An ego-network examination of friendship networks and the physical activity and screen time of grade five children


Developmental theory suggests friends emerge as important influence agents in middle-to-late childhood. It is currently unclear whether friends influence physical activity (PA) and screen time at this age. This study examined whether characteristics of the friendship network are associated with pedometer-measured PA and screen time of children. Participants (N = 801) from A Project Promoting healthy Living for Everyone in Schools (APPLE Schools) in 32 schools wore pedometers for nine consecutive days. Parents reported sociodemographic information and children reported their usual screen time and perceived friendship influences (i.e., co-participation in screen time, screen restriction support, support for PA). Children also listed the first and last names of their close friends in their school and grade. Social network variables, including in-degree and out-degree friendships, average friend behaviour (steps, screen time), diversity of friend behaviour (steps, screen time), proportion of same gender friends, and in-isolate and out-isolate status, were calculated. Path analyses controlled for clustering within schools, parent education, and child weight status. Friend support for PA, friend PA, and in-degree friendships were associated with ego PA in males, and friend PA was associated with PA in females. Friend screen co-participation was positively associated with PA in males and females, and friend screen restriction support was negatively associated with screen time in males. In-isolate status was associated with lower screen time in females and in- and out-isolate status was associated with lower PA in males. These findings can be used to incorporate friendship and network position components into health promotion programming.

Acknowledgments: We thank all of the students, parents and schools for their participation in the APPLE Schools evaluation, and the evaluation assistants and school health facilitators for their contributions to data collection. Thank you to Wendy Davis and Erin Faught for their work in coordinating the evaluation, Pamela Bailey for organizing logistics, and Connie Lu for data management.