CODEWITHSUNDEEP
recode
swhusky
swhusky

Reputation: 305

Recoding non-mutually exclusive variables into mutually exclusive variables

Any help with this would be much appreciated.

I have four overlapping race variables that I would like to make mutually exclusive and code the remainder into a new variable mixed race variable. I've been trying to use if and ifelse and failing miserably.

white<-c(1,1,1,NA)
black<-c(0,NA,1,0)
asian<-c(0,0,0,0)
aian<- c(0,0,0,0)       
white.n<-c(1,1,0,NA)
mix<-c(0,0,1,0)

df<-cbind(white,black,asian,aian,white.n,mix)
df

      white black asian aian white.n mix
[1,]     1     0     0    0       1   0
[2,]     1    NA     0    0       1   0
[3,]     1     1     0    0       0   1
[4,]    NA     0     0    0      NA   0

Any thoughts would be much appreciated.

Upvotes: 2

Views: 182

Answers (1)

Runic
Runic

Reputation: 149

You want a new column that codes everything into a mutually exclusive variable, right? Like this?

white<-c(1,1,1,NA,0)
black<-c(0,NA,1,0,0)
asian<-c(0,0,0,0,1)
white<-c(1,1,0,NA,0)

df<-data.frame(white,black,asian,white)
df

row.count <- rowSums(df, na.rm=T)
df$code[row.count > 1] <- "Mixed"
df$code[row.count == 1] <- names(which.max(df[row.count == 1, ]))

Upvotes: 1

Related Questions