Tukukkk

How can I avoid using a loop to subset a dataframe based on multiple factor levels?

In the following example my desired output is a dataframe. The dataframe should contain the rows of the original dataframe where the value in "Code" equals one of the values in "selected".

Working example:

#sample data

Code<-c("A","B","C","D","C","D","A","A")

Value<-c(1, 2, 3, 4, 1, 2, 3, 4)

data<-data.frame(cbind(Code, Value))



selected<-c("A","B") #want rows that contain A and B



#Begin subsetting

result<-data[which(data$Code==selected[1]),]

s1<-2

while(s1<length(selected)+1)

{

  result<-rbind(result,data[which(data$Code==selected[s1]),])

  s1<-s1+1

}

This is a toy example of a much larger dataset, so "selected" may contain a great number of elements and the data a great number of rows. Therefore I would like to avoid the loop.

You can use %in%

  data[data$Code %in% selected,]

  Code Value

1    A     1

2    B     2

7    A     3

8    A     4

Try this:

> data[match(as.character(data$Code), selected, nomatch = FALSE), ]

    Code Value

1      A     1

2      B     2

1.1    A     1

1.2    A     1

Here's another:

data[data$Code == "A" | data$Code == "B", ]

It's also worth mentioning that the subsetting factor doesn't have to be part of the data frame if it matches the data frame rows in length and order. In this case we made our data frame from this factor anyway. So,

data[Code == "A" | Code == "B", ]

also works, which is one of the really useful things about R.