RESUMO
While limited evidence is available, preliminary studies highlight the potential health benefits of risky play. However, most of the studies have used subjective methods (i.e., questionnaires) to evaluate children's risky play, which limits their validity and reliability. The purpose of the present study was to examine the relationship between the frequency of risky play and social behavior among Japanese preschoolers by using a valid and reliable method such as direct observation. A total of 32 Japanese preschoolers (71.4 ± 3.5 months old) participated in the study, and their social behaviors were measured by the Strength and Difficulties Questionnaire (SDQ). Data regarding the frequency of risky play was collected through direct observation. Results stated that, in a non-adjusted model, there was no significant association between children's risky play and prosocial behavior. However, the association became significant after adjusting for covariates such as gender, parental employment status, and physical activity. In contrast, there was no significant association between children's risky play and problem behavior (hyperactivity and aggression) after adjusting for covariates. In conclusion, covariates such as parental employment should be considered when examining the benefits of risky play.