Introduction

Early childhood educators are balancing a range of challenging responsibilities throughout the workday: providing learning experiences to children, responding to and comforting children, taking care of children’s health and hygiene, curriculum planning, preforming administrative tasks, communicating with caregivers, and more (Harrison et al., 2019). These educators are not only expected to keep 3–5-year-old children safe, but also to be inclusive and prepare all children to be successful when they enter kindergarten. Additionally, they are expected to accomplish these goals when the overall workforce is less well-paid, has fewer years of formal education, and fewer opportunities for professional development (PD) compared to educators working in elementary grades (Phillips et al., 2016; Rhodes & Huston, 2012).

In the face of all of these factors and demands, early educators’ well-being is negatively affected, which then impacts the quality of the education and care they provide (Cumming, 2017). Data from a nationally representative sample of Head Start teachers indicated that one in three teachers experience some depression over the course of the school year (Hindman & Bustamante, 2019), with up to 70% of some samples experiencing subclinical levels of depression (Roberts et al., 2016). Educators who are not happy are more likely to leave their jobs, which partially explains why about a third of early childhood educators quit within a year (Bassok et al., 2021; Rhodes & Huston, 2012; Wells, 2015). Unfortunately, work-related stress has only increased since the beginning of the COVID-19 pandemic (Crawford et al., 2021). Educators’ well-being has major implications for the children in their care. Educators experiencing high levels of stress are more likely to have classrooms with lower emotional climates (Friedman-Krauss et al., 2014a, 2014b) and higher levels of exclusionary discipline practices, such as expulsion and suspension (Gilliam & Reyes, 2018; Gilliam & Shahar, 2006). Overall, preschool children experience expulsion at higher rates than children in any K-12 grade (Gilliam & Reyes, 2018; Gilliam & Shahar, 2006), and Black boys receive far more suspensions than their female and White peers (U.S. Department of Education, 2016). The concerning and disproportionate rates of exclusionary discipline practices may be attributed to educators' levels of stress when they do not have the tools to manage challenging behaviors in their program (Gilliam & Reyes, 2018) and implicit bias, where educators are more likely to expect Black children and particularly Black boys to display challenging behaviors (Gilliam et al., 2016b).

Early childhood educators’ stress levels have been consistently connected to the level of challenging behaviors present in their classroom (Clayback & Williford, 2021; Friedman-Krauss et al., 2014a, 2014b). Behavior challenges are common for children enrolled in early care and education programs, who are new to the classroom environment and need to learn how to function in this setting, such as sharing toys or supplies with other children and following instructions from adults. These children are also going through a period of social–emotional development where they are learning how to identify and manage their emotions and interact with their peers. On top of this, 14–30% of children entering early care and education programs are exhibiting problematic, challenging behaviors (Barbarin, 2007; Yoder & Williford, 2019), many of which are disruptive in the classroom. Because of this, classroom management and implementing positive behavioral supports are consistently identified as areas in which early childhood educators need additional training after they have begun working in a classroom setting (Garrity et al., 2019; Hemmeter et al., 2008; Snell et al., 2012). However, many early childhood educators report that they do not have access to high-quality PD opportunities that meet their and their classroom’s needs (Linder et al., 2016).

Positive Behavioral Supports and Classroom Management

Positive Behavioral Supports (PBS) consist of the systematic adoption of evidence-based practices that have been shown to prevent and address challenging behavior in early childhood care and education settings (Hemmeter et al., 2007, 2022). Several PBS models for addressing challenging behaviors and other educational difficulties in children break down strategies into primary (targeting all children), secondary (targeting specialized groups with at-risk behavior), and tertiary (targeting individual children with high-risk behavior) tiers (Bayat et al., 2010; Fox et al., 2010; Greenwood et al., 2011; Sugai & Horner, 2006). Primary strategies are the basis of other intervention efforts, as they establish behavioral expectations for children and provide the supports necessary to prevent children’s behavior from rising to a level requiring more targeted or intensive intervention. PBS are particularly important during early childhood, as they support children’s social-emotional development and address the challenging behaviors common for young children (Hemmeter et al., 2007). Historically, many of these PBS strategies have been referred to as classroom management techniques. Although there has been an increasing shift in referring to them as PBS, both terms are represented in the literature and therefore utilized in the present review.

Classroom management and positive behavioral support techniques, part of the primary, universal base of PBS models, are strategies educators employ to manage children’s behavior and promote a positive classroom climate. They can be implemented with the entire class (i.e., creating classroom rules and routines) or across various children as educators are working with their class (i.e., praising children for positive behaviors). When children are provided with these supports, they are better able to engage in the classroom and more time is spent learning (Duda et al., 2004; Jolstead et al., 2017). Limited work has been done on how behavioral supports implemented in early education impact later outcomes for children. However, in elementary schools, including those with preschool classrooms, behavioral supports have been associated with later positive academic, behavioral, and social–emotional outcomes for children (Korpershoek et al., 2016). Teachers implementing these strategies in elementary classrooms have also been shown to feel more efficacious in their roles and experience lower levels of burnout (Ross et al., 2012).

These classroom management and behavioral support strategies appear across several prevention and intervention models (e.g., BEST in CLASS; Conroy & Sutherland, 2008; Incredible Years; Webster-Stratton et al., 2008; and The Pyramid Model; Hemmeter et al., 2016) and often have various terminology referring to similar practices (e.g., premack or when-then statements). In order to identify the various practices that have been studied in the early childhood literature and facilitate communication about these practices, McLeod et al. (2017) developed a set of common practice elements of the strategies that have been used to improve social, emotional, and behavioral outcomes in early childhood settings. Based on their review and consultation with experts in early childhood education, they identified 24 discrete practice elements, which were then broken into two categories. First, content items were defined as general principles which guide practice and included explicit instruction on behavioral skills (e.g., emotion regulation) and approaches to working with children (e.g., promoting educator child-relationships). Second, delivery items were defined as the way an educator provides instruction to children, which included a range of strategies such as ways to reinforce positive behaviors (e.g., praise or tangible reinforcements), provide feedback on behaviors (e.g., instructive feedback and error correction), give instructions to children (e.g., opportunities to respond and premack statements), and positive discipline (time-out). Collectively these strategies are understood to comprise the classroom management strategies and positive behavioral supports early childhood educators can employ in their centers.

Professional Development in Early Childhood Settings

One way early childhood educators develop and enhance their skills in classroom management and behavioral supports is through PD. There has been increased national focus on expanding access to high-quality early childhood programs, and early childhood researchers have called for more work investigating how PD is delivered to staff, how educators’ skills, behaviors, and dispositions are impacted, and how these changes go on to impact children’s behaviors and academic outcomes (Dunst, 2015; Sheridan et al., 2009). The early childhood workforce is particularly diverse, in terms of their racial background, reasons for entering the early care and education workforce, level of formal education, years of experience on the job, and certifications (Phillips et al., 2016). This means PD efforts must support individuals with a range of experiences and expertise in reaching the high expectations of a quality early childhood program (Fuligni et al., 2009). This requires having an understanding of the processes that result in significant changes in educators’ practices and a range of PD opportunities to meet the needs of different schools, centers, and individual educators.

There are several traditional forms of PD: formal education, credentialing, specialized on-the-job training, coaching, consultation, and communities of practice (Sheridan et al., 2009). Training associated with formal education and credentialing typically occurs before an individual is employed at a school or center. In contrast, specialized training, coaching, consultation, and communities of practice occur amongst staff already working with children in the field. This paper will focus on the latter areas of PD. Specialized trainings, consultation, and coaching approaches are typically short-term and involve an expert assisting an educator to develop their skills. Specialized trainings include a wide range of activities such as workshops, conferences, video presentations, discussions, and tutorials that are meant to provide information that will impact educators’ practices. Coaching and consultation practices involve an individual educator partnering with a coach or consultant to improve skills and practices through observations, reflections, problem-solving, goal-setting, and feedback (Artman-Meeker et al., 2015; Brennan et al., 2008; Elk & Page et al., 2019; Fuchs & Fuchs, 1989; Gilliam et al., 2016a; Sheridan et al., 2006; Snyder et al., 2015; Shannon et al., 2021). Typically, coaching directly targets the implementation of specific practices or interventions, while consultation aims to address a concern or goal determined collaboratively with the intent of building the capacity of educators and programs (Brennan et al., 2008; Sheridan et al., 2009; Snyder et al., 2015). Finally, communities of practice are groups that are formed based on shared professional interests, where members are learning from each other by raising questions, brainstorming, reflecting, and providing feedback (Buysse et al., 2003; Clarkin-Phillips, 2011; Kuh, 2012). For a full discussion of different PD practices in the early care and education setting, please see the work of Sheridan et al. (2009).

Within each of these categories of PD, there is a large amount of variation. Lengths of specialized training or coaching relationships can range from a single, one-hour interaction, or be implemented over several hours across an entire school year. Workshops can be exclusively didactic or integrate elements of discussion or practice through role-playing. Training can occur at a school during the workday or at a separate location during evenings or weekends. A specific early childhood setting, such as Head Start, may be targeted, or practitioners from across highly variable schools or centers may come together for training. A range of different practices the educator can implement may be presented, or there may be an exclusive focus on a single skill. What different PD packages contain and how they are delivered impacts their uptake and the implementation of the skill they are designed to promote (Brunsek et al., 2020; Downer et al., 2009; Dunst, 2015; Korpershoek et al., 2016; Pianta et al., 2014). Schools and centers are looking for PD opportunities that fit within the constraints of their schedules and promote skills they view as adding to their goals of providing inclusive, high-quality care and education, and educators will find different modalities of training to be more or less helpful, particularly when translating those skills to the classroom.

Snyder et al. (2012) provide a review of all empirical literature related to early childhood PD, with a particular focus on understanding the factors associated with learning and implementing new skills and evidence-based practices in the classroom. They broke these factors down into the characteristics and contexts of the learners, the content of the PD, and the organization and facilitation of learning experiences. However, when schools or programs are selecting PD series to incorporate into their staff’s training, typically there is a specific need or set of skills they hope to fill or teach. Further, previous work has shown that PD series with specific objectives lead to larger effects on educator practices (Zaslow et al., 2010). Classroom management and providing positive behavioral supports have been identified as needed areas of continued PD for early childhood educators (Hemmeter et al., 2008; Snell et al., 2012). Accordingly, there is a need for a comprehensive understanding of what PD packages exist, what evidence-based techniques they target, and what they comprise, so centers can evaluate their options for PD and additional programs can be adapted or developed to fit the currently unmet needs of centers and educators as it relates to classroom management and implementing positive behavioral supports.

For this purpose, this study consists of a systematic descriptive review of the empirical literature on PD for early childhood educators targeting classroom management and positive behavioral supports. The aim of this review was to characterize the current PD approaches reported in the literature, as PD approaches and the corresponding literature are highly variable, with differing component parts that have been implemented in various settings. We had three primary objectives in conducting this review. First, we aimed to identify what PD approaches to enhancing classroom management and positive behavioral support skills exist and what practices are included in their content. Second, we aimed to describe the context and characteristics of how PD approaches were delivered. Third, we aimed to characterize the research designs utilized and educator and child outcomes reported in these studies. Providing this review of the topography of the PD literature in early childhood should enable scholars and practitioners alike to identify existing approaches that have been studied, as well as potentially overlooked or understudied areas in need of additional evaluation.

Method

Studies to be included in the review were identified as follows. Positive behavioral supports and classroom management skills are of interest to both educational and behavioral sciences, so the Education Resources Information Center (ERIC) and the American Psychological Association (APA) PsychNet databases were selected for the search to ensure the inclusion of studies across both fields. The first author conducted an electronic search using these two databases on 25 May 2021. A combination of four categories of search terms were used: (a) child age (early education, early childhood, Head Start, preschool, PreK); (b) target sample (teacher, instructor, professional, personnel, workforce); (c) training (inservice, coaching, instruction, training, development); and (d) training topic (classroom management, behavioral support, universal, prevention, tier 1). Search term truncation and Boolean operators were utilized to search for all versions of search terms and their possible combinations. See supplementary materials for exact search terms used. The abstracts of identified articles were also reviewed to identify additional articles to be included in the review.

Following the search, articles were evaluated using the following inclusion criteria. (a) The study evaluated methods of training early educators on classroom management skills, use of positive behavioral supports, or approaches to improve children’s problem and/or challenging behaviors at a universal or whole-classroom level; (b) The evaluated training was delivered to at least one staff member who was working at a school or early childhood center with preschool-aged children (i.e., 3–5 years old); (c) The study included at least one educator outcome measuring the use of classroom management or positive behavioral support strategies; (d) The training was evaluated using one of the following designs: randomized controlled trial or cluster-randomized trial; quasi-experimental group comparison design in which groups were matched or tested for differences; open trial; or single-case experimental designs (e.g., multiple baseline); (e) The report was written in the English language; (f) The report was published in a peer-reviewed journal. There were no inclusion criteria related to publication date. See Fig. 1 for the diagram based on PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines (Page et al., 2021).

Fig. 1
figure 1

PRISMA flow diagram of study inclusion procedure

Full size image

Of the 3108 articles identified by the original search, 212 were identified as duplicate records and 44 were excluded as non-English records. The first author reviewed the remaining 2852 records’ titles and abstracts for potential eligibility. Twenty percent of the records (572) were randomly selected for review by the second and third author to establish inter-rater reliability in determining potentially eligible articles. Any disagreement was resolved through discussion after initial coding. Based on this review, 174 studies were determined to be potentially eligible. The first author reviewed the full text of these articles to establish if these studies met inclusion criteria. Any studies in question were discussed with all authors. After eliminating articles that did not meet inclusion criteria, there was a final list of 47 included articles. These articles were then evaluated for unique samples, yielding the 42 unique studies included for review (Fig. 1). A list of excluded studies may be obtained from the first author.

Coding Procedures

The research team worked collaboratively to develop a coding system to evaluate what PD approaches are available and the classroom management and behavioral support strategies they teach, the context and characteristics of how the PD approach was delivered, and the research design utilized and outcomes reported.

PD Approaches and Included Classroom Management & Behavioral Support Strategies

As early care and education settings may be interested in bringing in an established PD approach that has been shown to impact educators’ use of classroom management and behavioral supports, the name of the PD approach under investigation was coded for each study. In order to evaluate the strategies these PD approaches target, codes for behavioral management strategies were developed based on the common practice elements established in McLeod et al. (2017), see Table 1 for definitions.

Table 1 Definitions of classroom management and positive behavioral support strategies
Full size table

Context and Characteristics of the PD Approaches

The context and characteristics of the PD approaches were evaluated to determine what types of settings, educators, instructional practices, and formats have been included in studies to date. Codes were developed for the early care and education setting of educators participating in the PD (Head Start, Public Preschool, Private Preschool, Early Education Center, Daycare Center); if these centers included children with disabilities (no child identified as having a disability, mainstream classroom, inclusive setting, self-contained special education classroom), format of training administration (online, video, individual coaching, live practice, group instruction, role plays, interactive case studies, peer-coaching; see Table 2 for definitions); training administrator (research staff, school administrator, peer educator, behavioral health professional, accredited trainer from PD approach, self-guided); timing of training (during school day, inservice days [during school year], inservice days [before the start of school year], summer break, outside typical workday [evenings or weekends]); and length of training (measured in hours or days). No codes were mutually exclusive, so a single study could be coded for multiple categories within each code.

Table 2 Definitions of format of training administration
Full size table

Study Design and Reported Outcomes

Codes were also developed for important dimensions of study design and outcomes in order to provide descriptive information relevant to the strength of the empirical evidence related to PD effectiveness. These codes included educator sample size, child sample size, components beyond educator PD included in the intervention (parent training, administrative training, training in academic curriculums, targeted services for behavioral intervention); study design (randomized controlled trial/cluster-randomized trial, open trial/pre-post design, quasi-experimental, multiple baseline); reported measures of educator universal behavioral management, reported child outcomes, if fidelity to the training approach was reported, and if attrition to the training approach was reported. Codes for additional components of training, measure of classroom management and behavioral supports, and child outcomes were not mutually exclusive, so a single study could be coded for multiple categories within each code.

Inter-rater Reliability

All included articles sharing a sample were evaluated when coding for each unique study. The first author conducted the initial coding of all included studies. Ten (23.8%) of the included studies were randomly selected to be coded by the second and third authors to evaluate inter-rater reliability. Initial agreement across categories was 94.3%. Agreement was 100% for early educator sample size, 100% for child sample size, 80% for classroom management and behavioral support strategies, 100% for center type, 100% for format of administration, 100% for training administrators, 80% for training timing, 100% for training length, 100% for study methodology, 80% for additional intervention component, 90% for measure of classroom management and behavioral support strategy reported, 100% for child outcome reported, 100% for fidelity reported, and 100% for attrition reported.

Data Analysis

Codes and definitions were entered into a Qualtrics web-based survey (Qualtrics, 2022) that all coders completed while reviewing included studies. Responses were then downloaded as a spreadsheet and compared across coders. Discrepancies were checked, resolved through discussion, and updated on Qualtrics. Data were imported into SPSS (IBM Corp, 2020) for subsequent analyses. Descriptive statistics were run for each of the coding categories to assess what classroom management and behavioral supports have been targeted in the PD literature, the characteristics and context of these PD approaches, and the study designs and outcomes that have been included in previous studies.

Results

Forty-two unique studies evaluating professional development approaches targeting classroom management and behavioral support skills for early childhood educators were identified utilizing search procedures. Only four of the codes included in this study consisted of mutually exclusive categories: utilized methodology, fidelity reported, attrition reported, and child sample included. For the remaining codes, it was possible for one study to include multiple components of a single code, so the total sum of reported percentages may be greater than 100%. See Table 3 for all included studies.

Table 3 Studies evaluating professional development for enhancing classroom management and behavioral support skills in early childhood educators
Full size table

Overall, there was a great deal of variability across the literature, in terms of strategies targeted, approach components, context of training, study design, and reporting practices. In general, studies were more likely to include preventative and positive reinforcement strategies, as compared to approaches for teaching expected behaviors and skills and responding to challenging behaviors. Most studies were conducted in public preschool or Head Start settings, and trainings were typically delivered by research staff. Trainings primarily utilized group instruction and coaching formats during trainings, while few included online components. Most studies utilized observations to measure educators’ use of classroom management and behavioral support strategies, and studies that included a child sample also predominantly relied on observations. Reporting on training dose, timing of training, fidelity, and attrition were variable. See the following sections for further details.

PD Approaches and Included Classroom Management and Behavioral Support Strategies

In terms of classroom management and behavioral support strategies, PD approaches included preventative and positive reinforcement strategies more frequently, as compared to approaches for teaching expected behaviors and skills and responding to challenging behaviors. The most common classroom management and behavioral support strategy to be included in the professional development approach was positive reinforcement strategies, with 33 (78.6%) of the 42 included studies describing one of these strategies. This was followed by antecedent control (73.8%), techniques for teaching expected behaviors (57.1%), promoting positive relationships (57.1%), instruction on child skills (52.4%), and responding to challenging or problem behaviors (42.9%). There was a range of how many strategies a single PD approach targeted, from a single strategy to strategies from all six coding categories. See Table 4 for a summary of strategies included in the PD approach.

Table 4 Number and percentage of studies reporting classroom management and behavioral support strategies targeted in training
Full size table

Context and Characteristics of the PD Approaches

The included studies were assessed for characteristics describing the populations who have been included in evaluating the PD approach and how these approaches have been delivered, as shown in Table 5. Educators who participated in the reviewed studies predominantly worked in publicly funded programs. Over half of the reviewed studies included early educators from Head Start classrooms (52.4%), followed by studies including educators from public preK classrooms (42.9%). Fewer studies included educators from childcare centers (16.7%), early childhood education centers (11.9%), and private preK classrooms (7.1%). Only one study (2.4%) did not report the type of setting in which early childhood educators worked. Studies included educators from one to four different center types.

Table 5 Number and percentage of studies reporting aspects of the context and characteristics of PD approach
Full size table

In terms of the format of PD components, there was a great deal of variability. Reviewed studies ranged in the number of formats utilized in the PD approach, from one to five formats utilized. The majority included a group instruction component (73.8%) and individual coaching (66.7%). Role-playing with the educators was utilized in 35.7% of reviewed PD approaches. In-person coaching, where the educator received live feedback either during or immediately after implementing targeted strategies in a classroom, was utilized in 47.6% of studies. Peer-educators were utilized to implement that coaching in 7.1% of studies included. A video component, whether that be a demonstration of targeted strategies or video review of the educators’ instruction, was utilized in 47.6% of studies included. Two studies (4.8%) featured interactive case studies, where educators received feedback on strategies used in a set case study, and only one study (2.4%) included an online component.

Studies were also coded for information regarding the delivery of the PD approach. The majority of the studies (64.3%) included at least one component of training that occurred during the school day. Seven studies (16.7%) included components that were delivered outside of the typical workday (evenings or weekends). Several studies included components that occurred during set training days, either inservice days throughout the school year (9.5%) or preservice days held immediately before the start of the school year (4.8%). Two studies (4.8%) included components that were delivered during summer break. The timing was not reported for at least one component of training in 33.3% of included studies. The length of these components was highly variable, from 0.5 to 48 h, with 11 (26.2%) studies not reporting the length of at least one of the included components.

Most studies (71.4%) included research staff as training administrators. Accredited trainers from the PD approach were utilized to deliver training in 28.6% of studies. Several studies utilized staff employed by early childhood programs or the systems they are part of to deliver at least one component of the PD approach, including licensed professionals (e.g., school psychologist) employed by the school system (23.8%), school administrators (e.g., assistant principal; 9.5%), and peer-educators (7.1%). A self-guided component of training was included in 4.8% of reviewed PD approaches. Finally, 4.8% of studies did not report who administered the PD approach.

Study Design and Reported Outcomes

Reviewed studies were also coded for characteristics of their study design and included assessments of educators’ use of classroom management and behavioral support strategies and children’s outcomes, as shown in Table 6. The number of educators included in a study was highly variable, ranging from one to 444. The majority of studies also included a child sample (61.9%), ranging from two to 5074 children participants reported. A randomized control or cluster-randomized design was utilized in 42.9% of studies, an open trial or pre-/post- design in 28.6%, and a multiple baseline design in 28.6%. Considering only the PD approaches, not implementation of strategies throughout the study, half of the reviewed studies reported fidelity and 76.2% reported attrition. An additional component of the intervention beyond PD training in classroom management and behavioral supports for early educators was included in about a third of studies. These components included parent training (11.9%), training on academics (11.9%), targeted intervention for children needing more intensive support (9.5%), and administrative training (7.1%).

Table 6 Number and percent of studies reporting aspects of study design and outcomes
Full size table

There was also a great deal of variability in measured outcomes of the PD approaches. The majority of studies measured educators’ use of classroom management and behavioral support strategies using live observations (81.0%), with an additional 7.1% of studies utilizing coding from a recording of the educators’ instruction. A self-report measure of strategies was included in 23.8% of reviewed studies, and one study (2.4%) included a multiple-choice knowledge test. Observations were also the most common measure of child outcomes (40.5%), including observations of target children (21.4%), classroom-level observations (14.3%), and observations of a subset of children in the classroom not selected for their behavioral challenges (7.1%). An additional study (2.4%) coded classroom-level outcomes from a video recording. At least one educator-rating was included as a child outcome in 28.6% of studies, with 16.7% rating all children in their classroom, 9.5% rating target children, and 2.4% rating a subset of children not selected for behavioral challenges. A small portion of studies reported parent-ratings of a subset of children in a classroom, not selected for their behavioral challenges (4.8%), and direct assessments of all children in a classroom (4.8%). A child outcome was not reported in 42.9% of studies.

Discussion

The purpose of this review was to conduct a systematic descriptive review of available studies of PD approaches targeting classroom management and behavioral support strategies for educators working in early childhood settings. In particular, we were interested in what classroom management and behavioral support strategies are being implemented, how PD approaches are being delivered to which educators, and the research designs and outcomes that have been utilized to evaluate these approaches. We used systematic search and coding procedures to generate descriptive data on the current state of this literature, which will provide information useful to researchers doing work in this area and early childhood administrators interested in PD approaches that target classroom management and behavioral support strategies. In the subsequent discussion, we group the key findings of the review and provide contextual information to inform future research in advancing empirical support for PD approaches in ecologically valid settings and reporting practices.

PD Approaches and Included Classroom Management & Behavioral Support Strategies

Across studies, there was a wide range of which and how many classroom management and behavioral support strategies were targeted. The most commonly targeted strategies included positive reinforcement (i.e., strategies such as praise and tangible reinforcement, which are used in response to positive behaviors; 78.6%) and antecedent control (i.e., strategies such as rules or monitoring, which shape the environment before a behavior occurs; 73.8%). Given the preventative nature of PBS (Hemmeter et al., 2007; Sugai & Horner, 2006), it is fitting that strategies implemented before child behaviors or to elicit more positive behaviors are among the most commonly included in the reviewed studies. Conversely, strategies for responding to negative behaviors (i.e., ignoring or time-out) were the least targeted approach (42.9%), included in fewer than half of the reviewed studies. It is surprising that more PD approaches do not include strategies for responding to negative behaviors when they occur for several reasons. First, challenging behaviors are highly prevalent in the early childhood setting, given children’s developmental level at that age (Barbarin, 2007; Yoder & Williford, 2019). Second, these behaviors have been shown to be associated with early childhood educators’ stress levels and poor wellbeing (Clayback & Williford, 2021; Friedman-Krauss et al., 2014a, 2014b). Finally, if teachers are not equipped with effective behavioral support strategies for managing the most challenging behaviors, they may be more likely to rely on harsh punitive consequences, such as suspension or expulsion. Past work has shown that educators are more likely to utilize suspensions and expulsions when they view behaviors as more disruptive, are concerned that they will be held accountable for the harm these behaviors cause, and experience higher levels of stress due to these behaviors (Gilliam & Reyes, 2018). Indeed, the reliance on punitive or punishment approaches may lead to a coercive process (Patterson, 2016) wherein an educator may issue more commands, demands, and reprimands. This may result in the children exhibiting a greater frequency of more oppositional or aggressive behaviors, triggering additional teacher strategies that can be coercive. Therefore, one finding from this systematic review is that there appears to be a need for greater support of teachers’ management of difficult and challenging behaviors.

Targeted instruction on child skills (i.e., inclusion of a social–emotional learning curriculum) and techniques for teaching expected behaviors (i.e., instruction in classroom expectations, such as through modeling or providing feedback on behaviors) were each targeted by approximately half of the reviewed studies. This is potentially a missed opportunity as a meta-analysis of the effects of classroom management strategies studied in preschool to Grade 6 found that interventions targeting children’s social–emotional development resulted in the largest effect sizes, particularly on social–emotional outcomes (Korpershoek et al., 2016). Surprisingly, strategies that promote a supportive and nurturing child–educator relationship were also only included in about half of the reviewed studies. The Teaching Pyramid (Fox et al., 2003; Hemmeter et al., 2021), a framework for implementing PBS in preschool environments, conceptualizes nurturing and responsive caregiving relationships as the base of the pyramid, supporting the implementation of all other practices (Hemmeter et al., 2016). Further, the child–educator relationship has been shown to be an important factor in preschool children’s externalizing behavior problems and social and academic development (Pianta & Stuhlman, 2004), indicating that strategies targeting positive child–educator interactions and relationships are an important part of the larger classroom management and behavioral support toolkit.

Context and Characteristics of the PD Approaches

The vast majority of reviewed studies included early childhood educators working in publicly funded programs, either Head Start (52.4%) or public preschools (42.9%). In contrast, relatively few studies included educators working at childcare centers (16.7%) or private preschools (7.1%). However, over half of the national early childhood workforce is made up of childcare workers (McLean et al., 2021), and approximately 40% of three- and four-year-olds enrolled in preschool attend a private institution (U.S. Department of Commerce & Census Bureau, 2019). This indicates a large gap in the empirical literature of including child care and private early childhood professionals in studies of PD approaches targeting classroom management and behavioral support strategies.

There is a comparable gap in who is administering the PD approach. Nearly three-quarters of the reviewed studies included research staff as the training administrators, with only approximately a quarter of studies including each licensed professionals in the school system or accredited trainers from the PD approach. Very few studies included school administrators (9.5%), peer educators (7.1%), or self-guided approaches (4.8%). The majority of early childhood centers looking for PD approaches targeting classroom management and behavioral supports will not be able to find a university partner to conduct these trainings, and even if they do, it is unlikely that that partnership could be sustained for their long-term training needs. Research on intervention dissemination, implementation, and transportability has shown that characteristics of the intervention, practitioner, participant, and setting interact to impact a given intervention in practice (Schoenwald & Hoagwood, 2001). Applied to school settings, the implementation and sustained practice of evidenced-based interventions are affected by factors such as child characteristics, educator characteristics, attitudes and perceptions of the intervention, school characteristics, personnel expertise, classroom or school climate, administrative leadership, and policies and finances (Domitrovich et al., 2011; Forman, 2013; Forman et al., 2008). Therefore, it is notable that sustainable practices and the diverse characteristics of early childhood centers and workforces have not been included in much of the research on PD approaches targeting classroom management and providing positive behavioral supports to date.

In terms of how the PD approach was administered, there was a great deal of variety. The majority of reviewed studies utilized group instruction (73.8%) and individual coaching (66.7%), nearly half included in-person coaching and video components, and over a third included role-playing. In contrast, relatively few studies employed peer-coaching (7.1%), an online component (2.4%), or interactive case studies (4.8%), leaving opportunities for future work to explore different modalities of training, particularly those not requiring a live or synchronous training. As PD approaches aim to impact educators’ practice with children, there has been a focus on approaches that combine didactic workshops with practice elements such as role-playing, modeling, and coaching (Dunst, 2015; Zaslow et al., 2010); however, few studies on PD in early childhood education cite adult learning theories or their reasoning for how the PD approach was designed (Schachter, 2015).

The most common time for a component of training to take place was during the school day (64.3%). However, a third of the reviewed studies included at least one component where the timing was not reported. The length or dose of training was highly variable, ranging from half an hour to nearly 50 h, and also under-reported, with over a quarter of studies not reporting the dose of at least one component of training. Previous work has found that longer programs, with more opportunities to practice and gain proficiency in taught skills, are associated with more sustained changes in educator practices and effects on child outcomes (Brunsek et al., 2020; Dunst, 2015) These factors are critical information for early childhood centers in deciding what PD approaches will meet the needs of their centers, staff, children, and families, while also fitting within the constraints of their schedules and budgets.

Study Design and Reported Outcomes

When early childhood programs seek out professional development opportunities for their staff, they are typically hoping that these lead to improved outcomes for their program. Although some may be hoping to improve educator (proximal) outcomes, such as stress or retention, child (distal) outcomes, such as levels of challenging behaviors or school readiness, are important (Sheridan et al., 2009; Zaslow, 2009). In order to evaluate whether a PD approach is leading to these outcomes, high-quality empirical evidence and clear reporting practices are needed. In the current review, studies were variable in what information was reported in the publication. About 75% of studies reported attrition to the PD approach and only half reported the fidelity to the PD approach. Notably, we limited attrition to the PD approach, in order to evaluate if information on staffs’ completion of the training was available. We did not evaluate if the publication reported attrition to the whole study, including at follow-up. Providing data on attrition in the intervention is particularly important within this population, given the high turnover rates in the early care and education workforce (Bassok et al., 2021; Wells, 2015).

The current review revealed that there is a wide range in the quality of the empirical support for PD approaches targeting classroom management and behavioral support practices. Nearly half of studies utilized randomized controlled or cluster randomized trials, with slightly more than a quarter each utilizing open trials and multiple baseline designs. The number of educator participants was highly variable, ranging from one to 444; however, two-thirds included fewer than 50 educators. Promisingly, over 80% of studies included at least one observational outcome of educator classroom management and behavioral support practices. In terms of distal outcomes, the number of child participants was similarly variable, ranging from two to 5074; however, over a third of the studies did not include a child sample. Observations were also the most common measure of child outcomes, although only approximately 40% of reviewed studies included them. Educator-ratings were also regularly utilized, being included in over a quarter of recorded studies. Conversely, parent-ratings and direct-assessments were rarely used. Assessment of children’s behaviors and skills should match the dimension being measured, and each assessment method has strengths and limitations (McKown, 2017; Whitcomb, 2012). While observations and educator-reports can be reliable measures of children’s behaviors, programs interested in how improved classroom management and use of behavioral supports impacts children’s academic outcomes may also want to consider direct-assessment outcomes. The fullest understanding of outcomes will be provided by studies integrating different forms of assessment looking at multiple aspects of children’s behavior, skills, and knowledge (Denham, 2006).

Limitations

There are several limitations to consider when reviewing the results of this study. Although we conducted an expansive review of peer-reviewed studies targeting classroom management and behavioral support strategies for educators working in early childhood settings, there is likely relevant work that was not included in this review. Similar to other reviews investigating PD in Early Childhood Education (Snyder et al., 2012), we elected to exclude studies that were not peer-reviewed, such as dissertations. Further, there may be relevant studies that were never published and PD approaches targeting these skills that have not been formally studied, which could not be included in the review.

Additionally, we elected to only include studies with participants who were working with children aged three to five. We selected this inclusion requirement for several reasons. First, social–emotional competence and learning-related skills developed in this age range contribute to children’s school readiness when entering Kindergarten, an important predictor of later school success (Denham, 2006; Duncan et al., 2007; McClelland et al., 2006). Second, due to children’s rapid development from age zero to five, the developmental, educational, and supervision needs of children below the age of three are quite different from those aged three to five, which is reflected in the staffing recommendations for centers serving these different age groups (Administration for Children & Families, 2022). Accordingly, there are different professional development needs for staff working with these age groups, as evidenced by previous work showing that educators serving infants and toddlers have differing experiences with PD compared to educators working with preschoolers (Gardner-Neblett et al., 2021). Finally, there has been increasing recent momentum behind providing universal PreK to children aged three and four in the United States (Fact Sheet: The American Families Plan, 2021), which would require a massive expansion of the early childhood workforce and their training needs to provide high-quality programming (Barnett, 2021), a daunting task given the current unequal distribution of high-quality early childhood options across the country (Gomez et al., 2015). However, this does not diminish the professional development needs of early childhood educators working with children aged zero to three, and future work should consider the options targeting this population.

This study was also limited to reporting the information the authors of the reviewed studies included in their publications. As is evident from the review, several factors key to understanding the implementation of a PD approach, such as training length, when the training was conducted, attrition, and fidelity, were under-reported. As a result, our study was limited in that we were not able to evaluate the quality or rigor of the research designs or methodology beyond the study characteristics reported in Table 3. Similarly, the fullness of the description of the PD approach varied across reports. For example, some studies implementing the same PD approach described different strategies included (i.e., Baker-Henningham et al., 2009; Hutchings et al., 2013; Shernoff & Kratochwill, 2007; Webster-Stratton et al., 2008). It is unclear if this is due to differences in reporting or adaptations of the training. In their review of all early childhood professional development studies, Snyder et al. (2012) found that approximately a third of studies did not provide sufficient information on the adult learning strategies used to convey content. While all of the reviewed studies provided some information on content delivery (e.g., group instruction), many did not go into further detail about the adult learning strategies used during this instruction (e.g., role-playing, video examples). Finally, there were likely different degrees of overlap and intensity of implementation of the varied components of PD approaches. Based on the information reported within the primary papers, we were not able to explore whether these variables impacted early educator training or implementation of the strategies. Accordingly, we strongly encourage researchers conducting an empirical study on early childhood PD approaches to follow the recommendations for reporting practices outlined by Snyder et al. (2012).

Implications and Future Directions

The current study has several implications for researchers, school administrators, and mental health professionals. Beyond the reporting practices outlined in the previous section, future PD researchers should consider the needs of early childhood settings and workforce. Despite making up the majority of the early childhood workforce, few studies have included educators from child care or private preschool settings. Studies have also relied on research staff to administer PD approaches, rather than school administrators, mental health professionals, and accredited trainers, who are typically providing these trainings in everyday early childhood settings. Given the importance of these factors to the uptake and sustainability of evidenced-based strategies in regular practice (Domitrovich et al., 2011; Forman, 2013; Forman et al., 2008; Schoenwald & Hoagwood, 2001), future research should aim to diversify included settings and trainers. More work is also needed in understanding different formats of training and tailoring them to meet the needs of early care and education settings. The early childhood workforce is incredibly diverse with differing backgrounds, knowledge, and skills that they bring into a PD training (Fuligni et al., 2009; Phillips et al., 2016; Schachter, 2015). They are also working in a range of centers, with varying levels of resources, and unique school climates, all of which impacts how PD training is received (Hooper et al., 2022). Further, different center types have varying levels of constraints that require adaptive PD options which fit their needs (Koh & Neuman, 2009). Providing early care and education settings with a diversity of options for strategies targeted, format of administration, when the training can occur, and length of training will allow administrators and staff to select approaches that meet the needs of their educators, centers, children, and families (Schachter et al., 2019).

Future PD approaches may also address the diversity of early childhood care and education centers by incorporating discussions on implicit bias and cultural responsivity and humility in trainings on the use of positive behavioral supports and classroom management skills. Several calls have been made to incorporate trainings on these topics to address the disproportionate use of harsh and exclusionary disciplinary practices with Black children, particularly boys (Allen & Steed, 2016; Neitzel, 2018; Westerberg, 2016). Addressing implicit bias and cultural responsivity is a complicated task that calls for intentional and ongoing supports for early childhood educators. Research into how PD models can reduce implicit bias and associate harsh disciplinary practices in early childhood care and education settings is starting to become available (Davis et al., 2020; Fox et al., 2021). Accordingly, future approaches should consider how to intentionally incorporate topics of implicit bias and cultural responsivity and humility into PD targeting educators’ use of positive behavioral supports, as well as systematically evaluate the effectiveness of these trainings on educators’ use of positive and harsh disciplinary practices across demographic groups.

The current study also offers other insight to future trajectories of PD approaches for early childhood educators. First, the COVID-19 pandemic has greatly advanced the use of online and digital tools in education settings, including approaches to PD (Crawford et al., 2021). Previous work has shown that these digital tools can be used to effectively target preschool child–educator interactions (Pianta et al., 2014), but that other factors, such as use of a follow-up consultant, impact teachers’ engagement with training (Downer et al., 2009). More work is needed to fully understand how to utilize these tools and what other supports are needed for these PD approaches to impact educators’ practices and child outcomes. Future work should also consider and target other outcomes of PD for early educators, such as improving their wellbeing. Past reviews have called for research to investigate whether targeting educators’ well-being in PD also impacts children’s outcomes (Zaslow et al., 2010); however, educators’ well-being has been shown to be important to key areas of interest, such as classroom climate, levels of expulsion, and retention (Bassok et al., 2021; Friedman-Krauss et al., 2014a, 2014b; Gilliam & Reyes, 2018). Work with primary school teachers has shown that one PD approach targeting classroom management strategies decreased teachers’ work-related stress and increased wellbeing (Hayes et al., 2020), and coaches working with early childhood educators since the start of the COVID-19 pandemic have reported that providing encouragement and support to help early educators manage stress has been an important aspect of their role (Crawford et al., 2021). Finally, this review focused on preventative or universal practices; however, many children displaying challenging behaviors may also have developmental delays, learning disorders, or a disruptive behavior disorder (Barbarin, 2007; Yoder & Williford, 2019). Many of these children would benefit from more intensive forms of support, typically referred to as Tier-2 and -3 supports (Bayat et al., 2010; Greenwood et al., 2011). While specialized educators or behavior consultants may be necessary for implementing some of these supports, many classroom educators are involved with implementing tier-2 interventions, such as BEST in CLASS (Conroy & Sutherland, 2008) or a Daily Report Card (Daniel et al., 2021; Fabiano et al., 2010). Future research and reviews should also consider what PD and implementation support options are available for early childhood educators addressing challenging behaviors with more targeted supports.

With so many different options available for PD, it can be daunting for early childhood administrators and mental health professionals to select an approach that will accomplish their training goals for their center. Schachter et al. (2019) proposed guidelines for selecting PD content, including selecting formats that provide opportunities for knowledge gains and skill practice, are targeted on a specific content area for an extended time, and incorporate chances for educators to self-reflect and receive feedback. The current review is intended to be helpful to school administrators and mental health professionals in evaluating the extent to which PD approaches targeting evidence-based strategies for improving children’s behavior and social–emotional competence are in line with these guidelines, while also fitting the unique needs of their setting.