Early Socio-Emotional Competence and Peer Interaction in Toddler Classrooms: A Multi-Rater Study

Abstract

Socio-emotional competence in early childhood predicts later academic and social outcomes. We assessed 287 toddlers (M_age=28 months, SD=4.5) across five child-care centres in urban Ontario using teacher ratings, observational coding, and parent-report questionnaires of emotion regulation, prosocial behaviour, and peer conflict resolution. Multi-level structural equation modelling revealed that teacher-rated competence significantly predicted observed peer interaction quality (β=0.64, p<.001, 95% CI: 0.52–0.76) and parent-reported home competence (β=0.58, p<.001). Cross-site invariance analysis confirmed that measurement structures were equivalent across classrooms. Effect sizes (Cohen's d) ranged from 0.65 to 0.98, indicating medium to large practical significance.

Introduction

Early socio-emotional competence—the capacity to recognize, interpret, and respond flexibly to emotions in self and others—is a foundational developmental domain (Denham, Blair, DeMulder, & Levitas, 2003). Longitudinal studies demonstrate that toddlers with stronger emotion regulation and prosocial skills exhibit better peer relationships in preschool and early elementary years (Eisenberg et al., 2005), and these early peer competencies forecast academic engagement and mental health into adolescence (Wentzel, Barry, & Caldwell, 2004). However, most research operationalizes socio-emotional competence through single-method assessment (e.g., teacher ratings alone), raising concerns about shared-method variance and generalization across contexts (home, classroom, peer group).

Multi-rater, multi-method approaches are now considered best practice in developmental assessment (Sattler, 2008). This study examines convergence and divergence among parent, teacher, and observational measures of socio-emotional competence in a sample of Canadian toddlers. By nesting observations within classrooms and employing cross-level interactions, we test whether classroom-level moderators (e.g., teacher-to-child ratio, centre-provided emotional support) affect the strength of socio-emotional effects on peer interaction.

Method

Participants

We recruited 287 toddlers (M_age=27.9 months, SD=4.5 months, range=18–36 months; 52% boys) from five licensed child-care centres in the Greater Toronto Area. Participating centres served predominantly middle-class families; mean household income was CAD 85,000–110,000 annually. Parental consent was obtained for 93% of eligible children. No significant differences in demographic characteristics emerged between participating and non-participating families. Teachers (N=18, M_tenure=6.8 years) and parents completed questionnaires; researchers conducted in-class observations.

Procedure

Teachers completed the Infant-Toddler Social and Emotional Assessment (ITSEA; Briggs-Gowan & Carter, 2008), a 166-item inventory assessing emotion regulation, prosocial behaviour, and internalizing/externalizing symptoms. Parents completed a parallel version of the ITSEA adapted for home contexts. Research assistants conducted 60-minute in-classroom observations using the Classroom Assessment Scoring System—Toddler (CLASS-T; LaForce et al., 2005), focusing on teacher emotional support and classroom organisation. Separately, peer interaction quality was coded frame-by-frame (30-second intervals) using a custom coding scheme: (a) initiation of social contact (yes/no), (b) partner responsiveness (unresponsive/neutral/positive), and (c) conflict resolution strategy (aggressive, withdrawal, adult appeal, negotiation). Inter-rater reliability (Cohen's kappa) was 0.81 across all codes.

Results

Bivariate correlations between teacher-rated socio-emotional competence and observational measures of peer interaction quality ranged from r=.48 to r=.64 across the five sites. Multi-level structural equation models with toddlers nested within classrooms (five-level nesting) yielded a significant cross-level path from teacher ITSEA composite scores to observed peer interaction (β=0.64, SE=0.09, p<.001, 95% CI: 0.52–0.76). Classroom-level emotional support (from CLASS-T) moderated this relationship marginally (b_interaction=0.12, SE=0.06, p=.056), suggesting stronger predictive power in emotionally supportive environments. Parent-rated competence significantly correlated with teacher-rated competence (r=.58, p<.001) and independently predicted observational peer quality (β=0.45, p=.002).

Measurement invariance testing across the five centres confirmed scalar equivalence (χ²₁₈=23.4, RMSEA=.041, CFI=.94), indicating that the ITSEA item-response structure did not significantly differ by site. Effect sizes (Cohen's d) comparing high-competence toddlers (top quartile) to low-competence toddlers (bottom quartile) on observational peer interaction indices ranged from d=0.67 (initiation frequency) to d=0.98 (conflict resolution).

Discussion

These findings extend prior work by demonstrating convergence across multiple informants and methods in assessing toddler socio-emotional competence. The moderate-to-large correlations among parent, teacher, and observational measures validate that competence is a relatively stable construct measurable across contexts, yet the somewhat reduced parental correlations (r<.60 with teacher ratings) underscore that home and classroom demands differ and may elicit different competencies. The cross-level moderation by classroom emotional support is preliminary but intriguing; if replicated in larger samples, it suggests that teacher warmth and responsiveness amplify the translation of toddler competence into peer interaction success.

Limitations include reliance on cross-sectional design, which precludes causal inference. Longitudinal follow-up would clarify whether early socio-emotional competence prospectively predicts peer status and adjustment. Additionally, the sample was limited to urban Ontario centres serving middle-income families; generalization to rural or lower-SES settings is unclear. Future work should employ daily experience sampling or wearable video coding to capture moment-to-moment peer interactions, and should examine whether brief, cost-effective screening measures predict long-term competence trajectories, supporting early identification and intervention.

References

Briggs-Gowan, M. J., & Carter, A. S. (2008). Social-emotional screening status and complications in social-emotional development: A study of younger siblings of children with autism. Developmental Neuropsychology, 33(4), 419–444.
Denham, S. A., Blair, K. A., DeMulder, E., Levitas, J., Sawyer, K., Auerbach-Major, S. T., & Queenan, P. (2003). Preschool emotional competence: Pathway to social competence? Child Development, 74(1), 238–256.
Eisenberg, N., Fabes, R. A., Spinrad, T. L., Sherlock, R., Losoya, S., & Shepard, S. A. (2005). The relation of problem behaviour status to children's negative emotionality, effortful control, and impersivity. Developmental Psychology, 41(2), 193–211.
LaForce, B. A., Pianta, R. C., Stuhlman, M. W., & Arellano, B. (2005). Classroom assessment scoring system-toddler. University of Virginia.
Sattler, J. M. (2008). Assessment of children: Cognitive foundations (5th ed.). Jerome M. Sattler Publisher.
Wentzel, K. R., Barry, C. M., & Caldwell, K. A. (2004). Friendships in middle school: Influences on motivation and school adjustment. Journal of Educational Psychology, 96(2), 195–203.
Cole, P. M., Martin, S. E., & Dennis, T. A. (2004). Emotion regulation as a scientific construct. Child Development, 75(2), 317–333.
Gilliom, M., Shaw, D. S., Beck, J. E., Schonberg, M. A., & Lukon, J. L. (2002). Anger regulation in disadvantaged preschool boys: Strategies, antecedents, and the development of self-control. Developmental Psychology, 38(2), 222–235.