6 Like no other? 183 mapping relations among variables that share measure or method variance (e.g., positive and negative affect or associations among parenting practices). Explaining Differences between Subgroups and Individual Families In the second step of the pre-registered analysis, differences between subgroups were explored in terms of personal and family characteristics. That is, subgroups were compared based on several variables: average levels of parenting and affect variables, indicators of adolescent psychological functioning (i.e., depressive and anxiety symptoms, and self-esteem), adolescent demographic characteristics (i.e., age, sex, educational level), legitimacy beliefs about parental authority, and personality traits (i.e., environmental sensitivity, neuroticism). More information about these measures is provided in the online Supplementary Information. To test for group differences in these variables, t-tests (with continuous variables, e.g., age) and chi-square tests (with categorical variables, e.g., sex) were conducted. If no subgroups were identified that shared similar parent-adolescent dynamics, it was pre-registered to describe differences between individual families instead. Specifically, the correlation between the density of parenting-affect associations and the variables described above was calculated. Density was calculated for each family by dividing the number of parenting-affect associations by the total number of associations in their family-specific network model (excluding autoregressive effects). Parenting-affect density thus reflects the extent to which temporal associations between perceived parenting and adolescent affect contribute to the overall family-specific network. RESULTS Intraclass and Bivariate Correlations To assess the extent to which parenting practices and affect fluctuated from day to day, intraclass correlation coefficients were calculated (ICCs; see Figure 1 for an example of the data). The ICCs ranged between .49 and .64 (see Table 1). In other words, 49–64% of the variance in perceived parenting practices and adolescent affective well-being was due to stable differences between families. Day-to-day fluctuations within families accounted for the remaining 36%–51% of the variance. Descriptive statistics and bivariate correlations among parenting and adolescent affect variables are provided in Table 1. Perceived parenting practices and adolescents’ affective well-being were weakly to moderately correlated at the within-family level. On average,
RkJQdWJsaXNoZXIy MTk4NDMw