Changes in social responsibility, teachers’ socio-emotional learning strategies and aggressive behaviors: A three-month longitudinal study

• In Event: Contextual Influences on Youth Character Development: Evidence from Kenya, Brazil, and Canada

Abstract

Rationale: Interventions aimed at improving social responsibility have largely been tailored to the classroom or community level (Leadbeater, Thompson, & Sukhawathanakul, 2016). However, when assessing the impact of such interventions, the analyses have primarily focused on individual level change. In the current paper, we present findings of the effects of a community-based intervention that took place in Curitiba, Brazil, by examining the factors that explain classroom level differences in the change in social responsibility over time.

Method: Data was collected from 44 schools, with 8,338 children aged between 7 to 17 (Mage = 9.84; SD = 1.19) in 164 classes across two time points (three months apart). Ultimately, we used data from 1,850 children for whom parental consent was provided and had relevant data from both time points. Children provided self-reports of social responsibility (ex.: “I search for ways to help and include others”, α(T1) = .60, α(T2) = .65), aggressive behavior (ex.: “I push others”, α(T1) = .79, α(T2) = .67) and peer victimization (ex.: “Other kids threaten to hurt me”, α(T1) = .81, α(T2) = .83). Children also rated their teacher’s socio-emotional learning strategies (ex.: “Students are taught to understand how others think and feel”, α(T1) = .65), classroom structure (ex.: “Students know how they expect them to behave at school.”, α(T1) = .86) and support (ex.: “Teachers listen to students when they have problems.”, α(T1) = .86). Multi-level modeling was used to assess classroom level change in social responsibility while structural equation modeling was used to explain the differences across classrooms.

Results: Multilevel modeling of the classroom level change in social responsibility revealed an increase over time (b = .10, t(2294) = 6.06, p < .05) with significant differences across the classrooms (χ2(2294) = 4435.72, p < .05). Classroom structure (β = .25, z = 2.09, p < .05) and support (β = .37, z = 3.19, p < .05) were both positively related to teacher socio-emotional learning. In addition, classrooms that increased in social responsibility were also those that started with more socio-emotional learning (β = .36, z = 5.14, p < .05) and classroom structure (β = .36, z = 5.13, p < .05). Part of the association between earlier socio-emotional learning and less aggression later on was explained by the indirect effect of classroom level increases in social responsibility (b(indirect) = -.02, z = -2.15, p < .05). The resulting model was a good fit to the data (χ2(8) = 10.25, p > .05, CFI = .96, RMSEA = .08, SRMR = .10).

Discussion: The findings from the current paper provide two pieces of compelling evidence for the mechanism by which community-based interventions may improve individual behavior, in this case as measured by social responsibility. Classrooms that are high in structure and possibly support as well, may create “permissive substrates” for socio-emotional learning strategies from which improvements in social responsibility can be derived. Moreover, the association between teacher’s socio-emotional learning strategies and lower aggressive behavior later is indirectly related to increase in social responsibility.

Authors

• Jonathan Bruce Santo, University of Nebraska - Omaha
  Presenting Author

• Josafá M. Da Cunha, Federal University of Parana
  Non-Presenting Author