Recode categorical variable to binary (0/1)

Question

Could someone help me regarding the use of ifelse.

I have a data.frame (dat) with a categorical variable/factor called Q1 (dat$Q1). dat$Q1 was coded as 1,2,3 or 4. I need to create a new column data$new1 based on the following rule:

if dat$Q1 == 3 then dat$new1 should be 1. otherwise, dat$new1 should be 0.

What is the most efficient way of doing this please?

score 8 · Answer 1 · answered Apr 16 '13 at 01:58

8

Use ifelse as in:

dat$new1 <- ifelse(dat$Q1==3, 1, 0)

answered Apr 16 '13 at 01:58

Tyler Rinker

108,132
65
322
519

Thanks for the speedy response!! New to this site but I am very impressed :-) – user2284663 Apr 16 '13 at 02:07

score 6 · Answer 2 · answered Apr 16 '13 at 02:18

6

Just:

 dat$new1 <-  0+(dat$Q1==3)  # or use as.numeric(.)

answered Apr 16 '13 at 02:18

IRTFM

258,963
21
364
487

`as.numeric` has the edge using `microbenchmark`. It seems more idiomatic as well. +1 – Zelazny7 Apr 16 '13 at 03:19
Agreed, and if the goal were to have the values be 2 and 1 then use `1+as.numeric(dat$Q1==3)`. – IRTFM Apr 16 '13 at 05:45

Recode categorical variable to binary (0/1)

2 Answers2

Linked

Related