[R] comments on documentation

John Hendrickx john_hendrickx at yahoo.com
Tue Feb 27 10:37:24 CET 2001

I've been experimenting with R during the last week and I thought I'd
pass on some comments on issues on the documentation that left me a
bit confused. To start with (literally), the ReadMe.rw1021 file says
that the easiest way to install R is to download the rwinst.exe
installer program. That gave me the impression that the program would
connect to an R download site, download the software and install it,
along the lines of the Internet Explorer installation program for
example. The Readme should be clearer on this and instruct newbies to
download the zipfiles to a folder first. I also found that the
installer didn't work if the files were downloaded to c:\, it worked
fine if everything was downloaded to a folder.

I also had a good deal of trouble trying to import an SPSS file. I
found and installed the foreign package without too much trouble but
didn't realize I had to load it before I could use it :-(. Reading
The Fine Manual helped eventually, but it would be useful to provide
a bit more information in the "R Import/Export" documentation. 

After reading an SPSS file, what data structure is it? It's not a
data frame. In any event, I found it useful to convert the SPSS data
to a data frame since that also handles invalid names, e.g.
containing underscores. You can analyze the SPSS data as read in, but
any names containing underscores cause an error. I also found that
the SPSS names are all upper case. Maybe that could be changed in new
versions of read.spss. A workaround is:

indat <- read.spss("c:/data/ozs/eind.sav")
dat <- as.data.frame(indat)
names(dat) <- tolower(names(dat))

An example along these lines would help newbies a lot, I think. I
also disagree with the import/export documentation, that importing
files from foreign packages should be avoided (chapter 3). This will
often be the starting point for new users who have most of their data
in a foreign format and will usually be the quickest and most
practical method in any case.

John Hendrickx

Do You Yahoo!?
Get email at your own domain with Yahoo! Mail. 
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch

More information about the R-help mailing list