4 For better, for worse, or both? 123 worried about the future; and (3) I was nervous. The internal consistency of the scale was good at the between-family (ω b = .87) and sufficient at the within-family level (ω w = .71). Moreover, MCFA indicated good fit for a one-factor model of the GAD subscale (CFI = .92, TLI = .90, RMSEA = .04) and sufficiently high factor loadings: between .34 and .66 at the within-family level and between .71 and .97 at the between-family level. Meta-analytic work has shown that the SCARED is a valid self-report to screen for anxiety symptoms in adolescents (Hale et al., 2011). Adolescent Environmental Sensitivity Environmental sensitivity of the adolescent was assessed at T1 with the 12-item Highly Sensitive Child Scale (12-item HSC; Pluess et al., 2018; Weyn et al., 2021). The HSC aims at measuring trait environmental sensitivity, specifically sensory processing sensitivity, which is characterized by greater awareness of subtle environmental cues, behavioral inhibition, deeper cognitive processing, higher emotional and physiological responsivity, and ease of overstimulation (Aron et al., 2012; Pluess, 2015). The scale consists of three subscales: Ease of Excitation (5 items, e.g., “I get nervous when I have to do a lot in little time”), Aesthetic Sensitivity (4 items, e.g., “I notice when small things have changed in my environment”), and Low Sensory Threshold (3 items, e.g., “I don’t like loud noises”) (Weyn et al., 2021). The 12 items of the scale were rated on a scale from 1 (not at all) to 7 (extremely). In line with earlier work showing good psychometric properties in adolescent samples (Weyn et al., 2021), the internal consistency in the current sample was good (α = .80). Moreover, a CFA of a bifactor model (i.e., a general sensitivity factor and three group factors) showed a good fit (CFI = .96, TLI = .94, RMSEA = .05), with the general factor loadings between .23 and .72. In the current study we used the total scale score, in which a higher score indicates higher sensitivity to both negative and positive environmental influences. Preregistered Statistical Analyses To estimate parenting effects for each adolescent separately, in addition to the average effects in the sample, Dynamic Structural Equation Modelling (DSEM; Asparouhov et al., 2018; Muthén & Muthén, 2020) was employed, which combines the strengths of SEM, multilevel, and N = 1 timeseries. We preregistered our hypotheses and analytical approach (https://osf.io/8egxf/), which was based on similar preregistrations of (Beyens et al., 2021) and Bülow, Van Roekel et al. (2022). First, we checked whether the mean-level structure of the data was stationary. Because measurement occasion explained less than 10% of the variance (0.7% to 2.4%) in
RkJQdWJsaXNoZXIy MTk4NDMw