Maartje Boer

CHAPTER 8 246 and life satisfaction were explained by adolescents’ average level of upward social comparisons (M7a-f), after which we extended these models with SMU problems (M8a-f). All models were run and read with the MplusAutomation- package in RStudio 1.4.1106 (Hallquist & Wiley, 2018; RStudio Team, 2021). To interpret effect sizes, coefficients in the fixed effects models (M1-4) were STDYX-standardized, whereby 0.1 denoted a small effect, 0.3 moderate, and 0.5 large (Cohen, 1988). All models were estimated using Maximum Likelihood estimation. Codes for all data handling, imputation of missing data, and analyses are publicly available at https://osf.io/3fn2s/. Results Bivariate Associations The intra class correlations of the study variables ranged from 0.195 (SNS posting) to 0.453 (IM viewing). This means that 54.7 to 81.5 percent of the variance of the study variables was related to changes over time, which is considered substantial. Table 8.2 shows the correlations between the study variables. At the within-person level (level 1), adolescents’ life satisfaction decreased over time, with an almost moderate effect size. The more active SMU activities showed different changes over time: While the intensity of SNS responding and SNS liking increased over time with (very) small effect size, SNS posting decreased over time with small effect size, and IM sending did not change over time. The intensity of more passive SMU activities, namely SNS and IM viewing, increased over time with moderate and small effect sizes, respectively. In addition, increased intensity of all SMU activities, except for SNS posting, were associated with decreased life satisfaction, but with very small effect sizes. Increased SMU problems were associated with decreased life satisfaction and increased intensity of all SMU activities except for SNS posting, with small to moderate effect sizes. At the between-person level (level 2), the higher adolescents’ average intensity of more passive SMU activities (i.e., SNS and IM viewing) and IM sending, the lower was their average level of life satisfaction, although effect sizes were small. The higher the adolescents’ average level of upward social comparison, the lower was their average level of life satisfaction, with a large effect size. There was a moderate to large negative correlation between adolescents’ average level of SMU problems and life satisfaction. For all six