Dear Weichao-<br><br>Our apologies for a delay in answering your question.<br><br>You can absolutely use continuous treatments. In the latest release of the package, version 3.0 (to automatically load this library use install.packages("mediation")
using R 2.13, the latest version of R, or higher in the future).<br><br>In 3.0 this is done through treat.value and control.value arguments. More details are available here:<br><a href="http://cran.r-project.org/web/packages/mediation/vignettes/mediation.pdf">http://cran.r-project.org/web/packages/mediation/vignettes/mediation.pdf</a><br>
<br>We've setup a mailing list that you should join so you can stay abreast of changes and updates to the package. This is available here<br><a href="https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/mediation-information">https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/mediation-information</a><br>
<br>Note that sensitivity analyses are only worked out for binary 0/1 treatments.<br><br>best,<br>Dustin<br><br clear="all">Dustin Tingley<br>Government Department<br>Harvard University<br><a href="http://scholar.harvard.edu/dtingley" target="_blank">http://scholar.harvard.edu/dtingley</a><br>
<br>
<br><br><div class="gmail_quote">On Mon, Jul 4, 2011 at 4:39 PM, Kosuke Imai <span dir="ltr"><<a href="mailto:kimai@princeton.edu">kimai@princeton.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
What's the answer to this FAQ? Maybe, one of you can answer the question and cc to the mediation email list?<br>
<br>
Thanks,<br>
Kosuke<br>
<br>
Begin forwarded message:<br>
<br>
> From: "BAO, WEICHAO" <<a href="mailto:baow@email.sc.edu">baow@email.sc.edu</a>><br>
> Date: July 5, 2011 2:32:01 AM GMT+09:00<br>
> To: "<a href="mailto:kimai@princeton.edu">kimai@princeton.edu</a>" <kimai@Princeton.EDU><br>
> Subject: mediation R Package<br>
><br>
> Dear Dr. Imai,<br>
><br>
> I am trying to use your R package to do a mediation analyis. I have a question about the variable treatment(i.e. exposure ). I know the variable treatment in your example is discrete. Can the package still be used if the treatment variable is continuous (such as the exposure of some metals)?<br>
><br>
><br>
> Thanks you very much.<br>
><br>
> Sincerely,<br>
><br>
> Weichao<br>
><br>
><br>
> Weichao Bao<br>
> Aronal School of Public Health<br>
> University of South Carolina<br>
> <a href="tel:803-445-6087" value="+18034456087">803-445-6087</a><br>
<br>
</blockquote></div><br>