I'm trying to recode this variable, after transforming it in numeric:
e18$AntPT <- recode(e18$AntPT, 1 <- 0, 0 <- c(2:10))
but it returns
Error in 1 <- 0: invalid (do_set) left-hand side to assignment
Also, the final result should include these 3 parts of the code:
e18$AntPT <- as.numeric(e18$Q1501)
e18$AntPT <- recode(e18$AntPT, 1<-1, 0 <- c(2:10))
e18$AntPSDB <- as.numeric(e18$Q1505)
e18$AntPSDB <- recode(e18$AntPSDB, 10<-1, 0 <- c(2:10))
e18$AntPMDB <- as.numeric(e18$Q1502)
e18$AntPMDB <- recode(e18$AntPMDB, 100<-1, 0 <- c(2:10))
The last thing I have to do is:
e18$Ant_Part <- (e18$AntPT + e18$AntPSDB + e18$AntPMDB)
The final result of this sum must give me the numbers 0, 1, 10, 11, 100, 101, 110, 111 - which one of the numbers give me a different interpretation of my initial question.
If there are other numbers in the column and want to change only specific numbers
-before
e18
# AntPT AntPSDB AntPMDB
#1 1 1 100
#2 2 14 1
#3 4 2 2
#4 3 15 105
#5 1 7 24
#6 2 10 7
#7 3 11 8
Now, we loop over the columns of interest with lapply and replace the values in the column to 0 where they are from 2 to 10
e18[c("AntPT", "AntPSDB", "AntPMDB")] <- lapply(e18[c("AntPT",
"AntPSDB", "AntPMDB")], function(x) replace(x, x %in% 2:10, 0))
and the specific numbers in columns 'AntPSDB', 'AntPMDB' (10, 100) to 1
e18$AntPSDB[e18$AntPSDB == 10] <- 1
e18$AntPMDB[e18$AntPMDB == 100] <- 1
-after
e18
# AntPT AntPSDB AntPMDB
#1 1 1 1
#2 0 14 1
#3 0 0 0
#4 0 15 105
#5 1 0 24
#6 0 0 0
#7 0 11 0
and then do the sum
e18$Ant_Part <- (e18$AntPT + e18$AntPSDB + e18$AntPMDB)
e18 <- data.frame(AntPT = c(1, 2, 4, 3, 1, 2, 3),
AntPSDB = c(1, 14, 2, 15, 7, 10, 11),
AntPMDB = c(100, 1, 2, 105, 24, 7, 8))