[R] Re: [ -R] Summary converting among modes

Martin Henry H. Stevens hstevens at muohio.edu
Fri Jul 27 19:58:30 CEST 2001

Thank you very much for the response.
The way to get rid of those pesky characters is to first convert to
character (from the default factor), THEN convert to numeric. I had tried
using as.is= in read.table, but I used it incorrectly (as.is=TRUE instead of
as.is= <vector of col indices>).

as.numeric( as.character( vect.dat ) )
NOT simply    as.numeric(vect.dat)

I used as.is= incorrectly. I simply used as.is=TRUE, and did not specify
----- Original Message -----
From: "Thomas Lumley" <tlumley at u.washington.edu>
To: "Martin Henry H. Stevens" <hstevens at muohio.edu>
Cc: <r-help at stat.math.ethz.ch>
Sent: Friday, July 27, 2001 1:47 PM
Subject: Re: [R] converting among modes

> On Fri, 27 Jul 2001, Martin Henry H. Stevens wrote:
> > Windows 2000, R 1.3.0
> >
> > I have been given a data set (ASCII tab delimited) in which many
> > that are supposed to be numeric actually contain characters. I would be
> > happy to have those character entries disappear and become NA.
> >
> > When I use read.table and then attempt to convert these using
> > of course it doesn't work, and I get the ASCII? representation? maybe?
> > I use scan() and specify the data type (what=...), I get an error
> > saying that I have character data in the numeric variable.
> If there are only a few different character strings causing problems you
> can use the na.strings option of read.table().
> Your data are actually read as *factors*, not as characters, and
> as.numeric() gives the factor codes.  Converting factors to numeric is a
> FAQ (7.13); use as.numeric(as.character(the.factor))
> -thomas
> Thomas Lumley Asst. Professor, Biostatistics
> tlumley at u.washington.edu University of Washington, Seattle

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