Consistent with this definition, research has documented short- and longer-term outcomes associated with self-regulation as a resilience factor in early childhood. For example, Hardaway et al. (2012) found that family chaos was negatively associated with children’s self-regulation which in turn influenced externalizing behaviors. Their findings indicated that children’s self-regulation played a full mediating role in the relationship between family chaos and children’s externalizing behavior. This suggests that children’s self-regulation skills could serve as a crucial factor in mitigating externalizing behaviors within chaotic family environments. Sciaraffa et al. (2018) conceptualized that children’s self-regulation served as a protective factor promoting resiliency in learning, behaviors, and health associated with toxic stress caused by adverse childhood experiences. In fact, MacPhee et al. (2015) have reviewed multiple studies showing self-regulation acts as a resilience factor impacting children’s developmental outcomes. They reviewed multiple intervention programs designed to promote emotional regulation which was related to children’s behavior problems, adolescents’ substance use, and parents’ depression and anxiety. These studies highlight that self-regulation skills can function as a resilience factor under conditions of family adversity including families’ economic hardship.

Prior research has also documented the importance of self-regulation in the school context. For example, McClelland et al. (2006) found that children who enter kindergarten with high self-regulation skills also show higher achievement in math and literacy through sixth grade, with the academic achievement gaps increasing through second grade between children with higher and lower levels of kindergarten self-regulation. Using a large, nationally representative Australian sample, Howard and Williams (2018) found that self-regulatory skills demonstrated in preschool and primary grades were significantly associated with a range of adolescents’ outcomes in academic achievement (reading, numeracy), school truancy, health (weight, smoking, alcohol use), and mental well-being a decade later. This study also documented the potential for changes in self-regulation relatively early in life, between ages 4 and 7, to support positive developmental outcomes by reducing the risk of poor academic achievement and demonstration of risk behaviors that begin in adolescence (e.g., suicidal thoughts, smoking, property crimes, and school truancy) and have significant consequences for lifelong health and well-being.

As illustrated by these examples, a body of research has established that early self-regulation predicts important short- and long-term outcomes across a wide range of areas including school readiness, academic achievement, feelings of higher self-worth, a better ability to cope with stress, better mental health, and more positive adolescent and adult outcomes (Moffitt et al., 2011; McClelland et al., 2015; Montroy et al., 2016; Howard and Williams, 2018). This body of research has prompted some to conclude that the development of self-regulation in young children is an early life marker for later life success (Montroy et al., 2016). Importantly, this body of research also documents that self-regulation is malleable, sensitive to developmental processes to enhance resiliency. Several theories have postulated about the underlying developmental processes.

Developmental Systems Theory (DST) proposes that individual development emerges from myriad interactions across system levels, all shaped by the interaction of processes within and between individuals and their contexts, and at multiple levels of systems including culture, society, and ecology (Masten and Barnes, 2018). Through this dynamic process, resilience and its development then become not only a feature of the individual child, but rather the complex experiences the child has through relationships and interactions with other people in their system. Accordingly, self-regulation—and resilience, more broadly—within a child shapes the environment and interactions they experience, which in turn, shapes the ongoing development of self-regulation and overall resilience. For example, Blair (2009) emphasizes that age-related changes in self-regulation and individual differences in self-regulation at a given age or stage fundamentally shape children’s experiences and the responses that children evoke from caregivers and others, which further shape self-regulation.

To date, most research on the development of self-regulation has been conducted with preschool age children versus younger children. Previous research suggests a qualitative shift at age 3 with rapid, nonlinear development due to the integration of individually developing components of self-regulation (Diamond, 2002; Montroy et al., 2016). This observation and other research findings document that self-regulation is developing during the infant-toddler period and is impacted by the child’s context.

The existing literature focused on infants and toddlers has typically examined the home/family context and its associations with the development of early self-regulation. For example, Morawska et al. (2019) note that young infants have very limited capacity to self-regulate but rapidly develop it in response to parental socialization. Between 12 and 18 months of age children develop self-regulation skills (Kochanska et al., 2000) with evidence that specific parent–child interactions foster its development (Valcan et al., 2017). Specifically, meta-analyses document that parental use of positive strategies are associated with better child self-regulation with use of negative parenting strategies (e.g., coercive behaviors) being associated with weaker child self-regulation (Valcan et al., 2017). Even those sources highlighting parental impacts acknowledge the role of more distal contexts including features of the home, non-parental care settings, and peer groups (McCabe et al., 2004).

Despite this acknowledgement, few studies have examined the role of infant-toddler group care in the development of self-regulation. One of the few exceptions (Salminen et al., 2021) examined associations between teacher–child interaction quality and children’s self-regulation in Finnish and Portuguese toddler classrooms and found that specific features of teachers’ classroom interactions can promote the development of children’s self-regulation skills in these two sociocultural contexts. Given the increasing enrollment of infants and toddlers in out-of-home center-based early care and education (ECE) programs in the US (Horm et al., 2016), with infants spending significant time in these settings (Forry et al., 2018), a gap exists in our understanding of the development of self-regulation in this important infant-toddler context.

The data used in this study were drawn from a larger evaluation of a statewide initiative supporting early childhood programs serving children birth to age 3 growing up in low-income contexts in the state of Oklahoma (OECP, 2021–2022). The participating programs met the Early Head Start Performance Standards, operated at least 8 hours per day for at least 44 weeks of the year, and staffed each classroom with two teachers with at least one of these teachers holding a minimum of a CDA credential. A total of 253 classrooms across 9 agencies participated in this statewide initiative during 2021–22. From these participating classrooms, the program evaluator selected a total of 52 classrooms to be part of their 2021–22 evaluation study including infant, toddler, and preschool classrooms. Because the research questions posed in the current study focus on the potential associations of infant-toddler classroom quality and the development of self-regulation as a component of resilience in young children from low-income families, we selected only the 37 infant and toddler classrooms that were part of the larger evaluation study. The 37 infant and toddler classrooms represented three large Early Head Start/Head Start agencies in two large metropolitan cities in Oklahoma. Children who were enrolled in infant-toddler classrooms over a year were included in the final analyses (n = 269).

As noted above, the 269 children included in this study were enrolled in infant or toddler classrooms participating in a statewide initiative (OECP, 2021–2022) to expand and enhance ECE programming for children ages birth to 3 growing up in low-income contexts. As shown on Table 1, the participants included boys (55%) and girls (45%) between the ages of 3 to 37 months. Our sample included Spanish-speaking children (33%) and was racially and ethnically diverse.

Due to the fact that this study was a secondary analysis of program evaluation data, additional demographic information on the specific participants was not available. However, the initiative required that participating families met the eligibility criterion of having incomes at or below 185% of the federal poverty level. The statewide initiative reported that for the 2021–2022 school year, 3,011 children were enrolled in rural and urban settings across the state and that a total of 12,023 referrals for basic needs, emergency interventions, and other community service assistance were made (OECP, 2021–2022). This state program was modeled on Early Head Start in terms of eligibility and services provided and thus the families and children enrolled have the demographic characteristics and needs of a population that is marginalized and low resource.

The design of this study involved accessing existing data on children’s social-emotional development over a program year and information about the quality of the infant-toddler classrooms participating in the state initiative. Standardized measures were used to collect these existing data. Teachers rated children’s social-emotional development, including self-regulation, at the beginning and end of the 2021–22 year. Classrooms were observed in the winter to document quality.

Children’s self-regulation was rated twice, fall and spring, by their infant-toddler classroom teacher. These ratings were completed electronically by each classroom’s lead teacher. Teachers experienced a brief training on how to use the rating scale prior to completing their ratings. Teachers were given 1 month to complete the ratings for the children in their classrooms. Each teacher rated approximately 3 to 17 infants and toddlers, depending if the child’s age conformed to the age range measured by the social-emotional rating scale used in the larger evaluation study. The average number of children in the classrooms was 8 (S.D. = 2.4) and teachers rated all children’s social-emotional skills in fall and spring. Across the classrooms used in this study, enrollments varied based on agency’s grouping patterns and COVID-related low enrollments, but self-regulation scores were available for most children with ratings missing for only 1 or 2 children per classroom. The fall ratings were completed between October 2021 and November 2021, allowing teachers time to experience and know the children’s typical behaviors to produce valid ratings; the spring ratings were collected between April 2022 and May 2022.

Classroom observations were conducted between late November 2021 and March 2022; and did not overlap with the teachers’ ratings of children’s self-regulation. Data collectors attended a 4-day training in the fall of 2021 and completed a video certification to document reliability prior to data collection. To be certified as reliable, data collectors needed to agree on all items of the measure within 1 point on the 7-point scale (80%) and obtain at least 75% agreement in each of the three domains of the measure. Additionally, trained and reliable data collectors attended an in-person refresher training to review the observation guide prior to starting the classroom observations. Each classroom was observed one morning, with these observations lasting approximately 2–2.5 h. Following the procedure outlined by the developers of the classroom observation measure, observations consisted of six 10-min cycles with 5 to 10 min allocated between cycles for scoring and notetaking. Observers rotated observing and scoring each teacher present in the classroom and scored each cycle based on the observed teacher’s interactions with the enrolled infants and toddlers. After scoring the last cycle, observers rated “across the visit” items using evidence from the entire observation period.

For this study, we used both classroom-level average self-regulation scores and individual-level self-regulation scores in the analytical models. We aggregated self-regulation skills of children who were enrolled in the same classroom in fall to examine how the classroom-level average of children’s self-regulation skills at the beginning of the year were associated with classroom quality. As noted previously, the number of children in a classroom ranged from 3 to 17 with an average of 8 children per classroom. The classroom-level average self-regulation scores in fall were used as a classroom-level predictor (i.e., level-2) for classroom quality. This study created a factor representing classroom quality using the three domains (Support for Social-Emotional, Cognitive, and Language Development) obtained from QCIT. We conducted a confirmatory factor analysis (CFA) to confirm that the classroom quality latent variables demonstrated acceptable loadings on the three domains.

To address the two research questions, we conducted multi-level models employing Mplus (Muthén and Muthén, 1998–2017). For RQ1, we conducted a multi-level direct model by using the average of classroom-level self-regulation (level 2) as a predictor and gains in individual children’s self-regulation scores between fall and spring as an outcome. We included classroom ID in fall 2021 as a cluster ID and ran a two-level model with BETWEEN and WITHIN model commands. Individual children’s self-regulation skills in spring were regressed on the classroom-level average children’s self-regulation and children’s individual self-regulation skills in fall. We also controlled for children’s gender and home language in the direct model (see Figure 1).

For RQ2, we examined how classroom quality (level-2) mediated between the initial classroom-level average children’s self-regulation (level-2) and individual children’s gain in self-regulation skills (level-1) across a year after controlling for children’s gender and home language. We used a multi-level mediation model to address this research question (see Figure 2). In the Mplus mediation model, level-2 variables were included in BETWEEN to examine a level-2 mediational path with an exogeneous variable (i.e., classroom-level self-regulation) and an endogenous variable (i.e., classroom quality) and the intercepts of level-1 outcome (i.e., self-regulation in spring). In WITHIN command, individual children’s self-regulation in fall, and gender and home languages were included as exogenous variables (i.e., predictors) and self-regulation in spring as an endogenous variable. We used the model constraint command to test the significance of the indirect paths from classroom-level children’s self-regulation skills to individual children’s gains in self-regulation skills through classroom quality (RQ2: 2–2-1 multilevel mediational path).

Several limitations of the current study should be noted. First, given the use of a secondary dataset that was created for program evaluation purposes and not purposely designed to investigate the development of self-regulation in infant-toddler classrooms contexts, our study was limited to the available data. As noted previously, we had limited demographic information on the sample. Future research designed specifically to investigate the development of self-regulation in infant-toddler classroom contexts with diverse samples is needed to deepen our understanding.

Second, and relatedly, although measures of self-regulation are limited, especially those designed for use with infants and toddlers, our study would have been strengthened with additional measures of self-regulation. Although the DECA adopts a strengths-based approach consistent with our view of self-regulation as an important component of resilience and demonstrated strong internal validity in our study, it is based on teachers’ ratings of children’s behavior and thus shares the limitation of potential bias common to all behavior ratings. Although teacher ratings are often used to capture children’s social-emotional development, including self-regulation, this approach has known limitations. The teacher ratings used in this study could potentially be influenced by teachers’ perceptions, dispositions, or biases and thus may not be a pure measure of children’s self-regulation skills. Given that young children’s self-regulation skills can be observed and assessed, alternative methods such as direct observation could serve as effective measures for evaluating the self-regulation abilities of infants and toddlers in group care settings. Furthermore, incorporating parental reports of their children’s self-regulation skills could offer a comprehensive multi-informant assessment of self-regulation abilities in infants and toddlers. Future research incorporating these and other approaches to the measurement of self-regulation in infants and toddlers will likely broaden our understanding.

Third, our study could not isolate the influence of teachers versus peers on the development of self-regulation in the context of the studied classrooms. It is acknowledged that both teacher-child and child–child interactions contribute to classroom quality, the experiences of individual children in the classroom context, and the development of self-regulation in older preschoolers (Choi et al., 2018). Methods to disentangle these components of classroom experience is an area ripe for future research and development.

Fourth, the data used in this study were obtained from centers participating in a statewide initiative to enhance infant-toddler programming. Thus, it is possible that they may have offered different, potentially better, services than infant-toddler programs not participating in such an initiative. The average QCIT scores reported on Table 1 for the classrooms used in this study are similar to, but slightly higher than those reported by the QCIT’s developers (Atkins-Burnett et al., 2015). But, we need to acknowledge that the classrooms included in this study may offer programming that is not typical for infants and toddlers in the US, especially those enrolling children from low-resource contexts (Ruzek et al., 2014). Thus, generalizability of our findings could be limited and considered program specific because of the professional development and support the classrooms in our study were receiving.

Last, given this was an observational study that investigated associations, causality should not be implied. More research is needed to build the evidence base and deepen our understanding of the development of self-regulation in infants and toddlers, especially in varied contexts including classrooms. As an example, a bidirectional research design, encompassing both classroom quality and children’s self-regulation in both fall and spring, can offer a more nuanced understanding of the causal relationships between classroom quality and the development of children’s self-regulation.

Despite these limitations, our study had notable strengths that position it to make contributions to the existing body of literature. First, our study investigated the development of self-regulation, a skill that has received heightened attention as a key mechanism predicting a variety of important outcomes (McClelland et al., 2015), in an understudied age group, infants and toddlers. Beyond the focus on the development of an important skill in a unique age group, our study considered an understudied context, the early childhood classroom. Even with the body of the more prevalent preschool studies, it has been observed that the focus has generally not be placed on the ECE classroom context. In fact, Montroy et al. (2016) have called for more investigation of early school contexts to better understand the development of self-regulation. Our study focused on the context of infant-toddler center-based classrooms, using a relatively new tool that provided nuanced information about the nature and quality of the interactions and opportunities the children experienced. As discussed above, infant-toddler classrooms are an important context given the growth in infant-toddler group care and the amount of time increasing numbers of children spend in these settings (Forry et al., 2018).

Our results suggest that teachers adjusted their levels of support for social-emotional, cognitive, and language development based on the average level of self-regulation demonstrated by the infants and toddlers in their classrooms in the fall. Our inference is that the teachers whose classrooms scored high on the QCIT measure of quality were skilled in developmentally appropriate practices (NAEYC, 2020) which guide teachers to assess and incorporate information about children’s developmental skill levels and unique characteristics into their planning and implementation of classroom experiences and learning opportunities. Indeed, the lead teachers participating in this study were involved in a statewide initiative that provided training and other supports to improve and enhance infant-toddler classrooms. However, this is not typically the case. Past research has documented that teacher-child interactions and instructional support are typically lower quality in infant-toddler compared to preschool classrooms (La Paro et al., 2014; King et al., 2016). Infant-toddler teachers typically have lower levels of formal education than preschool, PreK, and K-12 teachers (Madill et al., 2016) with the infant-toddler workforce having some college but no degree (36%), or a high school diploma (28%) with fewer having an AA (17%) or bachelor’s degree or higher (19%) (NSECE, 2013). Additionally, fewer professional development (PD) opportunities are available that are tailored to the infant-toddler age group than those available for teachers who work with older children and research suggests the infant-toddler PD that is available is often delivered in formats that have questionable effectiveness and cover content focused on very basic skills (Madill et al., 2016). Pre-service preparation in institutions of higher education also has limitations with few infant-toddler focused courses, practicum or field experiences, or degree programs available (Horm et al., 2013; Chu, 2016). Thus, the findings of this study suggest what is possible with the supports of a state initiative offering ongoing and appropriate PD with content tailored to the infant-toddler age group receiving group care and education in center-based classrooms. Unfortunately, as documented above, this is not what is commonly available to infant-toddler teachers. Fewer PD opportunities are available for infant-toddler teachers overall and what is available is often offered as stand-along sessions without ongoing support (Madill et al., 2016).

A common observation is that the infant-toddler workforce is under-valued, under-compensated, under-professionalized, and under-supported (NSECE, 2013; Whitebook et al., 2016; Kwon et al., 2019; Vallotton et al., 2019). This exists despite the growing recognition that birth to age 3, as a unique developmental period, serves as the foundation for later development (Horm et al., 2016; Chazan-Cohen et al., 2017). The unique developmental characteristics of infants and toddlers, combined with the malleable nature of their development, necessitates that the adults in their lives, including nonparental caregivers such as infant-toddler teachers, are knowledgeable about how best to support and facilitate individual development. The findings of this study highlight the importance of infant-toddler teachers understanding how to implement developmentally appropriate classroom practices and ensure high-quality experiences tailored to the rapidly changing developmental skills of infants and toddlers. The measure of classroom quality used in this study captures how teachers adapt to the wide range of skills exhibited in infant-toddler classrooms and adjust to meet the needs of the enrolled infants and toddlers. Relative to self-regulation, strategies that are appropriate for non-mobile or pre-verbal infants may not be helpful or appropriate for a mobile infant and those with more language skills. Our results suggest that teachers who were skilled in creating a developmentally appropriate classroom supported the development of self-regulation in infants and toddlers through offering a higher quality of classroom experience. The higher quality classroom setting was associated with the further development of self-regulation skills in the enrolled infants and toddlers. Given the current nature of the infant-toddler workforce, this level of teacher expertise is a lofty expectation in most settings. Others have sounded the call for reform, but efforts to professionalize and appropriately compensate the infant-toddler workforce will take time. In the meantime, our findings suggest the development of classroom activities, curriculum, and associated PD to enhance quality in infant-toddler classrooms and expand teachers’ knowledge of child development, including the development of self-regulation, is paramount. Morawska et al. (2019) published an overview of the role of parenting in the development of self-regulation in children that highlights the evidence base for parenting interventions designed to promote the development of self-regulation in children under age 8, including a focus on infants and toddlers. The ECE field would profit from a similar review. Although some resources are available, a more intentional focus on accessible and easily implemented resources is crucial given the likely role of high-quality classrooms in supporting self-regulation development in conjunction with the current status of the infant-toddler workforce. Once again, it is worth noting that the classrooms in this study were drawn from centers that served infants and toddlers growing up in low-income contexts. In an analysis focused on a sample of toddlers from ECLS-B, Ruzek et al. (2014) found that toddlers in nonparental care typically experienced medium quality care (61%) rather than high quality (26%) or low quality (13%) care. Additionally, they found that children from low-resource homes that were enrolled in out-of-home care were more likely than their non-low-income counterparts to be in lower quality care. While the quality of care is important for all children and families (Vandell et al., 2010), it is especially important for young children growing up in the contexts of poverty and associated developmental risks because research has shown these children gain the most from high-quality care (La Paro et al., 2014). Thus, resources that are easy to access and use are critically important given the more typical low resource programs many infants and toddlers growing up in low-income contexts experience.

The topic of self-regulation as a protective factor that promotes resilience as early as the infant and toddler years is important, including examination of its development within early care and education settings. Thus, the observation that little research has focused on the development of self-regulation in infant-toddler classrooms, to date, calls for more research. Additionally, the findings of the one existing published study we located (Salminen et al., 2021) that studied associations between teacher–child interaction quality, as measured by the CLASS-Toddler (La Paro et al., 2009), and several aspects of children’s self-regulation in Finnish and Portuguese toddler classrooms suggests a deeper and more nuanced examination is needed. Interestingly, the results found by Salminen et al. (2021) regarding which aspects of teacher behavior, engaged support for learning or emotional and behavioral support, associated with specific components of child self-regulation including attention and inhibitory control varied by country. While documenting that features of teachers’ classroom interactions can promote the development of children’s self-regulation skills overall, their findings indicate more research is needed in different sociocultural contexts and with measurement tools that offer more specificity in children’s developing skills and teachers’ classroom interactions.

Our study used a relatively new measure of infant-toddler classroom quality—the Quality of Care for Infants and Toddlers (QCIT; Atkins-Burnett et al., 2015). The availability of the QCIT, which was specifically developed to measure the quality of caregiver-child interactions for infants and toddlers across group settings, including center-based care and family child care homes, has implications for future research. The QCIT was developed based on an extensive review of the caregiver-child interaction literature and measures three important domains of caregiver support including social-emotional, cognitive, and language/literacy development, as well as areas of concern. The goal of the developers was to provide “early childhood professionals and researchers the means to obtain a better understanding of how caregivers and young children interact in child care settings and improve child care services in the future” (Atkins-Burnett et al., 2015, p. 1). As a tool developed specifically for infant-toddler ECE settings, it offers more nuanced information than other tools that were adapted for infant-toddler settings after being originally designed for classrooms serving older children. We advocate that the QCIT is a contemporary, comprehensive, and research-based tool that offers fresh insights on infant-toddler settings. For example, the existing data used in this study restricted our investigation to social-emotional aspects of development given the DECA was the only child outcome measure available to us. With its comprehensive focus including indicators of teachers’ support for cognitive and language development, and inclusion of areas of concerns across the observational visit, the QCIT offers possibilities for similar studies investigating other important child outcomes. Additionally, because the QCIT can be used in both center- and home-based programs, it expands research opportunities. Thus, the QCIT is positioned to contribute to and enrich the existing scant body of infant-toddler research examining impacts of early group ECE experiences, both center- and home-based.

Given the importance of self-regulation as a critical component of resilience, the field would profit from additional studies that deepen understanding of the mechanisms supporting its early development. We took advantage of an existing data set to explore associations of characteristics of enrolled infants and toddlers and features of their classroom context. Future research using different measures, designs, analytical strategies, and diverse samples and contexts is needed to further explain young children’s development of self-regulation. For example, different measures of infant-toddler self-regulation and classroom quality could be used. Longitudinal designs following the effects of various classroom characteristics on young children’s self-regulation development over time are needed. Analyses that test moderation as well as mediation and include diverse samples and classroom contexts are also needed. Relative to classroom context, it is important to investigate different grouping strategies such as mixed-age grouping that is often used in infant-toddler settings. Additional studies could investigate individual children’s experiences with teachers and control for parenting practices to further our understanding. These few examples illustrate the many open questions in the current literature.