# this script creates the dataset for the mediation exercise

library(dplyr)
library(faux)
library(lavaan)

# data setup
set.seed(123)
n <- 100
dat <-
  rnorm_multi(n = n,
              r = .5,
              varnames = c("x", "m", "y")) |> 
  mutate(y1 = rnorm_pre(y, r = .4),
         y2 = rnorm_pre(y, r = .5),
         y3 = rnorm_pre(y, r = .6),
         y4 = rnorm_pre(y, r = .7)) |> 
  select(-y)

cfa("y =~ y1 + y2 + y3 + y4", dat) |> summary(fit.measures = T)

syntax <- " 
y =~ y1 + y2 + y3 + y4
y ~ c*x
m ~ a*x
y ~ b*m
ab := a*b
total := c + (a*b)
"

sem(syntax, data = dat) |> summary(fit.measures = T)

dat <- 
  dat |> 
  mutate(m = ifelse((x > quantile(x, .5)) & (runif(n) > .5), NA, m))

sem(syntax, data = dat) |> summary(fit.measures = T)

# write.csv(dat, "exercise_03.csv", row.names = F)