83 The feasibility of daily monitoring 4 2.6 Analyses All analyses were performed in RStudio-2022.02.2-458 (RStudio Team, 2022), which runs on R software (version 4.2.0; R Core Team, 2020). The data are available upon request from https://doi.org/10.17026/dans-zkw-fbvs. R scripts are publicly available from https://doi.org/10.17605/OSF.IO/DPBK8. Descriptive statistics on drop-out and compliance rates were obtained for the whole sample and per type of care facility (ambulatory care, residential care, juvenile detention center). Bivariate associations between various demographic and contextual factors and compliance rates were calculated. A Kruskal–Wallis test was used to test whether compliance rates differed between the three care types. Two point-biserial correlations were calculated to assess whether compliance rates differed as a function of gender (dummy coded as 0=female, 1=male) and whether or not the participant completed diaries on their own or a group mobile phone device (dummy coded as 0=group device, 1=own mobile phone). Pearson correlation coefficients were used to determine whether compliance was associated with age, total IQ, the frequency of reminders from caretakers, and the frequency of diary response integration in treatment. We evaluated the significance level of p-values at <.05, after controlling for false positives due to multiple testing using the Hochberg–Benjamini correction method. Results from the structured interviews are presented descriptively. To assess the extent to which the eight standardized diary items captured temporal fluctuations, we computed the mean squared successive difference (MSSD) for all available data-points per individual and per item. The MSSD is a measure of dispersion on a timeseries (Von Neumann et al., 1941). As opposed to the variance statistic, which in insensitive to periodic fluctuations on a timeline, MSSD captures variability between adjacent timepoints while taking into account gradual mean shifts. Specifically, it calculates the average of all differences between successive observations at timepoints i and i +1 on a time series of n timepoints, which is given by: x x n MSSD= ∑ ( − ) −1 i n i i =1 −1 +1 2 A higher MSSD thus reflects an instable pattern with high variability between answers from day-to-day, whereas lower values indicate that the temporal pattern of the answers was stable with relatively few day-to-day variability.
RkJQdWJsaXNoZXIy MTk4NDMw