Skip to main content

Table 3 Bootstrap mediation and moderated mediation effect

From: Mediation and moderation analyses: exploring the complex pathways between hope and quality of life among patients with schizophrenia

  Point Estimate Product of confidents BOOSTRAP 5000 TIMES 95% CI
bias corrected percentile
S.E. Est./S.E. Lower Upper Lower Upper
Mediated model Indirect effect
QOL – Resilience - Hope −0.071*** 0.027 −2.659 −0.130 − 0.026 −0.126 − 0.022
QOL – Depression - Hope − 0.142*** 0.034 −4.187 − 0.212 −0.078 − 0.208 −0.073
Total indirect effect
Total indirect effect −0.213** 0.043 −4.929 −0.303 −0.134 − 0.295 −0.127
Moderated mediation model Indirect effect
QOL – resilience - hope −0.046* 0.023 −1.973 −0.104 −0.009 − 0.096 −0.004
QOL–depression - hope −0.121*** 0.033 −3.686 −0.195 −0.064 − 0.188 −0.058
Total indirect effect
Total indirect effect 0.167*** 0.043 −3.875 −0.265 − 0.094 −0.265 − 0.086
Contrasts
Resilience vs depression 0.730* 0.356 2.050 0.057 1.466 0.033 1.438
  1. Percentile 95% CIs for bootstrap distributions are defined using the values that mark the upper and lower 2.5% of each distribution
  2. There were differences between the percentile and bias corrected methods, depending on the size of the paths in the model, with the percentile method generally showing a slight superiority in conditions in which the bias corrected method was slightly liberal
  3. CI confidence interval, QOL quality of life, SE standard error
  4. *p<0.05; **p <0.01; ***p<0.001