Maartje Boer

CHAPTER 6 170 Table 6.2 Descriptive Analyses (n observations = 6,327, n individuals = 2,109) SMU problems SMU intensity Depressive symptoms Life satisfaction Upward social comparisons Cyber- victimizaton Subjective school achievements Face-to-face- contact Level 1 (observations) β (SE) β (SE) β (SE) β (SE) β (SE) β (SE) β (SE) β (SE) T2 (ref.: T1) -0.053 (0.063) 0.309*** (0.053) 0.200*** (0.037) -0.332*** (0.039) -0.020 (0.043) 0.062 (0.038) -0.014 (0.050) -0.092* (0.044) T3 (ref.: T1) -0.106 (0.099) 0.324*** (0.075) 0.503*** (0.061) -0.519*** (0.051) 0.090 (0.071) 0.056 (0.051) 0.011 (0.046) -0.156** (0.049) Level 2 (adolescents) β (SE) β (SE) β (SE) β (SE) β (SE) β (SE) β (SE) β (SE) Girl 1 (ref.: boy) 0.163** (0.060) 0.418*** (0.048) 0.439*** (0.052) -0.119* (0.055) 0.382*** (0.055) -0.159** (0.058) 0.122 (0.063) 0.217*** (0.052) Pre-vocational educational level 2 (ref.: pre-university educational level) 0.431*** (0.091) 0.396*** (0.081) -0.076 (0.085) -0.170 (0.087) -0.254* (0.098) 0.328*** (0.080) -0.496*** (0.107) 0.426*** (0.073) Intermediate educational level 2 (ref.: pre-university educational level) 0.199* (0.098) 0.131 (0.091) 0.130 (0.088) -0.169 (0.095) -0.141 (0.105) 0.197* (0.090) -0.480*** (0.113) 0.223** (0.080) Immigrant background 3 (ref.: native) 0.073 (0.059) -0.070 (0.057) -0.114 (0.059) 0.015 (0.067) -0.036 (0.065) -0.048 (0.065) -0.162* (0.074) -0.275*** (0.062) Random parameters Estimate ( SE ) Estimate ( SE ) Estimate ( SE ) Estimate ( SE ) Estimate ( SE ) Estimate ( SE ) Estimate ( SE ) Estimate ( SE ) Residual variance within adolescents 0.134*** (0.021) 0.422*** (0.029) 0.147*** (0.013) 0.545*** (0.019) 0.212*** (0.009) 0.551*** (0.038) 0.579*** (0.028) 0.435*** (0.024) Residual variance between adolescents 0.296*** (0.042) 1.131*** (0.068) 0.214*** (0.020) 0.515*** (0.036) 0.194*** (0.013) 0.438*** (0.037) 0.305*** (0.029) 0.624*** (0.036) Notes. β = STDY-standardized, because all covariates were binary; SE = standard error; Ref. = reference category. The models were examined with multivariate multilevel regression analyses (level 1 = observations, level 2 = individuals) using the 20 imputed datasets of plausible values. The models were estimated with Maximum Likelihood with robust standard errors. 1 Based on question whether respondent was a boy or a girl. 2 Based on question which level of education the respondent was following. Educational level was determined using the most recent reported educational level. 3 Based on question in which country parent(s) were born. *** p < 0.001; ** p < 0.01; * p < 0.05.