I am conducting a mediation analyses in R with incomplete data for my master's thesis. For the missings I use MICE multiple imputation according to van Buuren & Groothuis-Oudshoorn (2011).
Tingley et al. (2014; see: https://oar.princeton.edu/jspui/bitstream/88435/pr1gj2f/1/Mediation%20JSS.pdf) wrote about mediation and missing data:
"The mediation package includes a pair of utility functions – mediations and amelidiate – to facilitate such analysis. First, users simulate multiple data sets using their preferred imputation software. Next, run mediate on each data set by simply passing the data sets through mediations. Next, pass the output of mediations to the amelidiate function, which combines the components of the output from mediations into a format that can be analyzed with the standard summary and plot commands."
Unfortunately I was not able to write a correct R code. I have the following three variables:
Predictor: number of completed modules (ModuleEr)
Mediator: therapeutic alliance with eCoach (WAI_P)
Outcome: Severity of depression (QIDS_t1)
Suppose I do a simple imputation, how do I proceed with the mediation?
I was thinking of something like that:
df <- data
imp <- mice(df, m=5, seed = 1234)
mod1 <- with(data = imp, exp = lm(WAI_P ~ ModuleEr))
res1 <- pool(mod1)
mod2 <- with(data = imp, exp = lm(QIDS_t1 ~ ModuleEr + WAI_P))
res2 <- pool(mod2)
For mediation I was thinking of:
a) Med <- mediate(mod1, mod2, treat="QIDS_t1“, mediator="WAI_P", sims = 100)
Or b ). Med <- with(data = imp, exp = mediate(mod1, mod2, treat="QIDS_t1“, mediator="WAI_P", sims = 100))
But both, a) and b) showed following warning:
Error in formula.default(object, env = baseenv()) : invalid formula
I would be very happy if someone could share an R code with me or correct mine!
Tanks a lot!