(basic) column manipulation in R. help please! :(

[will22]

hello! i'm somewhat new to R so i've been trying to give myself little exercises to see if i can practice. say i have the following fake dataset:

FactorA <- as.factor(c("a","a","a","a","b","b","b","b","c","c","c","c"))
FactorB <- as.factor(c("d","d","e","e","d","d","e","e","d","d","e","e"))

x <- c(1:12)

datum <- data.frame(FactorA,FactorB,x)

and i would like it to look like

FactorA <- as.factor(c("a","a","a","a","b","b","b","b","c","c","c","c"))
FactorB <- as.factor(c("d","d","e","e","d","d","e","e","d","d","e","e"))

x1 <-c(2,3,3,4,5,6,9,10,12,13,11,12)

data.frame(FactorA, FactorB, x1)

so as you can see i need to selectively choose some columns with certain values of Factor A and certain values of Factor B and add either 1 or 2 or 3 to certain numbers. here's what i've tried so far:

for (i in length(datum$FactorA)) {

if (datum$FactorA[i]=="a" & datum$FactorB[i]=="d") {

datum$x[i] <- datum$x[i]+1

}
else {

if (datum$FactorA[i]=="b" & datum$FactorB[i]=="e")

datum$x[i]<- datum$x[i]+2

else {

if(datum$FactorA[i]=="c" & datum$FactorB[i]=="d") {

datum$x[i]<- datum$x[i]+3

}
}
}
}
datum

but as you can see nothing really happens when i run the code. does anyone know why is that? and does anyone know if there's an R package or function that has automated this? i could see this as being something potentially useful and having to go through each possible combination of in the Factor columns seems a llittle bit cumbersome.

help/suggestions are appreciated!

[smoens]

for (i in 1:length(datum$FactorA)) {

if (datum$FactorA[i]=="a" & datum$FactorB[i]=="d") {

datum$x[i] <- datum$x[i]+1

}
else {

if (datum$FactorA[i]=="b" & datum$FactorB[i]=="e")

datum$x[i]<- datum$x[i]+2

else {

if(datum$FactorA[i]=="c" & datum$FactorB[i]=="d") {

datum$x[i]<- datum$x[i]+3

}
}
}
}
datum

for (i in 1:length(datum$FactorA)) gives the requested result.
In your code i only gets 1 value (i.e. 12 of length(datum$FactorA)) if you make it 1:length(datum$FactorA) it lets i run through the values 1 to 12.

[will22]

oh god... i feel so... so... dumb, heh. thanks for pointing that out! i was going crazy over it since it seemed relatively straight forward.

do you happen to know if there's any R package or function that does this automatically instead of having to go through runs of if/then conditionals?

[smoens]

No problem!

Below an alternative way without for/if-loops, but I'm not aware of any straightforward functions. This is just a way I would approach is, probably someone else can provide a shorter solution.

FactorA <- as.factor(c("a","a","a","a","b","b","b","b","c","c","c","c"))
FactorB <- as.factor(c("d","d","e","e","d","d","e","e","d","d","e","e"))

FactorCombi <- paste(FactorA, FactorB, sep='')
  #> FactorCombi
  # [1] "ad" "ad" "ae" "ae" "bd" "bd" "be" "be" "cd" "cd" "ce" "ce"

refset <- data.frame(matchers = c('ad','be','cd'), vals=1:3)
  #data frame containing to be matched values with respective addition values
  #> refset
  #  matchers vals
  #1       ad    1
  #2       be    2 
  #3       cd    3

result <- sapply(seq_along(FactorCombi), function(x) x + match(FactorCombi[x],refset$matchers))
  #> sapply(seq_along(FactorCombi), function(x) x +
  #      match(FactorCombi[x],refset$matchers))
  #[1]  2  3 NA NA NA NA  9 10 12 13 NA NA

result[is.na(result)] <- which(is.na(result))  
  #assign index values to NA values for which no match was made

datum <- data.frame(FactorA,FactorB,result)