mediation2: Serial Mediation with Two Mediators
In leettran/MeMoBootR: Mediation-Moderation with Bootstrapping in R

Description Usage Arguments Examples

View source: R/mediation2.R

This function runs a complete serial mediation analysis with two mediators, similiar to model 6 in PROCESS by A. Hayes (2013). As part of the output, you will find data screening, all three models used in the traditional Baron and Kenny (1986) steps, total/direct/indirect effects, the z-score and p-value for the Aroian Sobel test, and the bootstrapped confidence interval for the indirect effect.

mediation2(
  y,
  x,
  m1,
  m2,
  cvs = NULL,
  df,
  with_out = T,
  nboot = 1000,
  conf_level = 0.95
)

`y`	The dependent variable column name from your dataframe.
`x`	The independent variable column name from your dataframe. This column will be treated as X in mediation or moderation models, please see diagrams online for examples.
`m1`	The first mediator for your model.
`m2`	The second mediator for your model.
`cvs`	The covariates you would like to include in the model. Use a `c()` concatenated vector to use multiple covariates.
`df`	The dataframe where the columns from the formula can be found. Note that only the columns used in the analysis will be data screened.
`with_out`	A logical value where you want to keep the outliers in model `TRUE` or exclude them from the model `FALSE`.
`nboot`	A numeric value indicating the number of bootstraps you would like to complete.
`conf_level`	A numeric value indicating the confidence interval width for the boostrapped confidence interval.

1
2
3

mediation2(y = "Q11", x = "Q151", m1 = "Q31", m2 = "Q41",
          cvs = c("Q121"), df = mediation2_data, nboot = 1000, with_out = T,
          conf_level = .95)