How to convert integer to factor?

Question

Say I have the following df

x<-c(1,3,2,4,1,3,2,4,1,3,2,4)
y<-c(rnorm(12))
df<-data.frame(x,y)

x is integer and I try to convert it to a factor using the suggestion from the R cookbook:

df[,'x']<-factor(df[,'x'])

But it still remains an integer. I have tried a solution found on the forum

df[,'x']<-lapply(df[,'x'], factor)

but it did not work either. Thanks.

the lapply approach gives an error when I try it, the second approach gives the desired output... — Liz Young, Jul 01 '15 at 15:12

erasmortg · Accepted Answer · 2015-07-01T15:18:49.540

11

It seems that your original solution works. Check the typeof() as well as the class():

x<-c(1,3,2,4,1,3,2,4,1,3,2,4)
y<-c(rnorm(12))
df<-data.frame(x,y)
typeof(df$x)
#"double"    
class(df$x)
#"numeric"
df[,'x']<-factor(df[,'x'])
typeof(df$x)
 #"integer"
class(df$x)
#"factor"

edited Jul 01 '15 at 15:18

answered Jul 01 '15 at 15:13

erasmortg

3,246
1
17
34

1

Thanks, erasmortg. When I check with class it shows 'factor'. I hope it is indeed a factor then. Do not understand why typeof shows integer. I also just tried str(df) and it shows that x is a factor. – Vasile Jul 01 '15 at 15:18

score 4 · Answer 2 · answered Jul 01 '15 at 15:07

4

may be try

df$x=as.factor(df$x)

answered Jul 01 '15 at 15:07

Batanichek

7,761
31
49

2

Tried did not work. Still an integer – Vasile Jul 01 '15 at 15:09
This more often returns NAs. – M-- Aug 15 '21 at 21:25
@M-- with df in example work without NAs , can you provide example with NAs? – Batanichek Aug 19 '21 at 09:08
Right, that's why I said more often (talking about the solution in general with other type of dataframes/columns) – M-- Aug 19 '21 at 16:39
@M-- can you provide example when you got NAs? – Batanichek Aug 23 '21 at 09:25

score 3 · Answer 3 · answered Jul 01 '15 at 15:22

3

Try this.

df[,'x'] <- as.factor(as.character(df[,'x']))

answered Jul 01 '15 at 15:22

Shiva

789
6
15

score 1 · Answer 4 · answered Jul 01 '15 at 15:16

The code

df[,'x']<-factor(df[,'x'])

does in fact work correctly. But its not so obvious using the x values you've chosen of 1,2,3,4. Try it with this instead:

x<-c(10,23,23, 12, 12, 12, 10, 10, 34, 45, 10, 23)
y<-c(rnorm(12))
df<-data.frame(x,y)
df[,'x']<-factor(df[,'x'])
str(df)

Now its easy to see that that the x column has Factor w/ 5 levels "10","12","23" ...

score 0 · Answer 5 · answered Jul 01 '15 at 15:10

0

this will work :

df[,'x'] <- as.factor(df[,'x'])

answered Jul 01 '15 at 15:10

Liz Young

408
5
17

Maybe I am doing something wrong. That is what I did and what I am getting – Vasile Jul 01 '15 at 15:14
df[,'x'] <- as.factor(df[,'x']) > typeof(df$x) [1] "integer" – Vasile Jul 01 '15 at 15:14
use class(df$x) [1] "factor" – Batanichek Jul 01 '15 at 15:14

How to convert integer to factor?

5 Answers5

Linked