If you have specific statistics you want to run in mind (or even just
want to redo some old analyses in R in order to make sure you
understand what you are doing) I wrote up the following quick and
dirty guide while I was teaching myself R. A few other folk have
found it useful when they had a specific "I want to do X" kind of
question. And if you have any questions/comments about it, I'd be
happy to chat off list.
http://homes.msi.ucsb.edu/~byrnes/rtutorial.html
-Jarrett
----------------------------------------
Jarrett Byrnes
Postdoctoral Associate, Santa Barbara Coastal LTER
Marine Science Institute
University of California Santa Barbara
Santa Barbara, CA 93106-6150
http://www.lifesci.ucsb.edu/eemb/labs/cardinale/people/byrnes/index.html
On Jun 3, 2009, at 1:16 AM, Gavin Simpson wrote:
On Tue, 2009-06-02 at 17:27 -0500, malcolm McCallum wrote:
I want to sit down and learn R.
Where is the best place to start?
R has a steep learning curve. In my experience in helping colleagues
to
start using R, unless you can set aside a good chunk of time to
learn by
yourself and you stick at it you'll end up getting frustrated that
something you know how to do in <insert software name> takes seconds
you
can't get R to do it. (Annoyingly, R will probably do it more quickly,
and with less effort, once you know the correct incantation.) For
those
of us with day jobs (I learned to use R when a grad student and my
supervisor had just move to another institution) setting aside this
time
may be difficult.
If you can get yourself onto a short course near to you, you'll
probably
find the initial slog up the learning curve easier.
If that is not possible, I'd recommend Peter Dalgaard's book
Introductory Statistics with R (2008, Springer, Second Edition) as an
excellent, low level intro to R, which covers a good range of common
statistical methods. Peter is one of the R Core Development team so he
knows what he is taking about.
You don't need to buy a book of course. There is a lot of contributed
documentation:
http://cran.r-project.org/other-docs.html
The first 3 on there are good introductory texts, depending on your
level or what you wish to learn. IIRC, all three have been
updated/improved into published books --- I have copies and they are
all
very good. More R books can be found here:
http://www.r-project.org/doc/bib/R-books.html
Don't forget the manual 'An Introduction to R' that comes with R or
you
can download from:
http://cran.r-project.org/manuals.html
The 'R Data Import/Export' manual also contains useful info on getting
data into and out of R.
There is also now a growing number of titles in Springers useR series.
These are meant to be low cost (!?) books that focus on a particular
topic and show how R is used to apply techniques common to the topic:
http://www.springer.com/series/6991
Someone mentioned Ben Bolker's book - it might sound appealing as it
has
the words "Ecological" in the title, but don't buy this without
looking
at it first. Don't get me wrong - I have a copy and think it is
great -
but unless you are a theoretical ecologist and/or are familiar with
writing out likelihood functions and optimising them given some data,
this may not be the book for you (yet). It certainly isn't a book to
learn how to use R with, and spends a lot of time on constructing
analysis methods for non-standard problems rather than say teaching
models such as linear, or mixed effects etc that might be fit with
standard R functions.
Once you've taken the plunge, sign yourself up for the R-SIG-Ecology
list, which is fairly low traffic. If you don't mind lots of email,
then
the R Help list is a goldmine of information and I learned a lot
just by
following the discussions and trying to understand the solutions to
posted questions. Just do your homework first and read the posting
guide! The mailing list details can be found here:
http://www.r-project.org/mail.html
To start orienting your way through the ~2000 add-on packages on CRAN,
take a look at the Task Views, which arrange information into subject
areas and list packages for common analyses encountered within each
subject area:
http://cran.r-project.org/web/views/
That should be enough to get you started ;-)
G
Malcolm
On Tue, Jun 2, 2009 at 4:58 PM, Gavin Simpson <gavin.simp...@ucl.ac.uk
> wrote:
On Tue, 2009-06-02 at 14:15 -0700, AdRiAnA HuMaNeS wrote:
Dear Listers:
I am writing to ask if anyone knows a statistical program besides
PERMANOVA=
that can do ANOVAS of mixed designs with four factors (two
orthogonals and=
two nested) and unbalanced data,
Best Regards
Adriana Humanes
Function adonis() in the vegan package for R can fit this type of
model.
You can find out more here:
http://cran.r-project.org/package=vegan
HTH,
G
--
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~
%~%
Dr. Gavin Simpson [t] +44 (0)20 7679 0522
ECRC, UCL Geography, [f] +44 (0)20 7679 0565
Pearson Building, [e] gavin.simpsonATNOSPAMucl.ac.uk
Gower Street, London [w] http://www.ucl.ac.uk/~ucfagls/
UK. WC1E 6BT. [w] http://www.freshwaters.org.uk
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~
%~%
--
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%
Dr. Gavin Simpson [t] +44 (0)20 7679 0522
ECRC, UCL Geography, [f] +44 (0)20 7679 0565
Pearson Building, [e] gavin.simpsonATNOSPAMucl.ac.uk
Gower Street, London [w] http://www.ucl.ac.uk/~ucfagls/
UK. WC1E 6BT. [w] http://www.freshwaters.org.uk
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%