[R-sig-Geo] ANOVA (two way) and Friedman Test

[Eric Boeker]

Dear list members,


I have a set of soil samples tested for one Element taken at 3 different
depths (respectively at depth A, B and C) on a transect.
I want to determine whether the results for the different layers (depths)
are comparable.
>From the literature I found that 2 tests can to be employed:

1) An extension of ANOVA that accounts for a blocking effect (different
samples taken at same position) = ANOVA (two way)
2) Non parametric equivalent of ANOVA with blocking the Friedman test

In both case, I got stuck. In the case of Friedman (
http://stat.ethz.ch/R-manual/R-patched/library/stats/html/friedman.test.html)
the example is given with a matrix and so far I was only dealing with
data.frame. I didnt figure out how to transform my data into a matrix or if
the data frame can be used directly.
With the two way ANOVA
(http://cran.r-project.org/web/packages/car/car.pdf(p.11)), I did not
managed to figure out which argument to use for the
linear model (lm).

If someone is interested to give me a hand find attached the surveys data
at different depth.


-- 
Mit freundlichen Grüßen / Kind regards / Meilleures Salutations


Eric Böker

Project Manager
Eiseb Prospecting and Mining (Pty)Ltd
Gobabis, Namibia
<eric at eiseb.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://stat.ethz.ch/pipermail/r-sig-geo/attachments/20111118/bc666683/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: soil.csv
Type: text/csv
Size: 2268 bytes
Desc: not available
URL: <https://stat.ethz.ch/pipermail/r-sig-geo/attachments/20111118/bc666683/attachment.bin>