Skip to main content

Table 1 Results of the final mixed-effects regression models for aggression and depression

From: A 9-year longitudinal study on trajectories of aggressive and depressive symptoms in male and female children with overweight

Parameters Aggression Depression
Estimate S.E. p Estimate S.E. p
Intercept 2.8 0.35 < .0001 2.11 0.28 < .001
Time − 0.27 0.26 0.3028 − 0.57 0.2 0.0053
Time^2 − 0.36 0.1 0.0003 − 0.003 0.08 0.9632
Age 0.19 0.08 0.018 0.21 0.06 0.0013
Sex − 0.46 0.31 0.1433 − 0.19 0.24 0.4361
Group 0.35 0.31 0.2652 0.18 0.24 0.4603
Age*sex − 0.09 0.08 0.2372 − 0.05 0.06 0.3747
Age*group 0.01 0.08 0.8732 − 0.08 0.06 0.2225
Sex*group − 0.47 0.43 0.2775 0.68 0.33 0.0434
Time^2*age 0.03 0.01 0.0013 − 0.004 0.01 0.5721
Time^2*sex 0.18 0.11 0.1034 0.1 0.09 0.2581
Time^2*group 0.27 0.11 0.0185 0.5 0.09 < .001
Age*sex*group − 0.12 0.11 0.2795 0.12 0.08 0.1642
Time^2*age*sex − 0.01 0.01 0.1658 − 0.01 0.01 0.2206
Time^2*age*group − 0.04 0.01 0.0009 − 0.07 0.01 < .001
Time^2*sex*group 0.56 0.16 0.0004 − 0.52 0.12 < .001
Time^2*age*sex*group − 0.09 0.02 < .001 0.08 0.01 < .001
Model fit
Intercept VAR 0.13 0.14
Residual VAR 0.26 0.14
Residual SD 0.51 0.38
AIC 1339.17 1029.52
BIC 1448.5 1138.85
− 2LL 1291.17 981.52
  1. Significant estimates are italic (p < 0.05). Aggression and depression scores were transformed to lower skewness. Time^2 is the quadratic term for time. Age was centered on the grand mean. Sex was binary coded: 0 = female, 1 = male. Group was binary coded: 0 = control group, 1 = clinical group. VAR variance, SD standard deviation, AIC Akaike information criteria, BIC Bayesian information criterion. − 2LL = negative two log likelihood