R help - Sep 2012 - Recoding categorical gender variable into numeric factors

I currently have a data set in which gender is inputed as "Male" and
"Female"
, and I'm trying to convert this into "1" and "0".

I found a website which reccomended using two commands:

data$scode[data$sex=="M"] <- "1"
data$scode[data$sex=="F"] <- "2"

to convert to numbers, and:

data$scode <- factor(data$scode)

to convert this variable to a factor.



My issue is that, after I use the first command, *only* the female values
get converted to a number. I am left with a column filled with 2's and blank
spaces. Instead of typing both lines of the first command, I copy and pasted
the first line and changed the letter representing gender. I also made sure
that both letters were exactly as they appear in the dataset.

My questions are: is there any visible issue with my syntax, and are there
any other methods to accomplish this?

I'm also very new to R, so complex syntax is beyond me.

Conrad Baldner



--
View this message in context:
http://r.789695.n4.nabble.com/Recoding-categorical-gender-variable-into-numeric-factors-tp4642316.html
Sent from the R help mailing list archive at Nabble.com.

Hi Conrad,

On Wed, Sep 5, 2012 at 3:14 PM, Conradsb <csbaldne at vt.edu>
wrote:
> I currently have a data set in which gender is inputed as "Male"
and "Female"
> , and I'm trying to convert this into "1" and "0".
This is usually not necessary, and makes things more confusing. "Male"
and "Female" is clear and self-explanatory: "0" and
"1" are not.
>
> I found a website which reccomended using two commands:
>
> data$scode[data$sex=="M"] <- "1"
> data$scode[data$sex=="F"] <- "2"
Nope, "1" is the character 1, not the number 1 in R. Also, you said
the values were "Male" and "Female", not "F" and
"M". To convert
"Male" to 1 and "Female" to 2 you can use

data$scode[data$sex=="Male"] <- 1
> data$scode[data$sex=="Female"] <- 2
Notice "Male" and "Female", instead of "M" and
"F", and 1 and 2
instead of "1" and "2"

>
> to convert to numbers, and:
>
> data$scode <- factor(data$scode)
>
> to convert this variable to a factor.
No need to convert it to a factor first. Just use

data$sex <- factor(data$sex)
>
>
>
> My issue is that, after I use the first command, *only* the female values
> get converted to a number. I am left with a column filled with 2's and
blank
> spaces.
Strange, especially if sex is actually "Male" and "Female",
in which
case scode should be all NA. If you want to follow up on this, please
post the result of

dput(dat["sex"])

Instead of typing both lines of the first command, I copy and
pasted
> the first line and changed the letter representing gender. I also made sure
> that both letters were exactly as they appear in the dataset.
>
> My questions are: is there any visible issue with my syntax, and are there
> any other methods to accomplish this?
In this case you don't actually need to convert to numeric. Just use

data$scode <- factor(scode)

If you really need to convert characters to numbers, it is often
convenient to use factors as intermediate steps, like this:

dat <- data.frame(sex=sample(c("Male", "Female"), 10,
replace=TRUE))

dat$sex.n <- as.numeric(
  as.character(
    factor(
      dat$sex,
      levels = c("Female", "Male"),
      labels = c("0", "1"))))

Best,
Ista
>
> I'm also very new to R, so complex syntax is beyond me.
>
> Conrad Baldner
>
>
>
> --
> View this message in context:
http://r.789695.n4.nabble.com/Recoding-categorical-gender-variable-into-numeric-factors-tp4642316.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

I can't replicate your problem. I created a data set with "Male"
and
"Female" since that is what you indicate, but your commands use
"M" and "F"
which is different. When I use "Male" and "Female" the
recoding is just as
expected, but you don't even need to do this. You probably already have a
factor since R routinely turns character fields into factors:
> data <- data.frame(sex=c(rep("Male", 5),
rep("Female", 5)))
> data
      sex
1    Male
2    Male
3    Male
4    Male
5    Male
6  Female
7  Female
8  Female
9  Female
10 Female
> str(data)
'data.frame':   10 obs. of  1 variable:
 $ sex: Factor w/ 2 levels "Female","Male": 2 2 2 2 2 1 1 1
1 1

So data$sex is a Factor with two levels Female=1 and Male=2. If the result
of str(data) looks like this, you have a character array (chr):
> str(data)
'data.frame':   10 obs. of  1 variable:
 $ sex: chr  "Male" "Male" "Male" "Male"
...

If you want to convert a character array to a factor just use the command:

data$sex <- factor(data$sex)

By default, R orders the character strings alphabetically before converting
to factors so "Female" becomes 1 and "Male" becomes 2.

----------------------------------------------
David L Carlson
Associate Professor of Anthropology
Texas A&M University
College Station, TX 77843-4352
> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> project.org] On Behalf Of Conradsb
> Sent: Wednesday, September 05, 2012 2:14 PM
> To: r-help at r-project.org
> Subject: [R] Recoding categorical gender variable into numeric factors
> 
> I currently have a data set in which gender is inputed as "Male"
and
> "Female"
> , and I'm trying to convert this into "1" and "0".
> 
> I found a website which reccomended using two commands:
> 
> data$scode[data$sex=="M"] <- "1"
> data$scode[data$sex=="F"] <- "2"
> 
> to convert to numbers, and:
> 
> data$scode <- factor(data$scode)
> 
> to convert this variable to a factor.
> 
> 
> 
> My issue is that, after I use the first command, *only* the female
> values
> get converted to a number. I am left with a column filled with 2's and
> blank
> spaces. Instead of typing both lines of the first command, I copy and
> pasted
> the first line and changed the letter representing gender. I also made
> sure
> that both letters were exactly as they appear in the dataset.
> 
> My questions are: is there any visible issue with my syntax, and are
> there
> any other methods to accomplish this?
> 
> I'm also very new to R, so complex syntax is beyond me.
> 
> Conrad Baldner
> 
> 
> 
> --
> View this message in context: http://r.789695.n4.nabble.com/Recoding-
> categorical-gender-variable-into-numeric-factors-tp4642316.html
> Sent from the R help mailing list archive at Nabble.com.
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-
> guide.html
> and provide commented, minimal, self-contained, reproducible code.