R help - Feb 2003 - Periods instead of spaces in dataframe names?

Basic question:
when I use names() to extract the name of a dataframe element, why does it
have "." instead of " " between words?

Context:
I'm importing a CSV file of survey results for analysis. I read them like
this:

df <- read.csv("surveydata.csv",nrows=40,header=TRUE,
              
na.string=c("N/A",""),comment.char="",strip.white=TRUE)

To do a summary of the responses to a question, I can now do something like
this:

table(df[13])

 1  2  3  4
13 13  4  2

and can then do a barplot with:
barplot(table(df[13])))

Which is fine. But since the first row of my data file includes the questions
themselves, I want to use those questions as chart titles, like this:

barplot(table(df[13]),main=names(df[13]))

This gives an unexpected result: the title has a "." where every space
should
be between words, like this: "How.many.cats.do.you.own"

I can't figure out why, or how to get spaces instead of "." to
show up. I've
tried using gsub but without success, to substitute " " for
"."

I'm think I'm confused about something fundamental here, and hope
someone has
the patience to enlighten me. I confess to a being a bit vague in my
understanding of R's handling of arrays, dataframes, and vectors. My
background is in C programming and I keep looking for a "string"...

Thank you very much. Please reply or copy me directly if you respond.

Tom Arnold
Managing Partner, Summit Media Partners LLC
Visit our web site at http://www.summitmediapartners.com

Tom,

Tom Arnold wrote:
> Basic question:
> when I use names() to extract the name of a dataframe element, why does it
> have "." instead of " " between words?
> 
> Context:
> I'm importing a CSV file of survey results for analysis. I read them
like
> this:
> 
> df <- read.csv("surveydata.csv",nrows=40,header=TRUE,
>               
na.string=c("N/A",""),comment.char="",strip.white=TRUE)
> 
> To do a summary of the responses to a question, I can now do something like
> this:
> 
> table(df[13])
> 
>  1  2  3  4
> 13 13  4  2
> 
> and can then do a barplot with:
> barplot(table(df[13])))
> 
> Which is fine. But since the first row of my data file includes the
questions
> themselves, I want to use those questions as chart titles, like this:
> 
> barplot(table(df[13]),main=names(df[13]))
> 
> This gives an unexpected result: the title has a "." where every
space should
> be between words, like this: "How.many.cats.do.you.own"
> 
> I can't figure out why, or how to get spaces instead of "."
to show up. I've
> tried using gsub but without success, to substitute " " for
"."
> 
> I'm think I'm confused about something fundamental here, and hope
someone has
> the patience to enlighten me. I confess to a being a bit vague in my
> understanding of R's handling of arrays, dataframes, and vectors. My
> background is in C programming and I keep looking for a
"string"...
> 
> Thank you very much. Please reply or copy me directly if you respond.
You can use check.names = FALSE in your read.csv(...) call. Or you can 
use gsub as in:

names(df2) = gsub("\\.", " ", names(df2))

The "."  must be escaped in gsub if used in the pattern argument.

Regards,
Sundar

Dear Tom,

When you read the data into a data frame via read.csv, the character 
strings in the first row of the data file, which you've indicated is to be 
interpreted as a header, are used for column names; in the process, blanks 
are converted to periods, since nonstandard names including blanks are more 
difficult to deal with; names(df[13]) just returns the name of column 13 in 
the data frame. You could use the gsub function to recover the blanks -- 
something like gsub("\\."," ", names(df[2])). Alternatively,
you could
specify the argument check.names=FALSE to read.csv to avoid substituting 
periods for blanks in the first place, but this probably isn't a good idea. 
See ?read.csv for details.

I hope that this helps,
  John


At 09:33 AM 2/11/2003 -0700, Tom Arnold wrote:
>Basic question:
>when I use names() to extract the name of a dataframe element, why does it
>have "." instead of " " between words?
>
>Context:
>I'm importing a CSV file of survey results for analysis. I read them
like
>this:
>
>df <- read.csv("surveydata.csv",nrows=40,header=TRUE,
>               
na.string=c("N/A",""),comment.char="",strip.white=TRUE)
>
>To do a summary of the responses to a question, I can now do something like
>this:
>
>table(df[13])
>
>  1  2  3  4
>13 13  4  2
>
>and can then do a barplot with:
>barplot(table(df[13])))
>
>Which is fine. But since the first row of my data file includes the
questions
>themselves, I want to use those questions as chart titles, like this:
>
>barplot(table(df[13]),main=names(df[13]))
>
>This gives an unexpected result: the title has a "." where every
space should
>be between words, like this: "How.many.cats.do.you.own"
>
>I can't figure out why, or how to get spaces instead of "." to
show up. I've
>tried using gsub but without success, to substitute " " for
"."
>
>I'm think I'm confused about something fundamental here, and hope
someone has
>the patience to enlighten me. I confess to a being a bit vague in my
>understanding of R's handling of arrays, dataframes, and vectors. My
>background is in C programming and I keep looking for a
"string"...
>
>Thank you very much. Please reply or copy me directly if you respond.
>
>Tom Arnold
>Managing Partner, Summit Media Partners LLC
>Visit our web site at http://www.summitmediapartners.com
>
>______________________________________________
>R-help at stat.math.ethz.ch mailing list
>http://www.stat.math.ethz.ch/mailman/listinfo/r-help
-----------------------------------------------------
John Fox
Department of Sociology
McMaster University
Hamilton, Ontario, Canada L8S 4M4
email: jfox at mcmaster.ca
phone: 905-525-9140x23604
web: www.socsci.mcmaster.ca/jfox
-----------------------------------------------------