Table 3.
Timing of mediation effects—comparison of the mediation effect in multilevel structural equation models fitted to data containing different sets of measurement moments to investigate when the mediation effect occurs.
Combination of measurement moments | BADSa | GDGRSb | |||||
|
a × bc | SE | P value | a × b | SE | P value | |
Combination 1 | |||||||
|
Pretest | 0.03 | 0.06 | .66 | 0.06 | 0.04 | .17 |
|
Week 1–posttest | 0.44 | 0.13 | .001d | 0.43 | 0.14 | .003d |
Combination 2 | |||||||
|
Pretest–week 1 | 0.10 | 0.10 | .30 | 0.20 | 0.12 | .09 |
|
Week 3–posttest | 0.59 | 0.16 | <.001d | 0.50 | 0.18 | .005d |
Combination 3 | |||||||
|
Pretest–week 3 | 0.20 | 0.11 | .08 | 0.19 | 0.10 | .04d |
|
Week 5–posttest | 0.69 | 0.20 | <.001d | 0.60 | 0.27 | .03d |
Combination 4 | |||||||
|
Pretest–week 5 | 0.23 | 0.11 | .04d | 0.22 | 0.10 | .02d |
|
Posttest | 0.52 | 0.15 | <.001d | 0.24 | 0.10 | .02d |
aBADS: Behavioral Activation for Depression Scale.
bGDGRS: Goal Disengagement and Goal Re-engagement Scale.
cCoefficient for the product term testing the mediation effect.
dP<.05.