Crystal Smit

Chapter 2 40 increase in their water drinking (H1) and a decrease in their SSB consumption (H2) over time. These findings are in line with the growing body of peer influence research demonstrating that children model the consumption behaviors of their peers (Bevelander et al., 2012; Cruwys et al., 2015). That is, the children could have consumed more or less according to a potential social norm that was set by a peer. Previous studies have shown that encouragement by peers can increase acceptance and consumption of foods (Hendy, 2002; Hendy & Raudenbush, 2000) and even seeing peers consume food children do not like can increase their preferences and intake of these foods (Birch, 1980). This could also have been the case in the present study, given the fact that the influence agents successfully promoted water consumption which also led to their peers reporting a decrease in their SSB consumption. Nevertheless, we do not know whether children replaced their SSB consumption by drinking (more) water, or that the children who were exposed to the social network-based intervention modeled the influence agents in drinking less SSBs. Further research is needed to unravel by which of these mechanisms the decrease of SSB consumption could be explained. Notably, this was the first intervention study aimed at water drinking that incorporated the social modeling mechanism in conjunction with peer status among primary school children. The findings suggest that it is important to take the status of peers into account when targeting behavioral change processes in social networks. Previous experimental studies in adolescents have shown similar effects of peers on alcohol consumption (Larsen, Engels, Souren, Granic, & Overbeek, 2010; Teunissen et al., 2012), where high status peers were more influential than low status peers in reducing their willingness to drink alcoholic beverages (Teunissen et al., 2012). The adolescents accepted and internalized the anti-alcohol norms of their popular peers (Teunissen et al., 2012). This could also have been the case in the present study; however, we can only speculate whether the selected influence agents were more influential in improving children’s water consumption behaviors. Nevertheless, the children in the intervention reported an increase in their water drinking and a decrease in their SSB consumption