Dear Jan,
It depends what you are looking to do. For the studies that have
given you means and standard deviations, if you want a d effect size,
you should already be set (Mean1 - Mean2)/control standard deviation
or the pooled standard deviation if you prefer. For the studies where
you just have p-values, assuming you also have the sample size:
qt(p = .05/2, df = 42)
will give you the t for a two tailed p-value of .05 on 42 degrees of
freedom, and t is basically just d * (sqrt(df)/2), so it is not
difficult to get a d effect size once you have the t value.
Of course all that makes quite a few assumptions about what data you
have and what you are trying to do. For tools to help you in R, take
look at the rmeta package
http://crantastic.org/packages/rmeta
and I also recommend looking around crantastic for other packages that
might work for you. There are handy tags like "similar packages" that
are a nice way to browse around, and when you're done with your
meta-analysis, check back there to add ratings/a review of the package
you used.
Cheers,
Josh
On Fri, Oct 8, 2010 at 8:14 AM, kende jan <kendejan at yahoo.fr>
wrote:> Dear all,
>
> I am trying to do meta-analysis of continuous outcome data. Twelve studies
are
> selected but for six of them, i have only p-values and the six other means
and
> standard deviation for the two groups (Experimental and Control). How can I
do
> with R to take into account p-values and/or means and standard deviation to
> perform my meta-analysis.
>
> Thanks for your help
> Jan
>
>
>
> ? ? ? ?[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
--
Joshua Wiley
Ph.D. Student, Health Psychology
University of California, Los Angeles
http://www.joshuawiley.com/