The review, published in Educational Psychology Review, was conducted by a team at the University of Queensland led by Thuy-Anh Nguyen. It followed PRISMA guidelines, searched four major databases, and applied a three-level random-effects meta-analysis to 196 effect sizes drawn from 38 quantitative studies. The pooled Pearson correlation between teachers’ and students’ social-emotional competence, or SEC, came out at r = .26, a moderate association by conventional psychological standards. In plain terms, classrooms led by teachers who score higher on measures of emotional awareness, self-regulation, empathy, and relationship-building tend to contain students who score higher on similar measures.

What makes the finding notable is not simply its size but its breadth. Previous syntheses had zoomed in on narrow slices of the problem: teacher empathy alone, or early childhood educators only. One earlier meta-analysis of 19,738 early childhood educators found no significant overall link between teacher SEC and student social-emotional outcomes. The new review, by contrast, mapped the full five-domain framework developed by the Collaborative for Academic, Social, and Emotional Learning (CASEL): self-awareness, self-management, social awareness, relationship skills, and responsible decision-making. That structure allowed the researchers to ask not just whether teachers and students are connected, but which specific competencies travel together.

The domain-level results were striking. Teachers’ responsible decision-making showed the strongest observed relationship with students’ overall SEC, at r = .39, significantly outpacing teachers’ self-awareness, self-management, and social awareness. The authors suggest this may reflect the integrative nature of decision-making, which draws on all the other domains at once. A teacher who makes thoughtful, ethical, socially attuned choices in the moment of a classroom dispute is, in effect, deploying the whole competency stack, and modeling it for students in real time. On the student side, relationship skills showed the strongest link to overall teachers’ SEC, at r = .35, while students’ responsible decision-making showed the weakest. Intriguingly, teachers’ own relationship skills were not significantly correlated with students’ overall SEC, a puzzle the authors attribute partly to measurement differences: teachers’ relational skills were often assessed in teacher-student interactions, whereas students’ relational skills were usually measured among peers.

The theoretical scaffolding for these associations comes from several converging frameworks. Emotion socialization theory casts teachers as significant adults from whom children learn culturally acceptable ways of handling feelings, through modeling, explicit instruction, and contingent responses. The co-regulation framework emphasizes how teachers scaffold students’ emotional and behavioral regulation toward shared goals, guided by the teachers’ own beliefs about emotion. Jennings and Greenberg’s prosocial classroom model, meanwhile, proposes that teachers’ SEC operates through multiple channels: their own well-being, the quality of teacher-student relationships, classroom management, the delivery of social-emotional learning programs, and the overall classroom climate. The new meta-analysis provides the first quantitative evidence consistent with a direct association between teachers’ and students’ SEC, rather than exclusively indirect pathways through those mediators.

Mediation studies reviewed in the paper flesh out those channels. Nine of eleven studies that tested mediators found significant effects. Teachers’ self-efficacy, for example, was repeatedly linked to their instructional practices, which in turn predicted students’ self-efficacy, social skills, and growth mindset. The number of SEL lessons delivered, classroom organization quality, and positive teacher-student interactions all emerged as bridges between what teachers bring to the room and what students take away from it. Qualitative accounts reinforce the picture: teachers who completed SEC training described taking more mental health breaks, practicing breathing exercises, and encouraging their students to do the same, reporting classrooms where children felt secure enough to try new ideas.

The association was not uniform, however, and the moderators the team identified may matter as much as the headline number. The link was weaker in samples with higher proportions of students from low socioeconomic backgrounds, suggesting that structural conditions can constrain how teacher and student competencies connect. It grew modestly stronger with each additional year of teaching experience, consistent with the idea that seasoned teachers manage classrooms more efficiently and free up resources for modeling and reinforcing social-emotional skills. It was also stronger in secondary school samples than in primary and early-years settings, possibly because adolescents’ more advanced metacognitive capacities let them recognize and internalize what their teachers model, though the authors caution that younger children may simply be more malleable in ways this design cannot capture.

Measurement choices left clear fingerprints on the results. When students rated their own SEC, or when researchers observed classrooms directly, the teacher-student correlation was stronger than when teachers rated their students or when researcher-administered psychometric tests were used. Similarly, teacher SEC measured by student ratings produced the largest associations, exceeding teacher self-reports. The authors flag common method bias as a likely contributor: when the same perspective, or the same style of self-report, informs both sides of a correlation, the estimate can inflate. Neither student gender nor teacher gender moderated the association, indicating that the underlying mechanisms operate similarly across gender, at least as far as the data allow.

Robustness checks complicate the tidy story. Funnel plot asymmetry and a significant Egger test meant publication bias could not be ruled out, and the pooled estimate should be interpreted with caution. Yet the fail-safe N analysis suggested 2,005 additional null effect sizes would be needed to erase the association, and the effect remained positive and significant across sensitivity analyses that excluded global SEC measures, ambiguous constructs, unpublished studies, and statistical outliers. Heterogeneity was enormous, with roughly 90 percent of variance attributable to differences between studies rather than within them, a signal that the teacher-student SEC relationship is genuinely context-dependent rather than a single fixed quantity.

The authors are careful about what the evidence cannot yet say. Nearly all included studies were correlational, and only six intervention or experimental studies were identified, with mixed results: two mindfulness-based programs improved both teacher and student outcomes, one improved teachers but not students, and two small trials found no significant change in either. Only two studies examined the reverse pathway, in which students’ behavior shapes teachers’ competence, despite emerging evidence that such reciprocity exists. The team’s conclusion is measured but pointed: teachers’ SEC merits far closer attention in research on students’ social-emotional development, and future longitudinal and experimental work should test whether strengthening teachers’ competencies actually cultivates those of their students. If it does, the implications ripple outward, from teacher training curricula to the design of school-based SEL programs, and to a growing recognition that the emotional labor of teaching is not a private burden but a public variable in how the next generation learns to feel, relate, and decide.

Subject of Research: The relationship between teachers' and students' social-emotional competence across K-12 education

Article Title: Relations between Teachers’ and Students’ Social-emotional Competence: A Systematic Review and Multilevel Meta-analysis

Article References: Nguyen, T.-A., Hidajat, T., Zhou, T., Ma, T., Hoang, A., Poed, S., & Edwards, E. J. (2026). Relations between Teachers’ and Students’ Social-emotional Competence: A Systematic Review and Multilevel Meta-analysis. Educational Psychology Review, 38(1), Article 126. https://doi.org/10.1007/s10648-026-10218-4

Image Credits: AI Generated

DOI: 10.1007/s10648-026-10218-4

Keywords: social-emotional competence, teachers, students, meta-analysis, systematic review, CASEL framework, social-emotional learning, educational psychology, teacher-student relationships, classroom climate, moderators, publication bias