interactions domains could relate to all domains of school readi-
in the classroom above and beyond the broader ness. Below, we highlight areas in which we expect
context. There is growing recognition that individ- particularly strong associations.
ual children play a much more significant role in
shaping their own development than was previ-
Positive Engagement With Teachers
ously believed (Skinner & Belmont, 1993). Chil-
dren’s own initiative, active engagement, sustained We hypothesize that engagement with teachers
attention, and attunement in classroom-embedded will relate to the development of language and lit-
interactions are associated with heightened learning eracy skills, self-regulation, and the quality of
experiences within preschool settings and may con- teacher–child relationships. Children’s attunement
tribute to their own skill development in the pre- and positive interactions with their teachers help
school context (Curby et al., 2014). promote emerging language skills by providing
Notably, children’s engagement may be an impor- more opportunities for dyadic conversations and
tant precursor for developing more concrete skills exposure to rich language use (Justice, McGinty,
over time (e.g., literacy skills; Boivin & Bierman, Zucker, Cabell, & Piasta, 2013). Children who inter-
2013). In this study, we explore the extent to which act more positively with teachers often have more
children’s engagement in the classroom is associated individualized learning experiences that in turn
with children’s development in the preschool promote their acquisition of core content, which in
settings, beyond more typical measures of school preschool is often focused on emergent literacy
readiness (e.g., baseline literacy skills). Although skills (Palermo, Hanish, Martin, Fabes, & Reiser,
many of the components measured by children’s 2007). In addition, we posit that positive engage-
individual engagement, such as behavior control ment with teachers will relate to self-regulation
with tasks, may reflect conceptual overlap with more given that positive relationships with adults are the
traditional measures of school readiness, such as key context for children to develop self-regulation
measures of inhibitory control, we seek to test the skills (Vygotsky, 1986). Children who engage more
extent to which children’s observed engagement is positively with teachers may have more interactions
associated with school readiness skills above and with teachers aimed at fostering autonomy and
beyond their performance on more static measures self-directed learning, which in turn may help chil-
that may not take into account the interactional com- dren improve their own self-regulation skills (Liew,
ponent of these skills. We do acknowledge that there Chen, & Hughes, 2010).
may be a transactional relation among measures of Positive engagement with teachers may also pro-
school readiness skills and children’s observed vide increased opportunities for children to develop
engagement, which we cannot account for with our close relationships with their teachers (Liew et al.,
current measurement approach. However, we argue 2010). Although there is considerable conceptual
overlap between observed teacher–child interactions

Positive Engagement With Tasks
(either at the classroom or child level) and teacher–
child relationships (measured by teacher ratings and Children’s ability to persist in tasks, regulate
used as an outcome in this study), we do consider their behavior, and actively take the lead in their
them as distinct constructs. The pattern of moment- own learning is often related to academic and self-
to-moment interactions between children and teach- regulatory skills in the classroom (Blair & Diamond,
ers in classrooms forms the basis for teacher–child 2008). Children’s curiosity, motivation, and persis-
relationships. Children’s level of closeness or con- tence to participate and complete tasks help them
flict to their teacher then contributes to the develop- develop emotion regulation and autonomy (Fan-
ment of their working models of attachment and tuzzo et al., 2004). Children who are engaged with
relationships (Ahnert, Pinquart, & Lamb, 2006). tasks may have closer and less conflictual relation-
Children’s close relationships with their teachers ships with teachers given that they are more likely
are associated with improved academic performance, to be on task and demonstrate interest and initia-
lower externalizing behaviors, and better social skills tive in learning material (Hartz, Williford, & Koo-
from preschool through eighth grade (Ladd & Bur- men, 2015). They may also have more opportunities
gess, 2001; Pianta & Stuhlman, 2004). Thus, chil- to accrue new language and literacy skills through
dren’s closeness and conflict with teachers formed in sustained engagement with tasks or activities,
preschool represent important indicators of school although this may be a less consistent finding after
readiness, whereas children’s positive engagement accounting for engagement with teachers.
with teachers represents a process that contributes to
the development of children’s relational quality with
Negative Engagement With Teachers, Peers, and Tasks
their teachers. We posit that children’s observed pos-
itive engagement with teachers will be associated Children’s negative engagement with teachers
with children’s literacy and language skills at the (e.g., noncompliance or negative affect), peers (e.g.,
end of the preschool year, as well as the level of aggressive behavior) and tasks (e.g., lack of behav-
closeness and conflict with teachers, above and ioral control) may affect their ability to benefit from
beyond other domains of children’s engagement the social and instructional support in the class-
(e.g., task). We predict that this relationship contin- room (Ladd & Burgess, 2001; Pianta, 2001). Nega-
ues to hold after controlling for the quality of the tive engagement could lead to fewer opportunities
classroom environment and children’s baseline skills. to engage in activities that involve learning lan-
guage and literacy skills if children are engaged in
conflict with their teachers and peers or disengaged
Positive Engagement With Peers
with the classroom activity. In addition, negative
Children’s engagement with peers contributes to engagement could serve as a stressor that interferes
their development of language and self-regulation with children’s development of self-regulatory
skills. Children who are more engaged with peers skills, particularly if children’s behavior is misa-
may have more opportunities to participate in ligned with the expectations of the setting, creating
shared activities, such as dramatic play, where chil- a negative feedback loop that is associated with
dren need to use language to convey ideas and lower inhibitory skills and a less positive attitude
interpret responses (Fantuzzo, Perry, & McDermott, toward learning and reduced attention (Bulotsky-
2004). In addition, peers help to shape the behav- Shearer, Fernandez, Dominguez, & Rouse, 2011).
ioral and emotional regulations of children by mod- Because negative engagement encompasses chil-
eling self-control skills, providing emotional dren’s interactions across multiple individuals and
support, and motivating children to succeed in contexts, and could interfere with overall engage-
school (Riggs, Blair, & Greenberg, 2004). We ment, we hypothesize that negative engagement
hypothesize that children’s observed engagement may be related to poor school readiness develop-
with peers will be related to language and self-reg- ment across the multiple domains included in this
ulation after adjusting for classroom quality and study (language, literacy, self-regulation, and
previous performance in these domains. Although teacher–child relationships).
positive engagement with peers may relate to all
domains of school readiness, peer engagement may
The Current Study
be less of a robust predictor of teacher–child rela-
tionships, particularly when accounting for chil- In this study, we examine the extent to which
dren’s positive engagement with teachers. children’s engagement with teachers, peers, and
tasks is associated with their language and literacy describe the context within which child engagement
skills, self-regulation, and teacher–child relation- occurs, but a direct measure of children’s observed
ships at the end of the preschool year. Considerable individual engagement can help to fill in children’s
research has accrued regarding the relation among unique activities and behaviors within those con-
children’s positive engagement with teachers, peers, texts. We posit that observing classrooms through
and tasks with better academic and social function- two lenses—the global classroom environment and
ing, as well as the association between negative an individual child’s experiences in that environ-
engagement and less adaptive outcomes. However, ment—will provide a more comprehensive under-
most of this evidence is based on teacher report, standing of how children acquire new skills and
and engagement with multiple elements of the develop in preschool settings than previous meth-
classroom (e.g., teacher, peers, and tasks) is rarely ods have allowed (e.g., teacher report or checklists).
measured at the same time. We capitalize on a In addition, we explore the ways in which the mul-
recently developed observational tool, the Individu- tiple domains of children’s engagement (positive
alized Classroom Assessment Scoring System engagement with teacher, peers, task, and negative
(inCLASS), which uses real-time observational engagement) relate to school readiness, which is lar-
methods to measure children’s engagement within gely missing from previous studies (e.g., Williford
the classroom setting. The inCLASS opens up the et al., 2013).
opportunity to explore a relatively understudied Moreover, our sample consists primarily of low-
aspect of behavior within preschool settings, income children from racially and ethnically diverse
namely children’s engagement, and model the mul- backgrounds enrolled in state- and federally funded
tiple components of this engagement with teachers, preschool programs. This sample may be particu-
peers, and tasks simultaneously. We include all larly important to study given that low-income chil-
four domains (positive engagement with teachers, dren are more likely to enter kindergarten behind
peers, and tasks, and negative engagement) in the their peers and are increasingly enrolled in pre-
same model, which allows us to examine the school (Barnett & Yarosz, 2007). Our study helps to
unique association between each domain (e.g., add onto the understanding of how children’s own
teachers) while accounting for the other domains engagement in the classroom relates to their perfor-
(e.g., peers and tasks). mance over the preschool year.
Moreover, we explore whether the association
among children’s observed engagement and school
readiness holds after accounting for initial skills Method
and the broader observed classroom environment.
Participants
Very few past studies provide a full ecological per-
spective on classroom environment and account for Participants for this study were drawn from an
both individual-level interactions and the broader 18-month research project conducted by the
classroom interactions using observational methods. National Center for Research on Early Childhood
One exception is a study by Chien et al. (2010) that Education (NCRECE), which targeted improve-
used observational “snapshot” methods (observing ments in teacher–child interactions using indepen-
children’s behavior in 20-s intervals) to create pro- dent and combined course and coaching
files of children’s individual classroom engagement components. Recruitment for this project focused
in pre-kindergarten. They found that individual on early childhood education programs that served
classroom engagement predicted math, language, low-income children in 10 sites across the country.
and literacy skill gains over the preschool year even Recruitment took place in January 2008 and for a
after controlling for observed global classroom second cohort in January 2009, and data collection
quality (measured by the CLASS and the Early ended in 2011. Large-scale programs within a given
Childhood Education Rating System). In addition, locale (e.g., all preschool programs in a city) were
Williford et al. (2013) used a latent profile approach invited to participate, and then center directors and
to examine patterns of children’s engagement and teachers were contacted within each program or
found that children who were positively engaged in agency.
the classroom had greater gains in expressive In the NCRECE professional development study,
vocabulary, working memory, and inhibitory con- teachers were randomly assigned to two forms of
trol after controlling for classroom quality. professional development that focused on interac-
Collectively, these past studies suggest that mea- tions in the classroom. In Phase I, teachers were
sures of broad classroom quality may help to randomly assigned to a 14-week course or a control
condition. The course was designed to increase in- hour (10 min of observation, 5 min of coding) alter-
service teachers’ knowledge about the importance nating between selected children until the end of
of teacher–child interactions to help promote chil- the observation (approximately 4 hr), resulting in
dren’s learning and skill acquisition. After Phase I, eight inCLASS cycles per child (Downer, Booren,
teachers in each condition (i.e., course vs. control) et al., 2010). Classrooms were also observed with
were re-randomized into Phase II coaching sup- the CLASS in the winter on 2 separate days. Each
ports, which used teachers’ observed behaviors as day of classroom observation involved an average
the basis for feedback to promote teacher quality of three cycles over 1 hr (15 min of observation,
and in turn children’s school readiness skills. A full 5 min of coding for each cycle) for a total of six
description of the results of the intervention is pre- cycles across both days. Data collectors completed
sented in previously published studies (e.g., Hamre the direct assessments of children on a separate day
et al., 2012). from the classroom observations (conducted in fall
Teachers were then followed for an additional and spring of preschool year). For further informa-
year in which they received no additional services tion on data collector training and data collection
or materials (Phase III). This manuscript focuses on procedures, see Data S1.
data collected in Phase III of the NCRECE project
at one of the 10 sites, which was in one medium-
Measure of Observed Individual Children’s Engagement
sized city in the United States. We selected data
from this one site because a more extensive obser- Children’s individual engagement with teachers,
vational protocol of both children (e.g., multiple peers, and tasks within the preschool classroom was
observations across multiple days controlling for assessed using the inCLASS (Downer, Booren,
activity settings) and classrooms (e.g., multiple Hamre, Pianta, & Williford, 2011). The inCLASS
observations across multiple days across two time includes 10 dimensions: (a) positive engagement
points) was used compared with the other sites. with teachers; (b) communication with teachers; (c)
The total sample for this study was 211 children conflict with teachers; (d) sociability with peers; (e)
across 49 classrooms. These 49 classrooms were dis- assertiveness with peers; (f) communication with
tributed across 24 schools, with the number of peers; (g) conflict with peers; (h) engagement with
classrooms per school ranging from 1 to 8. Table 1 tasks; (i) self-reliance with tasks; and (j) behavior con-
presents the descriptive statistics for children and trol (see Table S1, for behavioral indicators for each
classroom characteristics, as well as their individual of the dimensions). Each dimension is rated during
engagement and classroom quality. Children in the each observational cycle on a scale of 1–7, with 1 or 2
study were predominately African American (49%), indicating the child is low on that dimension, 3–5
followed by Hispanic (40%), White (5%), and other indicating that the child is in the mid range, and 6–7
(5%). Children were 4.35 years old (SD = 0.32) at indicating the child is high on that dimension. The
baseline in the fall of the preschool year, and 35% only exception is for the two negative engagement
were English language learners. Children came dimensions—conflict with teachers and conflict with
from mostly low-income backgrounds, with 89% of peers—in which higher ratings indicate more conflict
children living at or below 200% of the poverty line or negative interactions.
(income-to-needs < 2.0), and 53% of mothers had a Previous psychometric work on the inCLASS
high school degree or less. Nearly all teachers in demonstrates the tool’s criterion and construct valid-
the sample had a bachelor’s degree at baseline ity, as well as its associations with children’s learning
(96%) and an average of 12 years of experience (Downer, Booren, et al., 2010; Williford et al., 2013).
(SD = 7.1). Twenty-six percent of the classrooms In addition, four conceptually and empirically valid
were Head Start, and 63% were in public schools, domains have been identified among samples with
with an average class size of 17 children. children from diverse backgrounds attending a vari-
ety of preschool settings. The four domains include
(a) positive engagement with teachers (positive
Procedures
engagement and communication with teachers); (b)
The normal protocol for the NCRECE project positive engagement with peers (sociability,
was to conduct 2 days of inCLASS observations in assertiveness, and communication with peers); (c)
the fall (which were 1–2 weeks apart) during the positive engagement with tasks (engagement and
activity the child happened to be participating in at self-reliance with tasks); and (d) negative classroom
time of observation. During each classroom visit, engagement (conflict with teachers and peers, and
data collectors conducted four 15-min cycles every behavior control reverse scored; Downer, Booren,
Table 1 Table 1
Descriptive Statistics for Children’s Characteristics, Engagement in the Continued
Classroom, and School Readiness, and Classroom Characteristics and
Quality (n = 211 Children; 49 Classrooms) N % M (SD)
N % M (SD) Teacher/classroom level

Teacher/classroom characteristics
Child level Race 49
Child characteristics Black 0.55
Boy 211 0.50 White 0.39
Age at baseline 211 4.35 (0.32) Hispanic/other 0.06
Race 208 Teacher has BA or above 49 0.96
Black 0.49 Teacher years of experience 49 12.06 (7.07)
Hispanic 0.40 Class size 49 16.61 (2.55)
White 0.05 Class income-to-needs 49 0.95 (0.41)
Other 0.05 Head start classroom 49 0.25
English language learner 198 0.35 Classroom in public school 49 0.63
Income-to-needs 185 0.88 (0.80) Observed classroom quality
Low income 185 0.89 Instructional support 49 2.59 (0.62)
(income-to-needs < 2.0) Classroom organization 49 5.25 (0.59)
Maternal education (years) 206 12.29 (2.29) Emotional support 49 5.36 (0.65)
Children’s observed engagement in classroom
Teachers 177 2.11 (0.58)
Peers 177 2.10 (0.54)
Assessment Scoring SystemTM- Prek (CLASSTM- Prek;
Tasks 177 3.73 (0.50)
Negative engagement 177 1.49 (0.30)
Pianta, La Paro, & Hamre, 2008). The CLASSTM
Children’s school readiness assesses the average classroom quality based on 10
Language skills dimensions that are composed of positive climate,
Receptive (fall) 198 36.33 (21.42) negative climate, teacher sensitivity, regard for stu-
Receptive (spring) 195 50.69 (19.87) dent perspective, behavior management, productiv-
Expressive (fall) 197 443.69 (29.63) ity, instructional learning formats, quality of
Expressive (spring) 195 457.10 (17.75) feedback, concept development, and language mod-
Literacy skills eling. The dimensions are each rated from 1 to 7,
Phonological awareness (fall) 161 12.16 (4.53) with higher scores indicating higher quality interac-
Phonological awareness (spring) 186 15.41 (5.77)
tions (the same coding scheme as the inCLASS).
Print knowledge (fall) 197 11.13 (9.63)
Research using the CLASSTM has provided evidence
Print knowledge (spring) 194 28.81 (9.31)
Self-regulation
supporting three domains: (a) emotional support
Inhibitory control (fall) 197 0.42 (0.30) (positive climate, negative climate reverse coded,
Inhibitory control (spring) 195 0.61 (0.34) teacher sensitivity, and regard for student perspec-
Approaches to learning (fall) 193 39.93 (6.80) tives); (b) classroom organization (behavior man-
Approaches to learning (spring) 196 41.48 (6.64) agement, productivity, and instructional learning
Teacher–child relationships formats); and (c) instructional support (quality of
Closeness (fall) 193 4.22 (0.69) feedback, concept development, and language mod-
Closeness (spring) 196 4.52 (0.52) eling). Numerous studies confirm the construct and
Conflict (fall) 193 1.53 (0.75) predictive validity of the CLASSTM with strong evi-
Conflict (spring) 196 1.51 (0.80)
dence on the relation among the three CLASSTM
domains and children’s learning and development
(Hamre & Pianta, 2005; Hamre et al., 2012).
et al., 2010; Williford et al., 2013). The twelve 15-min
cycles of inCLASS observations (conducted on three
separate days over 1–2 weeks in the fall) were aver- School Readiness Measures
aged within each domain for each child.
Language
The Peabody Picture Vocabulary Test, 3rd ed.
Measure of Observed Classroom Quality
(PPVT–III; Dunn & Dunn, 1997) measured chil-
The quality of teacher–child interactions in the dren’s receptive vocabulary by showing children a
classroom was measured using the Classroom card with four pictures and asking children to point
to a picture for an orally provided word. The raw

Teacher–Child Relationships
score of the PPVT–III was used. The Picture Vocabu-
lary subtest of the Woodcock–Johnson–III Psychoe- Children’s relationships in the classroom were
ducational Battery (WJ–III; Woodcock, McGrew, & measured by the Student–Teacher Relationship
Mather, 2001) was used to assess children’s expres- Scale (STRS; Pianta, 2001). The STRS is a 15-item
sive vocabulary, which asks children to name objects scale based on teacher report that examines teach-
in a series of pictures. The W-score of the WJ–III ers’ perspectives of their relationships with individ-
was used. The subscales of the PPVT–III and the ual children in their classroom, including scores for
WJ–III have high reliability with a Cronbach’s alpha conflict (the extent to which teachers perceive nega-
of .97 for the PPVT–III (Dunn & Dunn, 1997) and tive interactions and emotions with the child; seven
an alpha of .81 for the WJ–III Picture Vocabulary items; a = .89) and closeness (degree of warmth and
subscale among preschool-aged children (McGrew open communication with child; eight items;
& Woodcock, 2001). a = .84). Items were averaged within each subscale
and used for analysis.
Preliteracy
Covariates
Children’s preliteracy skills were assessed in the
areas of phonological awareness and print knowl- We included a series of child and classroom
edge using the Test of Preschool Early Literacy characteristics that were hypothesized to affect both
(TOPEL; Lonigan, Wagner, Torgesen, & Rashotte, children’s engagement in their classroom and their
2007). The phonological awareness subscale of the development and learning in the preschool year.
TOPEL focuses on children’s word elision and Parent-reported child-level characteristics included
blending skills. The Print Knowledge subscale of the child gender, race, income-to-needs, and mother
TOPEL measures knowledge of the alphabet, writ- education, as well as language of assessment. In
ten language conventions, and writing form. These addition, we controlled for class size and the per-
subtests have adequate internal consistency centage of children in the class who were at or
(a = .78–.89) and concurrent validity (.41–.43; Loni- below the poverty line based on income to needs.
gan, Keller, & Phillips, 2004; Lonigan et al., 2007). We also controlled for whether the preschool pro-
We used the raw scores for both of these subscales. gram was sponsored by state or local government
and was part of the public school system (class-
room is in public school = 1) and whether the pre-
Self-Regulation
school program received Head Start funding (Head
In this study, we focused on the components of Start = 1). In addition, we controlled for teachers’
self-regulation that could potentially be affected by Phase I treatment assignment (course or control)
children’s engagement in the classroom, including and Phase II treatment assignment (consultancy or
inhibitory control and approaches to learning. Inhibi- control).
tory control was assessed using the Pencil Tap test
(Smith-Donald, Raver, Hayes, & Richardson, 2007).
Analytic Strategy
This assessment asks children to inhibit their domi-
nant response and tap once when the assessor taps We tested whether children’s engagement with
twice and vice versa. A child’s score was computed teachers, peers, and tasks, as well as their negative
based on the percent of correct responses out of 16. classroom engagement, predicted school readiness
Percent of correct responses on the pencil tap has outcomes, while accounting for children’s perfor-
demonstrated good concurrent and construct mance in the fall of the preschool year (at baseline)
validity with other measures of inhibitory control and classroom quality. This allowed us to examine
(Smith-Donald et al., 2007). The Preschool Learning whether the inCLASS data were predictive of
Behaviors Scale (PLBS; McDermott, Green, Francis, & school readiness scores above and beyond class-
Stott, 2000) is a 24-item teacher rating scale of chil- room quality and fall school readiness scores. Data
dren’s approaches to learning in the classroom setting. were analyzed using robust maximum likelihood
The scale included three dimensions, competence (MLR) with Mplus 7.2 (Muthen & Muthen, 2012) in
motivation, attention persistence, and attitude toward which all school readiness outcomes (language
learning, which were averaged with higher scores skills, literacy skills, self-regulation, and teacher–
indicating more positive approaches to learning. In child relationships) were represented in a single
our sample, the scale had high reliability (a = .84). model that examined the relation among all four
domains of children’s engagement with school (range = .23–.74) and self-regulation and teacher–
readiness, after controlling for baseline performance child relationships were highly correlated (absolute
(fall school readiness scores) and observed instruc- value range = .01–.63), it was also the case that mea-
tional support, emotional support, and classroom sures were associated across domains. For example,
organization. The use of MLR arguably provides inhibitory control was more strongly correlated with
more accurate estimates of standard errors when measures of language and literacy (range = .30–.53)
observations are not normally distributed or are not than with teacher–child relationships (absolute value
fully independent (Muthen & Muthen, 2012). The range = .01–.17). We thus modeled all school readi-
multilevel nature of the data (children nested ness outcomes in a single model. Child and class-
within classrooms) was taken into account by using room demographic variables (gender, race, language
a sandwich estimator (type = complex command in of assessment, income-to-needs, mother education,
Mplus), a procedure that computes robust standard class size, percentage of class with income-to-needs
errors while clustering at the classroom level < 2.0, whether the program was sponsored by the
(Muthen & Muthen, 2012). Intraclass correlation public school system, and whether the program
coefficients (ICCs) in this study were .29–.37 for received funding from Head Start) were included as
language outcomes, .19–.30 for literacy outcomes, control variables in all models.
.20–.25 for self-regulation, and .22–.34 for teacher– Because our sample comes from classrooms with
child relationships. Children’s engagement (teacher, teachers who had previously participated in a two-
peer, task, and negative classroom engagement) phase professional development intervention, we
and classroom quality (instructional support, class- also include controls for each phase. All 211 teach-
room organization, and emotional support) vari- ers in this study were randomly assigned to Phase I
ables were centered at the grand mean. (97 course intervention teachers, 114 control teach-
The full sample size for this study was 211 chil- ers). In Phase II, all 211 teachers were re-rando-
dren. However, missing data did occur. Impor- mized into coaching versus control (102 coaching
tantly, there was no more than 15% of missing data intervention teachers and 109 control teachers). We
for any individual variable. Missing data were han- include controls for both Phase I and Phase II.
dled using full information maximum likelihood, Importantly, teachers were not receiving any inter-
which uses all available data for each case to esti- vention services in the preschool year in which the
mate each covariance increasing the statistical data for this study were collected.
power of estimated parameters to retain our full Model fit indices, including the comparative fit
sample of 211 children. In addition, the inclusion of index (CFI), the root mean squared error of approxi-
correlated variables (e.g., child sex and task engage- mation (RMSEA), and the standardized root mean
ment) as auxiliary variables in the analyses helped square residual (SRMR), were used to determine
to reduce bias (Enders, 2010), resulting in less whether the model being tested closely reproduced
biased estimates compared with other approaches the pattern of relationships found in the data. For
(e.g., listwise deletion, single imputation, or mean incremental fit indices (CFI), values above .90 signify
imputation) used to address missing data. We also good model fit; RMSEA values < .05 indicate close
examined the sensitivity of our results when we fit and values between .05 and .08 signify acceptable
limit our sample to only children who had the model fit; and for SRMR, an absolute measure of fit,
inCLASS (n = 177). values lower than .08 are generally considered a
We examined the relation among all four indica- good fit (e.g., Browne & Cudeck, 1992).
tors of children’s engagement—positive engagement
with teachers, peers, task, and negative engagement
—with a full set of child outcomes: (a) language (re-
Results
ceptive and expressive vocabulary) skills; (b) literacy
skills (phonological awareness and print knowl- We examined the relation among children’s
edge); (c) self-regulation (inhibitory control and observed engagement in the classroom and school
approaches to learning); and (d) teacher–child rela- readiness after accounting for the broader class-
tionships (closeness and conflict). Correlations room environment and children’s fall performance
among the measures of spring school readiness skills on school readiness measures. Bivariate correlations
ranged from .01 (approaches to learning and expres- indicated that the four dimensions of children’s
sive language) to .74 (receptive and expressive lan- engagement—positive engagement with teachers,
guage; see Table 2). Although it was the case that peers, task, and negative engagement—were mod-
language and literacy skills were highly correlated estly correlated (absolute value range r = .05–.59;
Table 2
Correlations Among Children’s Engagement in the Classroom, Classroom Instructional Quality, and Preschool Child Outcomes
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Children’s engagement in classroom

1. Teachers —
2. Peers .36** —
3. Tasks .39** .59** —
4. Negative engagement .13 .03 .32** —
Classroom quality
5. Instructional support .20** .10 .18* .07 —
6. Classroom organization .24** .04 .03 .02 .69** —
7. Emotional support .15* .07 .05 .04 .63** .81** —
Preschool child outcomes (spring)
Language
8. Receptive .31** .35** .16* .04 .09 .12 .02 —
9. Expressive .29** .32** .08 .08 .02 .01 .09 .73** —
Literacy
10. Phonological awareness .29** .31** .28** .26** .04 .08 .03 .64 .48** —
11. Print knowledge .04 .01 .06 .30** .30** .27** .13 .39 .22** .42** —
Self-regulation
12. Inhibitory control .13 .20* .16* .23** .11 .23** .09 .48** .28** .53** .42** —
13. Approaches to learning .07 .11 .07 .34** .01 .01 .01 .12 .04 .08 .29** .22* —
Teacher–child relationships
14. Closeness .17* .06 .15 .11 .004 .06 .001 .13 .09 .14 .17* .004 .21** —
15. Conflict .19* .22** .04 .43** .03 .04 .02 .002 .16* .05 .21** .16* .63** .09
*p < .05. **p < .01.

Children’s Engagement
565
see Table 2). The strongest correlations were among of standard deviations the raw score is from the
positive engagement with tasks and teachers grand mean. We also performed linear transforma-
(r = .39, p < .01) as well as positive engagement tions for several of our school readiness outcomes
with tasks and positive engagement with peers for ease of comparison across measures. For lan-
(r = .59, p < .01). Negative engagement was nega- guage and literacy measures, we performed a linear
tively correlated with positive engagement with transformation of the scores (e.g., we divided the
task (r = .32, p < .01), with positive, albeit small PPVT and PLBS by 10; divide the WJ–III by 100)
correlations with positive engagement with teachers for ease of comparison across measures. We did not
(r = .13, p > .05) and peers (r = .05, p > .05). The transform the teacher–child relationship scores (the
three dimensions of classroom quality—instruc- closeness and conflict subscales were based on
tional support, classroom organization, and emo- means of the items) or inhibitory control (percent
tional support—were highly correlated (range correct). A 95% confidence interval is presented in
r = .63–.81, p < .01). parentheses (in the table and the text following).
There were also interesting correlational patterns
among observed children’s engagement in the class-
Language
room and classroom quality, with modest associa-
tions overall. Children’s positive engagement with Overall, children’s positive engagement with
teachers was moderately correlated with the instruc- peers and negative engagement was related to
tional support, classroom organization, and emo- receptive language. More specifically, children’s
tional support in the classroom (range r = .15–.24, positive engagement with peers predicted higher
p < .05). Children’s positive engagement with peers receptive language, ES = .15 (.04, .26), p < .01. Chil-
and tasks were associated with slightly lower class- dren’s negative engagement with teachers, peers,
room quality (range r = .03 to .18), although the and task predicted lower receptive language skills,
magnitude of the correlations was small and the ES = .10 ( .19, .01), p < .01. Children’s positive
only significant association was between task engagement with teachers and task was not associ-
engagement and instructional support (r = .18, ated with children’s receptive or expressive lan-
p < .05). Children’s negative engagement was only guage skills. In addition, children’s positive
slightly correlated with classroom quality (range engagement with peers and negative engagement
r = .02–.07, all ps > .05). There was also a small to were not associated with expressive language skills.
moderate relation among children’s engagement in All findings were above and beyond children’s fall
the classroom and child outcomes in the spring of language skills, which were strongly associated with
the preschool year in expected directions (absolute children’s spring performance (effect size = .70–.80)
value range r = .02–.47; see table for p-values). and classroom quality, which was associated with
Classroom quality was also related to spring child children’s receptive language skills in mostly the
outcomes with somewhat smaller correlations (abso- expected direction, albeit the coefficients were often
lute value range r = .001–.30; see table for p-values). nonsignificant and modest in size (effect size of
classroom quality ranged from .06 to .12).
Observed Individual Children’s Engagement and School
Readiness Literacy
Table 3 presents the relation among the four Children with more positive engagement with
indicators of children’s engagement (positive their teachers had higher phonological awareness,
engagement with teachers, peers, task, and negative ES = .38 (.22, .54), p < .001, in the spring of the pre-
engagement) and school readiness outcomes, school year. Children’s engagement with teachers
including language and literacy skills, self-regula- was not associated with print knowledge. More-
tion, and teacher–child relationships, after account- over, children’s engagement with peers and tasks
ing for previous performance and classroom was not associated with either phonological aware-
quality. The fit indices for the final model were ness or print knowledge. Children’s heightened
generally good for CFI = .93 and SRMR = .03, with levels of negative engagement with teachers, peers,
slightly less fit for RMSEA = .09. All values pre- and tasks predicted lower phonological awareness,
sented below are from the regression model and ES = .27 ( .42, .12), p < .001, and lower print
presented as standardized effect sizes. knowledge in the spring of preschool, ES = .23
The inCLASS and CLASS predictors were stan- ( .37, .08), p < .01. All relations were above and
dardized using a z-score that represents the number beyond pretest scores (effect size range = .36–.39)
Table 3
Relation Among Children’s Observed Engagement and Language Skills, Literacy Skills, Self-Regulation and Teacher–Child Relationships in Preschool, Controlling for Observed Classroom Quality
and Previous Performance (n = 211)
Language Literacy Self-regulation Teacher–child relationships
Phonological Inhibitory Approaches

Receptive Expressive awareness Print knowledge control to learning Closeness Conflict
b (95% CI) b (95% CI) b (95% CI) b (95% CI) b (95% CI) b (95% CI) b (95% CI) b (95% CI)
Child engagement
Teacher .07 ( .05, .18) .02 ( .05, .10) .38 (.22, .54)*** .05 ( .20, .10) .02 ( .16, .20) 0 ( .18, .18) .09 ( .08, .26) .01 ( .18, .16)
Peer .15 (.04, .26)** .09 ( .02, .19) .14 ( .01, .30) .02 ( .14, .19) .17 (.01, .32)* .08 ( .27, .11) .11 ( .30, .35) .14 ( .06, .34)
Task .04 ( .15, .07) .03 ( .14, .08) .11 ( .29, .30) .03 ( .11, .19) .08 ( .28, .12) .03 ( .14, .20) .20* (.05, .35) .01 ( .24, .22)
Negative .10 ( .19, .01)** .07 ( .17, .03) .27 ( .42, .12)*** .23 ( .37, .08)** .22 ( .39, .06)** .24 ( .44, .03)* .20 (.05, .36)** .28 (.11, .44)**
engagement
Pretest .70 (.61, .79)*** .80 (.70, .91)*** .39 (.26, .52)*** .36 (.22, .51)*** .37 (.22, .52)*** .38 ( .24, .52)*** .41 (.27, .55)*** .36 (.22, .49)***
Classroom quality
Instructional .08 ( .03, .19) .12 (.01, .23)* .02 ( .13, .16) .18 (.03, .33)* .09 ( .27, .09) .04 ( .22, .14) .12 ( .32, .08) .12 ( .09, .34)
support
Emotional .06 ( .17, .05) .01 ( .13, .10) .02 ( .17, .13) .15 ( .35, .04) .14 ( .30, .01) .03 ( .26, .19) .15 ( .40, .11) .03 ( .18, .24)
support
Classroom .04 ( .08, .16) .02 ( .12, .07) .05 (.22, .11) .11 ( .11, .32) .32 (.10, .55)** .03 ( .24, .30) .19 ( .06, .50) .14 ( .38, .11)
organization
Note. Language skills were measured by the Peabody Picture Vocabulary Test, 3rd ed. (Receptive) and the WJ–III Picture Vocabulary subscale (Expressive). Literacy skills were
assessed based on phonological awareness and print knowledge using the Test of Preschool Early Literacy. Self-regulatory skills were measured by the pencil tap (Inhibitory con-
trol) and teacher report of approaches to learning measured by the Preschool Learning Behaviors Scale (Approaches to learning). For teacher–child relationships, teachers reported
on the closeness (Closeness) and conflict (Conflict) with the child. Effect sizes were based on standardized beta coefficients. Models control for child’s fall performance, gender, race,
language of assessment, income-to-needs, mother education, class size, % class income-to-needs < 2.0, whether the program was sponsored by the public school system, whether
the program received funding from Head Start, and intervention treatment assignment. Models also adjust for observed classroom quality across three domains: emotional support,
classroom organization, and instructional support. Model 1 fit: comparative fit index = .93, standardized root mean square residual = .03, root mean squared error of approxima-
tion = .09. *p < .05. **p < .01. ***p < .001.
Children’s Engagement
567
and classroom quality, where instructional support Encouragingly, for our main predictors of class-
was predictive of print knowledge, ES = .18 (.03, room engagement, there were no differences in the
.33), p < .05, with nonsignificant relations among pattern of significant findings. The only significant
other classroom domains and literacy outcomes. change was that instructional support no longer
predicted print knowledge, ES = .17, ( .02, .32),
p = .054, n = 177.
Self-Regulation
Children’s positive engagement with peers and
negative engagement with teachers, peers, and
Discussion
tasks were associated with inhibitory control, which
was above and beyond a significant association This study examined the extent to which young
between classroom organization and inhibitory con- children’s engagement was related to their own
trol, ES = .32 (.10, 55), p < .01, and controls for learning in preschool settings above and beyond
other domains of classrooms quality and pretest classroom quality and their skills at the start of the
scores. Children’s positive engagement with their preschool year. Among a predominately low-
peers was related to greater inhibitory control, income, racially and ethnically diverse sample, we
ES = .17 (.01, .32), p < .05. Children’s heightened found that children’s positive engagement with
levels of negative engagement predicted lower inhi- teachers was related to improved literacy skills,
bitory control, ES = .22 ( .39, .06), p < .01. Chil- children’s positive engagement with peers was
dren’s positive engagement with teachers or task related to improved language and self-regulatory
was not associated with inhibitory control. Chil- skills, and children’s positive engagement with
dren’s negative engagement in the classroom was tasks was related to closer relationships with teach-
associated with lower levels of approaches to learners. Children’s negative engagement in the class-
ing, ES = .24 ( .44, .03), p < .05, which reflected room was associated with lower language, literacy,
teachers’ report of children’s motivation, attention, and self-regulatory skills and more conflict and
and attitudes toward learning. Children’s positive closeness with teachers. The results of this study
engagement (with teachers, peers, and task) was highlight the importance of observing children’s
not associated with approaches to learning. engagement in the preschool environment to better
understand how children advance their school
readiness skills.
Teacher–Child Relationships
The relation among teacher–child relationships
Relations Among Children’s Individual Engagement and
with children’s positive task engagement and nega-
School Readiness
tive classroom engagement revealed an interesting
pattern of findings. Children who exhibited more Overall, children’s engagement with teachers
positive engagement with tasks were reported as was associated with increases in children’s literacy
having closer relationships with teachers, ES = .20 skills, in particular phonological awareness, over
(.05, .35), p < .05. In addition, children’s observed the preschool year (ES = .38) after accounting for
negative engagement predicted higher levels of con- their baseline performance. To put this effect size in
flict with teachers, ES = .28 (.11, .44), p < .01. Yet, context, it is equivalent to approximately a third of
unexpectedly, negative engagement also predicted a year of learning in preschool and is about one-
higher levels of closeness with teachers, ES = .20 fourth of the size of the achievement gap between
(.05, .36). Children’s engagement with teachers and high-income and low-income children in kinder-
peers was not associated with teachers’ report of garten (Reardon, 2011). Our results suggest that
closeness and conflict with children. All findings greater engagement with teachers may be associ-
were above and beyond pretest scores (effect size ated with children’s ability to decipher code-based
range = .36–.41) and classroom quality (effect size phonological structures and advance their phono-
range = .15– .19). logical awareness (Justice et al., 2013; Wasik &
Hindman, 2011).
Past work suggests that classroom environments
Sensitivity Test
play an important role in supporting children’s
We also tested whether the overall pattern of phonological awareness (Wasik & Hindman, 2011).
results was consistent when we limited the sample In particular, children’s code-related skills are pro-
to 177 children who had inCLASS data. moted when teachers provide meaningful exposure
to letters and sounds, including rhyming activities, past work that children who spend more time on
letter walks, joint writing activities, and shared task and demonstrate effortful engagement are typi-
book reading (Justice & Ezell, 2002). However, there cally viewed more positively by teachers (Hughes,
is less research on the role of individual children’s Luo, Kwok, & Loyd, 2008; O’Connor, 2010). Yet,
engagement within classroom environments. It may surprisingly, children’s negative engagement pre-
be the case that children who are positively dicted higher teacher-reported conflict as well as
engaged and have more communication with their higher teacher-reported closeness. We hypothesize
teacher may have more opportunities to engage in that children who are observed having more nega-
activities that promote their phonological develop- tive engagement with teachers may be more ori-
ment (e.g., shared writing experiences). However, ented toward the teacher in general, and teachers
future work may want to untangle the particular may monitor children more closely with whom
behavioral component of children’s engagement they have conflicts. This could lead children to have
that is driving this association (e.g., attunement to more opportunities for both positive and negative
teacher, sustained communication). relationships, and thus may help to explain why
Children’s engagement with peers was associ- negatively engaged children are reported to have
ated with development of their language skills and more conflict and closeness with teachers.
self-regulation over the preschool year. Children Overall, children who were negatively engaged
who are more engaged with peers may have in the classroom with teachers, peers, and tasks
increased opportunities for sustained and rich con- were at a comparative disadvantage in terms of
versation with classmates in both practical and their school readiness. Children with higher levels
social contexts, which may in turn help children of observed negative engagement exhibited lower
develop their language skills (Ladd, 2007). Accord- performance across nearly all of the academic, lan-
ing to sociocultural theory, children’s peer talk can guage, and social domains in our study. This con-
provide opportunities to learn from one another firms previous research that children who exhibit
(Vygotsky, 1986). Children who engage more with more conflict with teachers and peers, and who
their peers may need to work harder to initiate and have lower levels of behavioral control are at risk
sustain language interactions compared with for lower school readiness and represent an impor-
teacher–child interactions, which may help children tant group of children who are in need of support
more quickly learn how to understand language (Williford et al., 2013).
(Cekaite, Blum-Kulka, Grøver, & Teubal, 2014). Although our results were generally aligned with
Children’s engagement with peers was associ- our hypotheses, there were several associations
ated with increases in inhibitory control. between children’s engagement and child outcomes
Although inhibitory control is typically posited that we expected but did not find. One explanation
to capture both executive/cognitive and behavioral for the lack of hypothesized relations may be due to
control, our modeling approach helps to isolate the our analytic approach. We included controls for fall
behavioral component of inhibitory control by performance, which were strongly predictive of
adjusting for the variance that overlaps with lan- spring performance (effect sizes range = .36–.80). We
guage and literacy skills. The findings that positive also included controls for three dimensions of class-
peer engagement is related to higher inhibitory con- room quality—instructional support, emotional sup-
trol is aligned with previous research that interper- port, and classroom organization—and we included
sonal interactions with peers can help children all school readiness outcomes in one model, which
develop behavioral regulation (Gallagher, 1999). accounts for the correlations among spring out-
Children who have more positive engagement with comes. Our results on children’s engagement repre-
peers may have more opportunities to become sent the unique association with school readiness
exposed to the opinions, ideas, and feedback of after accounting for all of these important factors,
others, which may help them learn to regulate their which may have limited the variability available to
own behaviors (Ladd, 2007). predict and help to explain why we did not see some
We also found an interesting pattern of results associations where we may have expected them.
for predicting the quality of teacher–child relation- For example, observed engagement with teachers
ships, which may have been missed had we not did not predict teacher report of closeness and con-
measured children’s observed engagement in the flict with teachers. Fall levels of closeness and con-
classroom. As expected, children who were more flict were highly predictive of spring levels, which
engaged with tasks were reported as having closer is in line with previous research that past relational
relationships with teachers. This is aligned with patterns are generally good predictors of
subsequent relationships and that teachers’ report Children’s positive engagement with teachers may
of closeness and conflict, once established for a be more important for phonological awareness
child, are fairly stable throughout a school year compared with more basic skills like alphabet
(Ahnert et al., 2006). The type of activities and knowledge, which children may be able to acquire
interactions measured by the inCLASS measure- through other classroom activities (e.g., indepen-
ment tool to assess children’s engagement with dent work). In addition, past work suggests that
teachers may not capture the variation in teacher– the home literacy environment is related to print
child relationships that is unaccounted for by previ- knowledge but not phonological awareness
ous teacher report of closeness and conflict. In addi- (Senechal & LeFevre, 2002), suggesting that chil-
tion, it may be the case that the teacher-report dren’s engagement in school contexts may be par-
method used to measure teacher–child relationships ticularly important for more advanced preliteracy
is biasing the measure. Past work has found that skill accrual.
teachers’ judgments may include information that is In addition, it was surprising that children’s
unrelated to the competencies of a particular child engagement with teachers was not related to lan-
and instead reflect subjective biases in the way they guage development or teacher–child relationships
perceive the child’s behavior or skills (Duckworth given that children’s engagement with teachers
& Yeager, 2015). Thus, teacher–child relationships reflects their use of communication with teachers
may represent the teacher’s perceptions of the child, across varied purposes and the degree to which the
which may be colored by their general impressions child is emotionally connected to teachers (Roorda,
of the child rather than the child’s actual positive Koomen, Spilt, & Oort, 2011). For language devel-
engagement with the teacher on a given day, which opment, other domains of children’s engagement,
may help to explain the small, albeit significant cor- including positive engagement with peers and neg-
relations between observed teacher engagement ative engagement (which includes teachers, peers,
and teacher-reported closeness/conflict (correlation and tasks), were predictive of receptive language
range = .17–.19). skill accrual, highlighting the importance of peer
In addition, children’s engagement with task was talk for fostering language development (Cekaite
not related to self-regulation, which was unex- et al., 2014) and the factors that may put children
pected. Past research suggests that children who at risk for positive language development. Overall,
practice more in persisting at tasks and staying children’s engagement with teachers was only
engaged may have more opportunity to develop related to one (phonological awareness) of the eight
self-regulatory skills, emotion regulation, and school readiness outcomes. This suggests that chil-
autonomy (Blair & Diamond, 2008). The correla- dren’s engagement with teachers may not be as
tions between children’s engagement with task and strong of a predictor of children’s overall school
fall performance on self-regulatory tests were gener- readiness as are other domains (i.e., negative
ally modest (.15–.19), suggesting that pretest mea- engagement). This was not in line with our general
sures of these skills and observed engagement with theory regarding the importance of child–teacher
tasks are measuring different constructs. It may be interactions for promoting the acquisition of skills
the case that there was limited variation left over (Bronfenbrenner & Morris, 1998). However, results
after accounting for fall measures of self-regulation from other domains suggest alternative child-level
and other measures of children’s engagement. In interactions that are key for development and indi-
particular, the model included children’s negative cate that negative engagement across contexts may
engagement—typified by a lack of behavioral con- be a particularly relevant predictor of children’s
trol—which may be a more salient predictor of chil- development over the preschool year.
dren’s self-regulation than positive engagement
with tasks. For instance, Bulotsky-Shearer and Fan-
Importance of Dual Observational Lens on Both
tuzzo (2011) found that children with more nega-
Classrooms and Individual Children
tive engagement with tasks (based on teacher
report) were related to lower literacy outcomes in This study included both measures of children’s
preschool, kindergarten, and first grade. observed individual engagement and classroom-
There also may be more conceptual reasons that level interactions. Correlations among individual
we did not find expected relations. engagement and classroom quality were quite
For example, we did not find relations between small, suggesting that the two measures capture rel-
children’s engagement with teachers and print atively different components of the classroom envi-
knowledge, only with phonological awareness. ronment. Importantly, the quality of classroom-level
interactions in our sample was generally consistent preschool day. Results suggested that classrooms
with mean level averages across the country (based typified by high levels of emotional support and
on over 4,000 classrooms), with moderate to high classroom organization were related to more posi-
quality on emotional support and classroom organi- tive child-level engagement. In addition, support
zation, and lower quality on instructional support for a bidirectional association were found in which
(Hamre et al., 2014), supporting the generalizability children’s individual engagement was in turn asso-
of our findings. ciated with classroom-level emotional support and
Classroom quality was associated with children’s classroom organization.
language, preliteracy, and self-regulatory skills (but Future research may use this type of analytic
not teacher–child relationships), as is generally approach to explore the unidirectional and bidirec-
aligned with previous research (Burchinal, Vander- tional relations among classroom quality and chil-
grift, Pianta, & Mashburn, 2010; Hamre & Pianta, dren’s individual engagement, and subsequent
2005; Mashburn et al., 2008; Wasik & Hindman, implications for child development. Williford et al.
2011). However, in our study that controlled for (2013) did examine the extent to which classroom
previous performance and children’s observed quality strengthened or weakened the relation
engagement, effect sizes were generally small and between children’s engagement and outcomes, and
not consistent within across the school readiness found that children’s gains in expressive language
domains. These results suggest that only focusing were moderated by classroom quality, where chil-
on classroom quality, which is becoming increas- dren who were typically engaged made equivalent
ingly common in preschool settings, may miss gains to children who were more positively
important processes that are related to children’s engaged in the context of high-quality learning.
learning. These results suggest that children’s developmental
The importance of children’s engagement above gains in the classroom may be related to both the
and beyond classroom quality is consistent with individual child’s engagement and behaviors as
results from the only other studies, to our knowl- well as the type of experiences that the classroom
edge, to include both classroom-level and individ- affords. However, more nuanced measurement of
ual-level processes (Chien et al., 2010; Williford classrooms and children’s individual engagement
et al., 2013). Different from our study, Williford could further advance this area of research.
et al. (2013) combined the individual domains of There are a number of ways children’s individual
children’s engagement (teacher, peers, and task) engagement and classroom quality may interact to
into one measure of observed engagement and produce developmental outcomes. For example, it
found relations among individual positive engage- may be the case that certain children’s engagement
ment and gains in language and executive function- patterns fit well with the classroom quality environ-
ing skills, after controlling for observed classroom ment and even allow children to benefit from low
quality. Our studies collectively suggest that chil- to average quality classrooms. In addition, children
dren’s observed engagement is an important ele- who may have low levels of observed engagement
ment to measure in classroom settings. may particularly benefit from classrooms that offer
We also acknowledge that this type of work emotionally supportive environments. Future stud-
cannot totally unpack the bidirectional relations ies that include measurement of both individual
among global classroom quality and children’s and classroom-level processes have the potential to
behavior. For example, high-quality programs may explore these types of complex relations and con-
lead to more positive children’s behavior, which in tribute to the broader understanding of how these
turn may affect subsequent classroom quality and processes operate together for certain groups of
children’s later behavior. This study suggests that children.
children’s engagement predicts school readiness
skills after controlling for observed classroom qual-
Limitations and Future Work
ity but does not explain how they operate in con-
cert over time. Emerging work has explored the There are several limitations of this study. First,
bidirectional behavioral exchanges between chil- this is a descriptive study on the relation among
dren and their environments (Curby et al., 2014; children’s engagement and their academic, lan-
Justice et al., 2013). For example, Curby et al. guage, and self-regulation skills during the pre-
(2014) used cross-lagged autoregressive models to school year. Although we control for child-level
test for associations between classroom-level inter- and classroom-level factors, including children’s
actions and children’s engagement within a performance in the fall of the preschool year that
accounts for children’s time-invariant characteris- among other samples of teachers across a variety
tics, we cannot be certain that this relation is not of settings among diverse populations of teachers
spurious or unaccounted for by another unmea- and children.
sured factor. In this study, we modeled children’s observed
Second, some of the effects may be due to mea- engagement as a predictor of school readiness over
surement error. We used this data set because the time. However, children’s observed engagement
study conducted multiple days of inCLASS and may be an important indicator of school readiness
CLASS observations (conducted on three separate in and of itself. Future work may treat observed
days in the fall and controlling for activity setting) engagement as an outcome and examine growth
and multiple days of CLASS observations (con- over the preschool year, as well as the extent to
ducted twice in the winter), each with multiple which observed engagement predicts longer term
cycles of observations. This data collection is much outcomes (e.g., performance in elementary school)
more comprehensive than previous studies have to further develop the role of observed engagement
afforded (e.g., Williford et al., 2013). However, it in the conceptual model for measuring and assess-
remains an empirical question regarding how many ing school readiness.
cycles of observation and how many days are nec- In addition, we included all four domains of the
essary to provide a reliable estimate of a child’s inCLASS simultaneously in our models—positive
behavior in the classroom at one point in the year engagement with teachers, peers, and task—and
and to determine stability across points in the year. negative engagement. This allows us to examine
Understanding the optimal number of cycles and the unique relation among each individual domain
days and exploring stability over time would and school readiness. However, this approach does
require much more complex statistical modeling not allow us to examine how the domains operate
and is needed in future work. In addition, we do bidirectionally to promote children’s engagement in
not have measures of classroom quality on the the classroom. It is an important question and a
same days in which children were observed so key area for future research to explore whether or
there may have been varying quality on that day how these environmental components work in com-
that may have influenced their behaviors. Future bination to produce developmental gains over the
work should consider the optimal setup to school year.
observe both children and classrooms to produce Although the inCLASS opens up opportunities
the most reliable results while taking into account to study children’s engagement across multiple
cost constraints and trying to reduce classroom domains, it is not without its limitations. Past psy-
disruption. chometric work supports the four domains modeled
Third, our sample may have limited generaliz- in the current article (Downer, Booren, et al., 2010).
ability. In particular, our study included children However, in this structure, positive engagement is
from primarily low-income backgrounds. Although measured separately for engagement with teachers,
we do not hypothesize that the associations would peers, and tasks, whereas negative engagement is
look different among a more heterogeneous sam- an overall measure across contexts. Grouping nega-
ple, we are only able to generalize to low-income tive engagement across teachers, peers, and tasks
children in Head Start or other preschool class- may limit the interpretation of the role of negative
rooms. Moreover, our findings may not generalize engagement, as it is unlikely that negative engage-
to other child-care settings, such as home- or fam- ment operates in the same way for teachers, peers,
ily-based care. In addition, teachers in our study or tasks. Future work may explore the extent to
were part of a professional development interven- which negative engagement with teachers, peers,
tion in years prior to data collection. Past results and tasks uniquely relate to children’s performance.
suggest that the professional development inter- Children’s individual engagement was coded at
vention led to increases in observed teacher quality the dimension level (e.g., positive engagement with
as well as child outcomes during the intervention teacher, teacher communication, and teacher con-
year (e.g., Hamre et al., 2012). Importantly, no flict) for each individual domain. Yet, as presented
intervention occurred during the year of our study in Table S1, each dimension is made up of several
(because we use data from the postintervention behavioral markers (e.g., attunement, shared posi-
year), and we control for intervention treatment tive affect, and initiatives communication). A limita-
assignment, which limits the effect of the intervention of using a more global, integrative coding
tion on our results. Nonetheless, future work approach is that we are not able to untangle the
should explore whether our findings are replicated role of individual behavioral markers, which could
inform intervention work. Future work may exam- supports based on children’s performance. Observ-
ine the utility of coding at a more granular level. ing children’s engagement has the potential to help
Last, in terms of the implications for future inter- educators enact more useful individualized strate-
vention work, the pattern of results regarding the gies that can help guide decisions about where,
positive domains of the inCLASS (teacher, peer, when, and how to intervene with at-risk children.
and task) to individual school readiness domains Although questionnaires or diagnostic tests are
does not lend itself to a clear intervention plan easier and more cost effective to administer, observ-
(e.g., focus on improving children’s engagement ing children’ engagement with peers, tasks, and
with teachers to improve literacy skills) given that teachers may have an added benefit given the
four of the eight outcome variables are associated child-centered approach that emphasizes real-time
with children’s engagement. However, children’s behaviors in classrooms.
negative engagement is associated with seven of The associations found in this study among chil-
the eight outcomes, suggesting that this domain dren’s observed engagement and children’s school
could be a possible target for intervention. The readiness suggest the utility of focusing on observed
dimensions for negative engagement do include engagement as the target of intervention work. One
behaviors across multiple contexts (teacher conflict, route to improve this engagement could be to
peer conflict, and lack of behavioral control with enhance teachers’ skills in observing child-level inter-
tasks). However, these dimensions and subsequent actions and their subsequent interactions with indi-
behavioral markers do provide a possible starting vidual children. Past work suggests that professional
point for exploring intervention work in this area. development activities, including coursework, indi-
In addition, it is an open question about whether vidualized coaching, and access to video exemplars,
this observation system could serve as an intervention can help teachers become better observers of interac-
tool for teachers. In this study, we did not test tions within their classroom and can help improve
whether the inCLASS was related to children’s devel- their own classroom-level teaching strategies (Hamre
opment over time above and beyond teacher reports et al., 2012; Jamil, Sabol, Hamre, & Pianta, 2015; Van
of children’s classroom behaviors or engagement, or Es & Sherin, 2002; Wasik & Hindman, 2011). The
whether teachers’ behaviors with students changed as same type of professional developmental activities
a function of learning more about the pattern of chil- could be applied using a more child-focused lens,
dren’s engagement reflected in the inCLASS. Estab- although there is not empirical work to support this
lishing the added value of the inCLASS, above and idea. Focusing on child engagement within class-
beyond typically used teacher questionnaires, is an room settings as a possible intervention tool needs
important avenue for future research. more attention to move the field forward. Overall,
our results do suggest that in order to understand
how classrooms relate to children’s development, it
Conclusion
is important to look beyond early skills and overall
This study explored the association among chil- classroom quality and capture children’s individual
dren’s engagement in the classroom and develop- experiences in classroom settings.
ment. Our results generally confirmed that
although children’s skills at the start of the pre-
school year are a strong predictor of where they
will end up, children’s individual engagement was References
associated with school readiness above and beyond Ahnert, L., Pinquart, M., & Lamb, M. E. (2006). Security
previous skills and the average quality of classroom of children’s relationships with nonparental care provi-
instruction. The findings highlight the benefit of ders: A meta-analysis. Child Development, 77, 664–679.
capturing children’s individual engagement with https://doi.org/10.1111/j.1467-8624.2006.00896.x
teachers, peers, and tasks within the broader class- Barnett, W. S. (2011). Effectiveness of early educational
room context. These classroom-embedded skills intervention. Science, 333, 975–978. https://doi.org/10.
1126/science.1204534
may serve as a key target in interventions designed
Barnett, W. S., & Yarosz, D. J. (2007). Who goes to preschool
to prepare low-income children for school.
and why does it matter? Preschool policy matters. New
In early childhood education settings, the most Brunswick, NJ: The National Institute for Early Educa-
common format to identify and intervene with tion Research.
young children who are at possible risk for positive Bernstein, S., West, J., Newsham, R., & Reid, M. (2014).
development is to test all children using a diagnos- Kindergartners’ skills at school entry: An analysis of the
tic assessment tool and then provide individualized ECLS-K. New York, NY: Mathematica Policy Research.
Blair, C., & Diamond, A. (2008). Biological processes in preschoolers’ competence in classroom interaction.
prevention and intervention: The promotion of self-reg- Early Childhood Research Quarterly, 25(1), 1–16.
ulation as a means of preventing school failure. Devel- https://doi.org/10.1016/j.ecresq.2009.08.004
opment and Psychopathology, 20, 899–911. https://doi. Downer, J., Sabol, T. J., & Hamre, B. (2010). Teacher–child
org/10.1017/S09545790800043 interactions in the classroom: Toward a theory of
Boivin, M., & Bierman, K. L. (Eds.). (2013). Promoting within-and cross-domain links to children’s develop-
school readiness and early learning: Implications of developmental outcomes. Early Education and Development, 21,
mental research for practice. New York, NY: Guilford. 699–723. https://doi.org/10.1080/10409289.2010.497453
Bronfenbrenner, U., & Morris, P. A. (1998). The ecology of Duckworth, A. L., & Yeager, D. S. (2015). Measurement
developmental processes. In W. Damon (Series Ed.) & matters assessing personal qualities other than cognitive
R. M. Lerner (Vol. Ed.), Handbook of child psychology: Vol. ability for educational purposes. Educational Researcher,
1. Theory (5th ed., pp. 993–1028). New York, NY: Wiley. 44, 237–251. https://doi.org/10.3102/0013189X15584327
Browne, M. W., & Cudeck, R. (1992). Alternative ways of Duncan, G. J., Dowsett, C. J., Claessens, A., Magnuson,
assessing model fit. Sociological Methods & Research, 21, K., Huston, A. C., Klebanov, P., . . . Japel, C. (2007).
230–258. https://doi.org/10.1177/0049124192021002005 School readiness and later achievement. Developmental
Bulotsky-Shearer, R. J., & Fantuzzo, J. W. (2011). Pre- Psychology, 43, 1428–1446. https://doi.org/10.1037/
school behavior problems in classroom learning situa- 0012-1649.43.6.1428
tions and literacy outcomes in kindergarten and first Dunn, L. M., & Dunn, L. M. (1997). Peabody Picture Vocab-
grade. Early Childhood Research Quarterly, 26(1), 61–73. ulary Test–third edition: Manual. Circle Pines, MN:
https://doi.org/10.1016/j.ecresq.2010.04.004 American Guidance Services.
Bulotsky-Shearer, R. J., Fernandez, V., Dominguez, X., & Eccles, J. S., & Roeser, R. W. (1999). School and community
Rouse, H. L. (2011). Behavior problems in learning influences on human development. In M. H. Bornstein &
activities and social interactions in Head Start class- M. E. Lamb (Eds.), Developmental psychology: An advanced
rooms and early reading, mathematics, and approaches textbook (4th ed., pp. 503–554). Mahwah, NJ: Erlbaum.
to learning. School Psychology Review, 40, 39–56. Enders, C. K. (2010). Applied missing data analysis. New
Burchinal, M., Vandergrift, N., Pianta, R., & Mashburn, York, NY: Guilford.
A. (2010). Threshold analysis of association between Fantuzzo, J., Perry, M. A., & McDermott, P. (2004). Pre-
child care quality and child outcomes for low-income school approaches to learning and their relationship to
children in pre-kindergarten programs. Early Childhood other relevant classroom competencies for low-income
Research Quarterly, 25(2), 166–176. https://doi.org/10. children. School Psychology Quarterly, 19, 212–230.
1016/j.ecresq.2009.10.004 https://doi.org/10.1521/scpq.19.3.212.40276
Camilli, G., Vargas, S., Ryan, S., & Barnett, W. S. (2010). Fredricks, J. A., Blumenfeld, P. C., & Paris, A. H. (2004).
Meta-analysis of the effects of early education interven- School engagement: Potential of the concept, state of
tions on cognitive and social development. The Teachers the evidence. Review of Educational Research, 74, 59–109.
College Record, 112, 579–620. https://doi.org/10.3102/00346543074001059
Cekaite, A., Blum-Kulka, S., Grøver, V., & Teubal, E. Gallagher, T. M. (1999). Interrelationships among chil-
(Eds.). (2014). Children’s peer talk: Learning from each dren’s language, behavior, and emotional problems.
other. Cambridge, UK: Cambridge University Press. Topics in Language Disorders, 19(2), 1–15.
Chien, N. C., Howes, C., Burchinal, M., Pianta, R. C., Hamre, B., Hatfield, B., Pianta, R., & Jamil, F. (2014). Evi-
Ritchie, S., Bryant, D. M., . . . Barbarin, O. A. (2010). dence for general and domain-specific elements of
Children classroom engagement and school readiness teacher–child interactions: Associations with preschool
gains in prekindergarten. Child Development, 81, 1534– children’s development. Child Development, 85(3), 1257–
1549. https://doi.org/10.1111/j.1467-8624.2010.01490.x 1274. https://doi.org/10.1111/cdev.12184
Curby, T. W., Downer, J. T., & Booren, L. M. (2014). Hamre, B. K., & Pianta, R. C. (2005). Can instructional
Behavioral exchanges between teachers and children and emotional support in the first-grade classroom
over the course of a typical preschool day: Testing bidi- make a difference for children at risk of school failure?
rectional associations. Early Childhood Research Quar- Child Development, 76, 949–967. https://doi.org/10.
terly, 29, 193–204. https://doi.org/10.1016/j.ecresq. 1111/j.1467-8624.2005.00889.x
2014.01.002 Hamre, B. K., Pianta, R. C., Burchinal, M., Field, S., LoCa-
Downer, J. T., Booren, L. M., Hamre, B., Pianta, R. C., & sale-Crouch, J., Downer, J. T., . . . Scott-Little, C. (2012).
Williford, A. (2011). The Individualized Classroom Assess- A course on effective teacher–child interactions effects
ment Scoring (inCLASS). Unpublished technical manual, on teacher beliefs, knowledge, and observed practice.
Curry School of Education, University of Virginia, American Educational Research Journal, 49, 88–123.
Charlottesville, VA. https://doi.org/10.3102/0002831211434596
Downer, J. T., Booren, L. M., Lima, O. K., Luckner, A. E., Hartz, K., Williford, A. P., & Koomen, H. (2015). Changes
& Pianta, R. C. (2010). The Individualized Classroom in teacher’s perceptions of the teacher–child relationship:
Assessment Scoring System (inCLASS): Preliminary children’s engagement and teacher’s characteristics. Manu-
reliability and validity of a system for observing script submitted for publication.
Hughes, J. N., Luo, W., Kwok, O. M., & Loyd, L. K. Muthen, L. K., & Muthen, B. O. (2012). Mplus: User’s
(2008). Teacher–student support, effortful engagement, guide (5th ed.). [Software manual]. Los Angeles, CA:
and achievement: A 3-year longitudinal study. Journal Author.
of Educational Psychology, 100(1), 1–14. https://doi.org/ NICHD Early Child Care Research Network. (2002).
10.1037/0022-0663.100.1.1 Child-care structure ? process ? outcome: Direct and
Jamil, F. M., Sabol, T. J., Hamre, B. K., & Pianta, R. C. indirect effects of child-care quality on young children’s
(2015). Assessing teachers’ skills in detecting and iden- development. Psychological Science, 13, 199–206.
tifying effective interactions in the classroom. The Ele- https://doi.org/10.1111/1467-9280.00438
mentary School Journal, 115, 407–432. O’Connor, E. (2010). Teacher–child relationships as
Justice, L. M., & Ezell, H. K. (2002). Use of storybook dynamic systems. Journal of School Psychology, 48, 187–
reading to increase print awareness in at-risk children. 218. https://doi.org/10.1016/j.jsp.2010.01.001
American Journal of Speech Language Pathology, 11, 17–29. Palermo, F., Hanish, L. D., Martin, C. L., Fabes, R. A., &
https://doi.org/10.1044/1058-0360(2002/003) Reiser, M. (2007). Preschoolers’ academic readiness:
Justice, L. M., McGinty, A. S., Zucker, T., Cabell, S. Q., & What role does the teacher–child relationship play?
Piasta, S. B. (2013). Bi-directional dynamics underlie the Early Childhood Research Quarterly, 22, 407–422.
complexity of talk in teacher–child play-based conver- https://doi.org/10.1016/j.ecresq.2007.04.002
sations in classrooms serving at-risk pupils. Early Child- Pianta, R. C. (2001). Student–teacher relationship scale. Lutz,
hood Research Quarterly, 28, 496–508. https://doi.org/ FL: Psychological Assessment Resources.
10.1016/j.ecresq.2013.02.005 Pianta, R. C., Hamre, B. K., & Allen, J. P. (2012). Teacher-
Knudsen, E. I., Heckman, J. J., Cameron, J. L., & Shonk- student relationships and engagement: Conceptualiz-
off, J. P. (2006). Economic, neurobiological, and behav- ing, measuring, and improving the capacity of class-
ioral perspectives on building America’s future room interactions. In S. L. Christenson, A. L. Reschly &
workforce. Proceedings of the National Academy of C. Wylie (Eds.), Handbook of research on student engage-
Sciences of the United States of America, 103, 10155– ment (pp. 365–386). New York, NY: Springer Science.
10162. https://doi.org/10.1073/pnas.0600888103 https://doi.org/10.1007/978-1-4614-2018-7_17
Ladd, G. W. (2007). Social learning in the peer context. In Pianta, R. C., La Paro, K. M., & Hamre, B. K. (2008).
O. N. Saracho & B. Spodek (Eds.), Contemporary perspec- Classroom Assessment Scoring System [CLASS] manual:
tives on socialization and social development in early child- Pre-K. Baltimore, MD: Brookes.
hood education (pp. 133–164). Charlotte, NC: Pianta, R. C., & Stuhlman, M. W. (2004). Teacher–child
Information Age. relationships and children’s success in the first years of
Ladd, G. W., & Burgess, K. B. (2001). Do relational risks school. School Psychology Review, 33, 444–458.
and protective factors moderate the linkages between Raver, C. C., Jones, A. S., Li-Grining, C. P., Metzger, M.,
childhood aggression and early psychological and Smallwood, K., & Sardin, L. (2008). Improving pre-
school adjustment? Child Development, 72, 1579–1601. school classroom processes: Preliminary findings from
https://doi.org/10.1111/1467-8624.00366 a randomized trial implemented in Head Start settings.
Liew, J., Chen, Q., & Hughes, J. N. (2010). Child effortful Early Childhood Research Quarterly, 23(1), 10–26.
control, teacher–student relationships, and achievement https://doi.org/10.1016/j.ecresq.2007.09.001
in academically at-risk children: Additive and interac- Reardon, S. F. (2011). The widening academic achieve-
tive effects. Early Childhood Research Quarterly, 25(1), ment gap between the rich and the poor: New evidence
51–64. https://doi.org/10.1016/j.ecresq.2009.07.005 and possible explanations. In R. Murnane & G. Duncan
Lonigan, C. J., Keller, K., & Phillips, B. M. (2004). Assess- (Eds.), Whither opportunity? Rising inequality and the
ment of children’s pre-literacy skills. In B. Wasik (Ed.), uncertain life chances of low-income children (pp. 91–116).
Handbook on family literacy: Research and services (pp. New York, NY: Russell Sage Foundation.
525–550). Mahwah, NJ: Erlbaum. Riggs, N., Blair, C., & Greenberg, M. (2004). Concurrent
Lonigan, C. J., Wagner, R. K., Torgesen, J. K., & Rashotte, and 2-year longitudinal relations between executive
C. A. (2007). TOPEL: Test of preschool early literacy. Aus- function and the behavior of 1st and 2nd grade chil-
tin, TX: Pro-Ed. dren. Child Neuropsychology, 9, 267–276. https://doi.
Mashburn, A. J., Pianta, R. C., Hamre, B. K., Downer, J. org/10.1076/chin.9.4.267.23513
T., Barbarin, O. A., Bryant, D., . . . Howes, C. (2008). Roorda, D. L., Koomen, H. M., Spilt, J. L., & Oort, F. J.
Measures of classroom quality in prekindergarten and (2011). The influence of affective teacher–student rela-
children’s development of academic, language, and tionships on students’ school engagement and achieve-
social skills. Child Development, 79, 732–749. https:// ment a meta-analytic approach. Review of Educational
doi.org/10.1111/j.1467-8624.2008.01154.x Research, 81, 493–529. https://doi.org/10.3102/
McDermott, P. A., Green, L. F., Francis, J. M., & Stott, D. 0034654311421793
H. (2000). Preschool Learning Behaviors Scale. Philadel- Sabol, T. J., Hong, S. S., Pianta, R. C., & Burchinal, M. R.
phia, PA: Edumetric and Clinical Science. (2013). Can rating pre-k programs predict children’s
McGrew, K. S., & Woodcock, R. W. (2001). Woodcock– learning? Science, 341, 845–846. https://doi.org/10.
Johnson III. Technical manual. Itasca, IL: Riverside. 1126/science.1233517
Senechal, M., & LeFevre, J. A. (2002). Parental involve- Williford, A. P., Maier, M. F., Downer, J. T., Pianta, R. C.,
ment in the development of children’s reading skill: A & Howes, C. (2013). Understanding how children’s
five-year longitudinal study. Child Development, 73, engagement and teachers’ interactions combine to pre-
445–460. https://doi.org/10.1111/1467-8624.00417 dict school readiness. Journal of Applied Developmental
Skinner, E. A., & Belmont, M. J. (1993). Motivation in the Psychology, 34, 299–309. https://doi.org/10.1016/j.ap
classroom: Reciprocal effects of teacher behavior and pdev.2013.05.002
student engagement across the school year. Journal of Woodcock, R. W., McGrew, K. S., & Mather, N. (2001).
Educational Psychology, 85, 571–581. https://doi.org/10. Test record. Form A. Woodcock–Johnson III tests of achieve-
1037/0022-0663.85.4.571 ment [assessment tool]. Itasca, IL: Riverside.
Smith-Donald, R., Raver, C. C., Hayes, T., & Richardson, B.
(2007). Preliminary construct and concurrent validity of
the preschool self-regulation assessment (PSRA) for Supporting Information
field-based research. Early Childhood Research Quarterly,
22, 173–187. https://doi.org/10.1016/j.ecresq.2007.01.002 Additional supporting information may be found in
Van Es, E. A., & Sherin, M. G. (2002). Learning to notice: the online version of this article at the publisher’s
Scaffolding new teachers’ interpretations of classroom website:
interactions. Journal of Technology and Teacher Education, Table S1. Individualized Classroom Assessment
10, 571–595. Scoring System (InCLASS) Dimensions, Definitions,
Vygotsky, L. S. (1986). Thought and language (Rev. ed., A. and Specific Behaviors for Measuring Observed
Kozulin, Ed. and Trans.). Cambridge, MA: MIT Press.
Children’s Engagement Across Teacher Interactions,
Wasik, B. A., & Hindman, A. H. (2011). Improving vocab-
Peer Interactions, and Task Orientation
ulary and pre-literacy skills of at-risk preschoolers
through teacher professional development. Journal of Data S1. Participants, Data Collection Proce-
Educational Psychology, 103, 455–469. https://doi.org/ dures, and Measurement Information
10.1037/a0023067

Sabol, Et Al (2018)

Uploaded by

Copyright:

Available Formats

Sabol, Et Al (2018)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sabol, Et Al (2018)

Uploaded by

Copyright:

Available Formats

Child Development, March/April 2018, Volume 89, Number 2, Pages 556–576

Low-Income Ethnically Diverse Children’s Engagement as a Predictor of

classroom experiences in a preschool setting on primary mechanisms through which classrooms

overlap between observed teacher–child interactions

N % M (SD) Teacher/classroom level

to a picture for an orally provided word. The raw

Children’s engagement in classroom

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

Language Literacy Self-regulation Teacher–child relationships

Phonological Inhibitory Approaches

You might also like