R help - Oct 2012 - Replace column values in R conditional on values from different column

Dear List,

I am working on a stats project and have been stumped by the issue of
replacing values in a column conditional on values from a different column.
I searched the forum and google in general, and was able to put some code
together, but it's not working the way it's supposed to... I appreciate
any
help you can offer! A detailed description of the issue, and the code I came
up with, is below. 

I am trying to separate one variable (CohenSum) into separate values for
males and females, in two separate columns, so I can do computations on them
(i.e. mean, sd). I tried the code below, but the values I got didn't make
sense - the length of the M and F columns should have been about half of the
initial column length and NA for the rest (I have about half men and half
women in the sample) but instead every cell has values in it, except for the
initial NAs. I think R somehow inserted 0 instead of NAs for the gender
values which were not supposed to be in that column (i.e. replaced values
for females with 0 in the male column) but I don't know why. 

Thanks!!!!!

for(i in 1:length(Health2$CohenSum)){
  if (Health2$q_5a[i]==1) Health2$CohenM[i] <- Health2$CohenSum[i]
  else if (Health2$q_5a[i]==2) Health2$CohenF[i] <- Health2$CohenSum[i]
  else Health2$CohenF[i] <- NA
  }



--
View this message in context:
http://r.789695.n4.nabble.com/Replace-column-values-in-R-conditional-on-values-from-different-column-tp4646087.html
Sent from the R help mailing list archive at Nabble.com.

Hello,

No need for a loop. "else if" doesn't exist in R. And
Health2$CohenM and
Health2_CohenF have to be created first, in case of a loop

If Health2$q_5a only have 1 and 2 as possible values, just use "which"
and "-". And I would suggest to not add new variables in your data 
frame, because you will include some NA. Create 2 new variables, then do 
your different calculations on.

CohenM <- Health2$CohenSum[w]
CohenF <- Health2$CohenSum[-w]

Regards
Pascal


Le 12/10/13 20:59, GradStudentDD a ?crit :
> Dear List,
>
> I am working on a stats project and have been stumped by the issue of
> replacing values in a column conditional on values from a different column.
> I searched the forum and google in general, and was able to put some code
> together, but it's not working the way it's supposed to... I
appreciate any
> help you can offer! A detailed description of the issue, and the code I
came
> up with, is below.
>
> I am trying to separate one variable (CohenSum) into separate values for
> males and females, in two separate columns, so I can do computations on
them
> (i.e. mean, sd). I tried the code below, but the values I got didn't
make
> sense - the length of the M and F columns should have been about half of
the
> initial column length and NA for the rest (I have about half men and half
> women in the sample) but instead every cell has values in it, except for
the
> initial NAs. I think R somehow inserted 0 instead of NAs for the gender
> values which were not supposed to be in that column (i.e. replaced values
> for females with 0 in the male column) but I don't know why.
>
> Thanks!!!!!
>
> for(i in 1:length(Health2$CohenSum)){
>    if (Health2$q_5a[i]==1) Health2$CohenM[i] <- Health2$CohenSum[i]
>    else if (Health2$q_5a[i]==2) Health2$CohenF[i] <- Health2$CohenSum[i]
>    else Health2$CohenF[i] <- NA
>    }
>
>
>
> --
> View this message in context:
http://r.789695.n4.nabble.com/Replace-column-values-in-R-conditional-on-values-from-different-column-tp4646087.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.
>

HI,

Not sure how your dataset looks like and not very clear whether this is what you
want.

Try this:
set.seed(1)
Health1<-data.frame(q_5=sample(1:2,10,replace=TRUE),CohenSum=rnorm(10,25))
Health2<-data.frame(Health1,CohenM=NA,CohenF=NA)
idxM<-which(Health2$q_5==1)
?idxF<-which(Health2$q_5==2)
Health2$CohenM[idxM]<-Health2$CohenSum[Health2$q_5==1]
?Health2$CohenF[idxF]<-Health2$CohenSum[Health2$q_5==2]
?Health2
#?? q_5 CohenSum?? CohenM?? CohenF
#1??? 1 24.17953 24.17953?????? NA
#2??? 1 25.48743 25.48743?????? NA
#3??? 2 25.73832?????? NA 25.73832
#4??? 2 25.57578?????? NA 25.57578
#5??? 1 24.69461 24.69461?????? NA
#6??? 2 26.51178?????? NA 26.51178
#7??? 2 25.38984?????? NA 25.38984
#8??? 2 24.37876?????? NA 24.37876
#9??? 2 22.78530?????? NA 22.78530
#10?? 1 26.12493 26.12493?????? NA


A.K.



----- Original Message -----
From: GradStudentDD <dd7kc at virginia.edu>
To: r-help at r-project.org
Cc: 
Sent: Saturday, October 13, 2012 2:59 PM
Subject: [R] Replace column values in R conditional on values from different
column

Dear List,

I am working on a stats project and have been stumped by the issue of
replacing values in a column conditional on values from a different column.
I searched the forum and google in general, and was able to put some code
together, but it's not working the way it's supposed to... I appreciate
any
help you can offer! A detailed description of the issue, and the code I came
up with, is below. 

I am trying to separate one variable (CohenSum) into separate values for
males and females, in two separate columns, so I can do computations on them
(i.e. mean, sd). I tried the code below, but the values I got didn't make
sense - the length of the M and F columns should have been about half of the
initial column length and NA for the rest (I have about half men and half
women in the sample) but instead every cell has values in it, except for the
initial NAs. I think R somehow inserted 0 instead of NAs for the gender
values which were not supposed to be in that column (i.e. replaced values
for females with 0 in the male column) but I don't know why. 

Thanks!!!!!

for(i in 1:length(Health2$CohenSum)){
? if (Health2$q_5a[i]==1) Health2$CohenM[i] <- Health2$CohenSum[i]
? else if (Health2$q_5a[i]==2) Health2$CohenF[i] <- Health2$CohenSum[i]
? else Health2$CohenF[i] <- NA
? }



--
View this message in context:
http://r.789695.n4.nabble.com/Replace-column-values-in-R-conditional-on-values-from-different-column-tp4646087.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.

Thank you so much!!! This is exactly what I needed!! Have a wonderful rest of
the weekend!!



--
View this message in context:
http://r.789695.n4.nabble.com/Replace-column-values-in-R-conditional-on-values-from-different-column-tp4646087p4646091.html
Sent from the R help mailing list archive at Nabble.com.