R help - Oct 2011 - control the conversion of factor to numeric

Dear R-list,



I currently have to convert a data.frame with several factor-variables to a
numeric matrix.

Now the problem is, that the order of the factor-labels don't match the
order I would like to use.


for example, let's assume I have this factor-variable in my data-frame:

x <- factor(rep(1:4, 5), labels=c("slightly disagree",
"disagree", "agree", "slightly agree"))


Now I would like to convert this to a numeric vector so that disagree == 1,
slightly disagree == 2, slightly agree == 3, and agree gets the value 4.
but as.numeric(x) just converts the factor levels to numerical values according
to their label-order.


Is there a convenient, flexible function that let's you control how the
factor-levels get converted to numerical values?



thanks for any suggestions!

Add levels= to your factor() call.

E.g.,

x1 <- factor(rep(1:4, 5), labels=c("slightly disagree",
"disagree",
"agree", "slightly agree"), levels = c(2,1,4,3))
as.numeric(x1)
 [1] 2 1 4 3 2 1 4 3 2 1 4 3 2 1 4 3 2 1 4 3

Michael

On Tue, Oct 18, 2011 at 7:35 AM, Martin Batholdy
<batholdy at googlemail.com> wrote:
> Dear R-list,
>
>
>
> I currently have to convert a data.frame with several factor-variables to a
numeric matrix.
>
> Now the problem is, that the order of the factor-labels don't match the
order I would like to use.
>
>
> for example, let's assume I have this factor-variable in my data-frame:
>
> x <- factor(rep(1:4, 5), labels=c("slightly disagree",
"disagree", "agree", "slightly agree"))
>
>
> Now I would like to convert this to a numeric vector so that disagree == 1,
slightly disagree == 2, slightly agree == 3, and agree gets the value 4.
> but as.numeric(x) just converts the factor levels to numerical values
according to their label-order.
>
>
> Is there a convenient, flexible function that let's you control how the
factor-levels get converted to numerical values?
>
>
>
> thanks for any suggestions!
>
> ______________________________________________
> 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.
>

On Oct 18, 2011, at 7:35 AM, Martin Batholdy wrote:
> Dear R-list,
>
>
>
> I currently have to convert a data.frame with several factor- 
> variables to a numeric matrix.
>
> Now the problem is, that the order of the factor-labels don't match  
> the order I would like to use.
>
>
> for example, let's assume I have this factor-variable in my data- 
> frame:
>
> x <- factor(rep(1:4, 5), labels=c("slightly disagree",
"disagree",
> "agree", "slightly agree"))
>
>
> Now I would like to convert this to a numeric vector so that  
> disagree == 1, slightly disagree == 2, slightly agree == 3, and  
> agree gets the value 4.
> but as.numeric(x) just converts the factor levels to numerical  
> values according to their label-order.
>
>
> Is there a convenient, flexible function that let's you control how  
> the factor-levels get converted to numerical values?
I think you want to first change the ordering of the levels and then  
convert as desired.

 > levels(x) <- c("slightly disagree", "disagree",
"slightly agree",
"agree")
 > x
  [1] slightly disagree disagree          slightly agree    agree
  [5] slightly disagree disagree          slightly agree    agree
  [9] slightly disagree disagree          slightly agree    agree
[13] slightly disagree disagree          slightly agree    agree
[17] slightly disagree disagree          slightly agree    agree
Levels: slightly disagree disagree slightly agree agree

#  Now choose hoe to convert

 > as.numeric(x)
  [1] 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
 > as.character(x)
  [1] "slightly disagree" "disagree"          "slightly
agree"
"agree"
  [5] "slightly disagree" "disagree"          "slightly
agree"
"agree"
  [9] "slightly disagree" "disagree"          "slightly
agree"
"agree"
[13] "slightly disagree" "disagree"          "slightly
agree"    "agree"
[17] "slightly disagree" "disagree"          "slightly
agree"    "agree"
>
>
>
> thanks for any suggestions!
>
> ______________________________________________
> 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.
David Winsemius, MD
West Hartford, CT