The importance of social connections in combating compulsive exercise among group-exercise participants: a network analysis

Purpose: This paper aims to use social network analysis (SNA) to determine whether compulsive exercise (CE) was related to social connections and network position among participants of group-exercise programs. Design/methodology/approach: Members from two group-exercise programs (Gym 1: n= 103; Gym 2: n= 56) completed an online survey measuring their social connections within the program, CE, depressive symptoms and sense of belonging. Network position was calculated for each person based on network centrality scores (i.e. closeness, eigenvector centrality). Linear network autocorrelation models determined whether respondents reported similar CE as their network ties (i.e., network effects) and whether network position was related to CE in these networks. Findings: Eigenvector centrality (i.e., being connected to popular/important people within the network; Gym 1: parameter estimate [PE] = 0.51, p< 0.01, Gym 2: PE = 0.39, p= 0.02) and network effects (i.e. having similar CE scores as direct network ties; Gym 1: PE = 0.07, p< 0.01, Gym 2: PE = 0.19, p< 0.01) were related to CE among participants in these programs. Originality/value: This study builds on existing SNA research suggesting the importance of social connections and network position on CE, and, to the best of the authors’ knowledge, is the first to explore these effects among group-exercise participants. This study describes how the social environment can impact, both positively and negatively, someone’s susceptibility for CE and supports fostering social connections within group-exercise programs as a way to potentially combat harmful CE among its participants.