Stata version 15.0 se

First, a multivariate linear regression analysis was conducted to examine the association between bullying experience and self-esteem. The independent variable was the frequency of being bullied (i.e., never, occasionally, sometimes, and often). The dependent variable was the total score of self-esteem. Then, the child’s sex (“male” or “female”), BMI (“underweight,” “normal,” or “overweight”), grade (“5th” or “8th”), lack of daily necessities (“yes” or “no”), the experience of financial difficulties (“yes” or “no”), household income, mothers’ educational background, fathers’ educational background, experience of relocation, parents’ marital status, and school location were added as confounders.
Then, we also performed a multivariate linear regression analysis to examine the moderating effect of six types of coping as an interaction term (i.e., victimization * each type of coping) on the association between victimization and self-esteem. These analyses were weighted for response rates of each municipality (i.e., probability weight). A significant level in this study is 5%. All analyses were conducted by using STATA version 15.0 SE (StataCorp. 2017, StataCorp LLC, College Station, TX, USA).

Statistical analysis was undertaken in Stata version 15.0SE [54 ]. When comparing parametric continuous descriptive information looking for differences by both exposure and outcome, either an independent samples t-test or oneway ANOVA was used. When looking at non-parametric continuous data, a Mann-Whitney U test or Kruskal-Wallis test was used. When performing analysis on categorical descriptive variables, a Chi-square or Fisher’s exact test was used to look for significant differences. Logistic regression was used to model the association of interpregnancy weight change with PB. This model was built up by adding in a priori selected covariates in stages until all available variables hypothesised to affect the relationship had been adjusted for. Subgroup analyses were then undertaken as follows: