Crystal Smit

Chapter 5 122 their peers’ intrinsic motivation and some of them even used strategies that could be considered as the opposite of autonomy-support—controlling strategies (Bartholomew, Ntoumanis, &Thøgersen-Ntoumani, 2009)—for instance, by turning it into a challenge and promising rewards. However, previous research has shown that intrinsic motivation, in particular, is a strong predictor of long-term changes in water consumption (Smit et al., 2018). Therefore, the training activities could be improved by having a greater focus on teaching the influence agents to promote water drinking in a manner that fosters an intrinsically motivating environment for their peers. In relation to this, some influence agents also indicated that they had not sufficiently motivated their peers and had difficulty in doing so. Hence, another refinement in the content of influence agent training would be to provide more specific examples of how to promote water consumption but most importantly also practice real-life situations through role play (Audrey, Cordall, Moore, Cohen, & Campbell, 2004). To apply these refinements to the training and thus possibly make the intervention more effective, the contact moments could be extended. This could also contribute to the relatedness with the researchers and among the influence agents themselves (Teixeira et al., 2020). Unexpectedly, some influence agents indicated that having a secret mission together was the most enjoyable aspect of the training. No part of the training was specifically developed with this intention but having a secret with others— thus group collaboration—may have motivated the influence agents to promote the behavior. By facilitating group collaboration, individuals experience feelings of belonging (i.e., the need for relatedness; Gillison et al., 2019), which may ultimately strengthen their intrinsic motivation (Ryan & Deci, 2000). Therefore, an avenue for refinement could be to emphasize group collaboration among influence agents, by focusing on the secret mission aspect, in order to motivate them to promote the target behavior. In addition, the additional analysis showed that the training approach effected the greatest change for boys and children in higher grade levels. It is therefore essential to make modifications to the training content so that it fits the entire target group. Nevertheless, it is important to note that there appeared to be no differences in changes for children from different levels of family affluence.