Grouping variable using if statement in R

Question

I would like to do a simple if statement to group codes into groups. The variable has number codes and I would like to create a new variable that groups several number codes together. I have written the following if statement but because they are many code numbers (30 codes), I need help writing a more elegant code to group the variable rather than writing 30+ if statements.

Data2$RevisedSIC.Group <-c()
for (i in 1:length(Data2$SIC.Group )) {
if (Data2$SIC.Group[i] =="10110") Data2$RevisedSIC.Group [i]="Metal" else 
if (Data2$SIC.Group[i] =="10410") Data2$RevisedSIC.Group [i]="Metal" else
if (Data2$SIC.Group[i] =="10439") Data2$RevisedSIC.Group [i]="Metal" else
if (Data2$SIC.Group[i] =="14111") Data2$RevisedSIC.Group [i]="Stone" else
if (Data2$SIC.Group[i] =="10421") Data2$RevisedSIC.Group [i]="Stone" }

Pierre Lapointe · Accepted Answer · 2012-01-25 00:49:54Z

1

There is no need for a loop

Data2 <-data.frame(rep(c(10110,10410,10439,14111),2))
colnames(Data2) <-"SIC.Group"

Data2$RevisedSIC.Group[Data2$SIC.Group %in% c(10110,10410,10439)] <- "Metal"
Data2$RevisedSIC.Group[Data2$SIC.Group %in% 14111] <- "Stone"

  SIC.Group RevisedSIC.Group
1     10110            Metal
2     10410            Metal
3     10439            Metal
4     14111            Stone
5     10110            Metal
6     10410            Metal
7     10439            Metal
8     14111            Stone

answered Jan 25, 2012 at 0:49

Pierre Lapointe

16.3k2 gold badges47 silver badges59 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Justin · Accepted Answer · 2012-01-25 00:42:59Z

1

take a look at match

lookup <- data.frame(code=c('10110','10410','10439','14111','10421'), name=c('Metal','Metal','Metal','Stone','Stone'))

Data2$RevisedSIC.Group <- lookup$name[match(Data2$SIC.Group,lookup$code)]

answered Jan 25, 2012 at 0:42

Justin

43.4k9 gold badges96 silver badges111 bronze badges

1 Comment

Amateur Over a year ago

Thank you everyone for your time and answers!

mathematical.coffee · Accepted Answer · 2012-01-25 00:48:41Z

You can use %in%:

metals <- c("10110","10410","10439")
stones <- c("14111","10421")
# ... and so on for each group

if ( Data2$SIC.Group[i] %in% metals ) {
    Data2$SIC.Group[i] <- "Metal"
} else if ( Data2$SIC.Group[i] %in% stones ) {
    Data2$SIC.Group[i] <- "Stone"

} # ... and so on for each group

You'd still need to write as many if/else as there are groups, which is hopefully much less than the number of codes you have.

To condense it further you could dispense with the loop:

metals <- c("10110","10410","10439")
stones <- c("14111","10421")
# ... and so on for each group

Data2$SIC.Group[ Data2$SIC.Group %in% metals ] <- "Metal"
Data2$SIC.Group[ Data2$SIC.Group %in% stones ] <- "Stone"
# ... and so on for each group

To condense even further, you can define your groups in a list as below and then just have one line of code for all groups in the assigning step:

groupCodes <- list(
   metals=c("10110","10410","10439"),
   stones=c("14111","10421")
   # ... and so on for each group
)

for (n in names(groupCodes)) {
    # just once for all groups.
    Data2$SIC.Group[ Data2$SIC.Group %in% groupCodes[[n]] ] <- n
}

Collectives™ on Stack Overflow

Grouping variable using if statement in R

3 Answers 3

Comments

1 Comment

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

1 Comment

Comments

Your Answer

Sign up or log in

Post as a guest

Related