Crystal Smit

Chapter 2 38 This indicates that changes in SSB consumption differed for children in the intervention and control conditions 1 . Posthoc contrast comparisons showed a significant difference in SSB consumption over time for the intervention condition ( M baseline = 1.28 ± SEM .07; M post-intervention 1.06 ± SEM .06; p < .001), but not for the control condition ( M baseline = 1.18 ± SEM .07; M post-intervention 1.15 ± SEM .06; p = .596). This indicates that children who were exposed to the social network-based intervention reported drinking significantly less SSB over time than children who were not exposed to the intervention. Water Drinking Intentions There was a marginal significant main effect for condition ( F (1,206) = 3.34, p = .069, p η 2 = .02) and no significant main effect for time ( F (1,206) = .56, p = .454) on water drinking intentions. More importantly, there was a significant interaction effect between condition and time ( F (1,206) = 3.93, p = .049, p η 2 = .02) on water drinking intentions, indicating that changes in water drinking intentions differed for children in the intervention and control conditions. Posthoc contrast comparisons showed a significant difference in water drinking intentions over time for the control condition ( M baseline = 2.75 ± SEM .07; M post-intervention = 2.60 ± SEM .07; p = .026), but not for the experimental condition ( M baseline = 2.83 ± SEM .07; M post-intervention = 2.86 ± SEM .07; p = .576). The findings indicated that children who were exposed to the intervention did not report a change in their water drinking intentions over time, whereas the children in the control condition reported a decrease in their water drinking intentions. Additional Analyses The same analyses were performed in the sample excluding the influence agents ( n = 185) to investigate whether the effect of the intervention was driven by the behavior and intentions of the influence agents. Similar results were found for the multivariate model, with a significant interaction effect between condition and time ( V = .06, F (3,179) = 3.89, p = .010, p η 2 = .06) on the three dependent variables 1 The SSB consumption measure at both baseline and post-intervention were positively skewed. However, the results were identical when using the raw and logarithmically transformed scores; therefore, the reported analyses are based on the raw scores.