Skip to content

Control variables in mediation analysis

4 messages · F.Doerwald, Ista Zahn, John Fox +1 more

#
Hi,

Just about anything can be bootstrapped, that's one of the things that
makes bootstrapping great! Use e.g., the boot package, and bootstrap
the product of the coefficients.

Alternatively the mediation package
(http://cran.r-project.org/web/packages/mediation/index.html) might
help, though I'm not sure if it uses bootstrap.

Best,
Ista
On Thu, Mar 21, 2013 at 3:21 PM, F.Doerwald <f.doerwald at student.rug.nl> wrote:
#
Dear Ista and F.Doerwald,

F.Doerwald could also fit a model using sem() in the sem package, and then use deltaMethod() in the car package to get the estimated value and standard error of any function of model coefficients that he or she chooses.

Sorry, I didn't read the original posting.

I hope this helps,
 John

------------------------------------------------
John Fox
Sen. William McMaster Prof. of Social Statistics
Department of Sociology
McMaster University
Hamilton, Ontario, Canada
http://socserv.mcmaster.ca/jfox/

On Thu, 21 Mar 2013 16:49:46 -0400
Ista Zahn <istazahn at gmail.com> wrote: