123 Short-term outcomes and six-month follow-up of BEAR 6 is consists of 14 items, measured on a 5-point Likert-type scale with a higher total score meaning higher mental well-being. Additionally, the PSQ is used to determine parental experiences with their child, how they interact with their child and how parents feel about their own health. The PSQ consists of 34 items on a 4-point scale, with higher total t-scores indicating more parental stress. Table 2. Inter-rater reliability between coders Cohen’s kappa (κ) Weighted kappa Estimated accuracy Percent agreement Child Total joint engagement .73 1.00 ≥99% 81% Supported joint engagement .59 .90 97% 70% Coordinated joint engagement .79 .00 ≥99% 93% Symbol-infused joint engagement .81 .81 97% 93% Expressive language .94 1.00 ≥99% 96% Parent Scaffolding .57 1.00 ≥99% 70% Symbol highlighting .57 1.00 ≥99% 70% Following in .60 .90 97% 70% Caregiver affect .73 .86 96% 81% Dyad Fluency and connectedness .77 .80 96% 85% Notes. Interpretation of Cohen’s Kappa: values ≤ 0 indicate no agreement , 0.01–0.20 none to slight, 0.21–0.40 fair, 0.41– 0.60 moderate, 0.61–0.80 substantial, and 0.81–1.00 almost perfect agreement (Cohen, 1960). Dyadic Outcome Measure Overall impression of the parent-child interaction. To capture the flow of interaction between parent and child, ‘fluency and connectedness’ of the interaction was scored. Within this item, several variables were rated: the balance between partners, the connection of the partners to each other and their play, and the fluency of the interaction (Adamson & Suma, 2020). Fluency and connectedness were rated on a 7-point Likert scale and were based on the same videotaped dyadic interaction as before, with a score of 1 indicating that no interaction was established, a midpoint of 4 meaning that interaction lacked smoothness and appeared to be largely dominated by one partner and a score of 7 indicating that the interaction was fluid, balanced and sustained (Adamson & Suma, 2020). Inter-rating variability on the ‘fluency and connectedness’ item was κ = .77, indicating substantial agreement amongst coders. Statistical analysis Statistical analyses were conducted using SPSS Statistics (IBM-Corp., version 29), with a significance level set at p = .05. Initially, we intended to employ a generalized linear mixedeffects model for repeated measures with a random effect for cluster to account for the
RkJQdWJsaXNoZXIy MjY0ODMw