2.5. Data analysis plan
To examine whether fathers reported more challenging parenting behaviour than mothers (hypothesis 1), a paired-samples t-test was conducted. The relationship between parental anxiety and challenging parenting behaviour (hypothesis 2) was examined through hierarchical multiple regression analyses (MRA), whilst controlling for potential covariates (e.g. parental age). The hypothesis (3), that parents would report higher levels of challenging parenting behaviour for their male children compared to their female children, was examined via independent samples t-test.
To examine the association between challenging parenting behaviour and child anxiety (hypotheses 4 and 5), and whether child gender moderates this association, two dependent variables were examined across a series of regression analyses. Both dependent variables were measures of child anxiety: mother report of child anxiety symptoms on the PAS (a continuous measure); presence of an anxiety diagnosis based on the ADIS-IV-P (a dichotomous variable-anxiety group). For child anxiety symptoms, hierarchical MRA was used. For anxiety group, logistic regression was used. Of primary interest were the main effects of challenging parenting behaviour and child gender on child anxiety, as well as the interaction between child gender and challenging parenting behaviour.