Crystal Smit

Chapter 5 108 5.1) as with the influence agents (Markland & Tobin, 2004; Smit et al., 2018). A total score for the peers’ intrinsic motivation was constructed by averaging the three items, which demonstrated adequate internal consistency (Cronbach’s α T1 = .83; Cronbach’s α T2 = .87). Their perceived social support to drink water was measured at T1 and T2, using four items derived from a broader questionnaire on healthy behaviors (Kiernan et al., 2012), each rated on a 6-point scale ranging from 1 = ‘never’ to 6 = ‘always’. A total score for perceived social support was constructed by averaging the four items, which demonstrated adequate internal consistency (Cronbach’s α T1 = .79; Cronbach’s α T2 = .86). The peers’ perceived social norm was assessed at T1 and T2, based on their beliefs about how often one’s peers drink water (i.e., descriptive norm; response options ranged from 1 = ‘never’ to 6 = ‘always’) and their beliefs about the approval of one’s peer regarding drinking water (i.e., injunctive norm; response options ranged from 1 = ‘no, certainly not’ to 6 = ‘yes, certainly’; Smit et al., 2018). To evaluate the self-determination theory-based techniques that were implemented in the training to support the influence agents in motivating their peers, we measured on a 6-point scale (ranging from 1 = ‘never’ to 6 = ‘always’) whether the influence agents applied the water-promoting strategies discussed in the training: (1) drinking water themselves, (2) talking about water at school or home, (3) talking and (4) forwarding short videos about water on a social media platform ( Social Buzz ). The influence agents also provided responses to several open-ended questions concerning the strategies they implemented to motivate their peers to drink more water and their experiences therein. Statistical Analyses All data were analyzed using SPSS version 25 (SPSS, Inc., Chicago, IL, US). Significance was set at p < .05. For the close-ended (quantitative) data related to the first and third research aim, we computed both means ( M ) and standard deviations ( SD ) for the general experience measurements (i.e., influence agents’ enjoyment and perceived autonomy support) and water-promoting strategies, as well as the percentage (%) of influence agents with positive (score of 3 or