R help - Oct 2013 - Comparing two groups

Hi,
this might sound trivial, but I'm pretty new to R and statistics in general.
What I want to do is to compare two data values. The hook is, that they are
non-normally distributed and that one value is five times as big as the
other. The box-plots look like this 
<http://r.789695.n4.nabble.com/file/n4678190/mixture_vs_monoculture.png> 
.
Ignore the number at the bottom. I want to know if those two are
significantly different from one another. I tried it with the wilcox-test
(because it is advertised as a non-parametric test), but get a p-value of
0.0009 and naturally don't quite believe it to be true.
Do you have any suggestions how I can handle that problem?

Andrej



--
View this message in context:
http://r.789695.n4.nabble.com/Comparing-two-groups-tp4678190.html
Sent from the R help mailing list archive at Nabble.com.

Hi
> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> project.org] On Behalf Of Andrej
> Sent: Monday, October 14, 2013 9:49 AM
> To: r-help at r-project.org
> Subject: [R] Comparing two groups
> 
> Hi,
> this might sound trivial, but I'm pretty new to R and statistics in
> general.
So why not start with some statistical textbook? There are plenty of them
available in CRAN.
> What I want to do is to compare two data values. The hook is, that they
> are non-normally distributed and that one value is five times as big as
> the other. The box-plots look like this
> <http://r.789695.n4.nabble.com/file/n4678190/mixture_vs_monoculture.png
> >  .
> Ignore the number at the bottom. I want to know if those two are
> significantly different from one another. I tried it with the wilcox-
> test (because it is advertised as a non-parametric test), but get a p-
> value of
> 0.0009 and naturally don't quite believe it to be true.
Why? What leads you to this statement? Some other tests? Some other results?
> Do you have any suggestions how I can handle that problem?
You can try some normalising procedure like Box-Cox.

function (x, lambda, inv = F) 
{
    if (!inv) {
        if (missing(lambda)) 
            log(x)
        else (x^lambda - 1)/lambda
    }
    else (lambda * x + 1)^(1/lambda)
  }

or boxcox from MASS package.

and then to use standard t.test, but you will probably gat quite similar result
as with wilcox.test.

Regards
Petr
> 
> Andrej
> 
> 
> 
> --
> View this message in context: http://r.789695.n4.nabble.com/Comparing-
> two-groups-tp4678190.html
> Sent from the R help mailing list archive at Nabble.com.
> 
> ______________________________________________
> 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.