Questions:

How were the data analyzed?

What inferential analyses were performed?

Which statistics and/or tests of significance were used?

Here is the statistical analysis:

The baseline characteristics of the intervention group and the wait-list control group were compared by analysis of covariance for continuous variables and chi-square tests for categorical variables. The baseline factors included the age of the children and the parents, the sex of the children and the parents, the children's medication status, and the pretest scores of SWAN, CBCL, PSI, WHO-5, and ASRS. Intervention group participants were assessed at baseline (T1) and after the intervention (T2). Wait-list control group participants were assessed at the same time with the intervention group, and would receive the same program after posttest of intervention groups. The effects of FBMI were tested by ANCOVA, comparing the FBMI group (Arm 1) with the wait-list control group (Arm 2). All analyses were carried out according to the intention-to-treat approach. The participants' missing values were imputed using the last-observation-carried-forward method. A two-sidedpvalue of .05 or less was considered to be statistically significant. In the case of significant results, effect sizes (Cohen'sd) were calculated.Cohen (1988)suggested thatd= 0.2 be considered a small effect size, 0.5 a medium effect size, and 0.8 a large effect size.

This study further attempted to explore the mediating effects of child attention in other dependent variables. The PROCESS macro was used to test the mediating effects of child attention on the relationship between group difference as the independent variable, and parents' stress or well-being or children's behavioral problems as dependent variables (Hayes, 2013). Bootstrapped estimates of confidence intervals (CIs) for indirect effects were calculated. It is bias-corrected because this approach does not assume distribution normality of sampled indirect effects (Preacher, Rucker, & Hayes, 2007). If 95% CIs do not encapsulate 0, they are considered significant and mediating effects exist. All analyses controlled for the child age and pretest value of the corresponding dependent variable.

SPSS version 23 was used to administer the above statistical tests.