The co-occurrence of aggression and prosociality in kindergartners: A population-level latent class analysis

• In Event: The Developmental Intertwining of Aggression and Morality: Actions, Judgments, and Responses

Abstract

Traditionally, aggression (e.g., harming others) has been viewed as a maladaptive form of psychopathology reflecting deficits in morality, whereas prosociality (e.g., helping others) has been considered a key indicator of socio-moral competence. This view has been called into question in recent years, however, in light of research indicating that some aggressive children are prosocial, dominant, and socially adept in their peer interactions (Hawley, 2014). This aligns with a growing literature showing inconsistent relations between aggressive and prosocial conduct in early childhood; although some studies have found the two behaviors to be negatively correlated, others have found null or positive associations (Hay et al., 2021). This suggests there is considerable heterogeneity in the degree to which young children exhibit aggressive and prosocial tendencies.
The present study explored this heterogeneity in kindergarteners living in Ontario, Canada. Kindergarten is an insightful developmental period to examine these processes given that individual differences in aggressive and prosocial conduct begin to stabilize between 4 and 6 years of age. Using a person-centered methodology (latent class analysis; LCA) with population-level teacher-report data, our main aim was to determine the extent to which children who behave aggressively also exhibit a reduced capacity for prosocial interactions.
Data come from the 2018 Ontario implementation of the Early Development Instrument (EDI), a teacher-report measure of developmental health collected on children enrolled in publicly funded kindergartens across Canada. The current analyses included 130,489 children (Mage= 5.69 years, Range= 4.52-6.50; 48% female), representing nearly 90% of Ontario’s 2018 kindergarten population. Teachers reported on the frequency of aggression (5-items) and prosociality (8-items) using a 3-point rating scale (0= never, 1=sometimes, 2=often). See Table 1 for complete items frequencies.
Given well-established sex differences in teacher reports of both behaviors, we conducted LCA models on all items for males and females separately. A 5-class solution was optimal for both sexes (see Figure 1). Approximately 57% of males belonged to non-aggressive/prosocial classes. Approximately 23% of males were classified into one of two aggressive classes (i.e., > 50% average probability of scoring “sometimes” or “often” across items), yet only a minority¬¬¬¬––representing 9% of all males––were also low in prosociality (< 50% probability of sometimes/often). The final 20% of boys belonged to an “uninvolved” class low in both. Approximately 82% of females belonged to non-aggressive/prosocial classes. Only 6% of females were classified as aggressive, and they were equally as likely to behave prosocially as non-aggressive/prosocial girls. The remaining 12% of females fit a similar “uninvolved” profile of being relatively low in both behaviors.
This study constitutes one of the largest investigations into the co-occurrence of aggressive and prosocial behaviour in early childhood. With the exception of a small group of males, most aggressive kindergartners were also highly prosocial, and the majority of students who lacked prosocial skills rarely engaged in aggression. These findings cast doubt on the traditional conceptualization of aggression and prosociality as opposite ends of a single continuum and underscore the utility of person-centered approaches for understanding the complex interplay between these two behaviors.

Authors

• Marc Jambon, Wilfrid Laurier University
  Presenting Author

• Eric Duku

• Magdalena Z Janus, McMaster University