12.4 Mediated moderation
12.4.1 Mediated moderation as the indirect effect of a product.
Hayes explains this in the next subsection, but we’ve already fit this model, which we called model3. Here’s the summary.
print(model3, digits = 3)## Family: MV(gaussian, gaussian)
## Links: mu = identity; sigma = identity
## mu = identity; sigma = identity
## Formula: justify ~ 1 + frame + skeptic + frame:skeptic
## donate ~ 1 + frame + justify + skeptic + frame:skeptic
## Data: disaster (Number of observations: 211)
## Samples: 4 chains, each with iter = 2000; warmup = 1000; thin = 1;
## total post-warmup samples = 4000
##
## Population-Level Effects:
## Estimate Est.Error l-95% CI u-95% CI Eff.Sample Rhat
## justify_Intercept 2.449 0.150 2.162 2.739 4000 1.000
## donate_Intercept 7.293 0.276 6.735 7.823 4000 1.000
## justify_frame -0.556 0.219 -0.977 -0.128 3612 1.000
## justify_skeptic 0.106 0.038 0.030 0.180 4000 1.000
## justify_frame:skeptic 0.199 0.056 0.092 0.305 3302 1.000
## donate_frame 0.157 0.273 -0.380 0.694 3437 1.000
## donate_justify -0.924 0.082 -1.089 -0.763 4000 1.000
## donate_skeptic -0.043 0.047 -0.135 0.050 4000 1.000
## donate_frame:skeptic 0.016 0.070 -0.121 0.152 2988 1.000
##
## Family Specific Parameters:
## Estimate Est.Error l-95% CI u-95% CI Eff.Sample Rhat
## sigma_justify 0.819 0.041 0.742 0.902 4000 1.000
## sigma_donate 0.989 0.049 0.897 1.091 4000 1.001
##
## Samples were drawn using sampling(NUTS). For each parameter, Eff.Sample
## is a crude measure of effective sample size, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).12.4.2 Why mediated moderation is neither interesting nor meaningful.
If it helps interpret this section, take a long look at the model formula.
model3$formula## justify ~ 1 + frame + skeptic + frame:skeptic
## donate ~ 1 + frame + justify + skeptic + frame:skeptic