A multi-method exploration into the social networks of young teenagers and their physical activity behavior

• Published in: BMC public health 2021-01, Vol.21 (1), p.77-77, Article 77
• Main Authors: Montgomery, Shannon C, Donnelly, Michael, Badham, Jennifer, Kee, Frank, Dunne, Laura, Hunter, Ruth F
• Format: Article
• Language: English
• Subjects: Adolescence; Adolescent; Adolescent Behavior; Adolescents; Cross-Sectional Studies; Data collection; Demographic aspects; Exercise; Female; Females; Friends; Gender differences; Health aspects; Humans; Investigations; Male; Males; Multi-methods analysis; Norms; Physical activity; Physical fitness; Qualitative analysis; Regression analysis; Regression models; Schools; Social aspects; Social behavior; Social interactions; Social Networking; Social networks; Social organization; Social Support; Teenagers; Youth
• ISSN: 1471-2458; 1471-2458
• DOI: 10.1186/s12889-020-10081-0

There is a need for novel interventions to target inadequate levels of adolescent physical activity behavior. Previous research indicates that better understanding of the processes by which social networks influence physical activity behavior in adolescents may be useful to enhance intervention design. This study used a multi-methods approach to aid our understanding about the role of social networks for adolescent physical activity behavior. The quantitative phase of data collection was analyzed using a three-step linear regression model using cross-sectional data from the WiSe study (n = 529 participants, 48.6% female, mean age 14.38 years (SD 0.32)). A demographically reflective sub-sample of schools were invited to take part in the qualitative phase, which involved focus group discussions. Thematic analysis was used to explore findings from the quantitative phase in greater depth, and identify other themes pertaining to the association between social networks and physical activity behavior. Males' physical activity behavior was predicted by their friend group (0.46, p = 0.007) whereas females' physical activity was predicted by their best friend (0.21, p = 0.03). The three main findings that were uncovered by the regression analysis were explored during the qualitative phase: 1) friends have similar physical activity behaviors; 2) friendship social networks may influence differently early adolescent male and female physical activity behavior; 3) popularity and sociability were not associated with physical activity behavior. Two additional themes emerged from the analysis of focus group data: 4) social norms and 5) external factors that may impact the relationship between adolescent physical activity behavior and social networks. The investigation of the interplay between the findings from each phase of the inquiry indicated that social networks influence in different ways and to different degrees the physical activity of adolescent males and females. In turn, these insights point to the need for a systematic tailoring process for the development and implementation of physical activity behavior interventions.