Removing all binary variables from the data

Question

I have data as follows:

df <- data.frame(A=c(1,2,3), B=c(1,0,1), C=c(0.1, 0.011, 0.3), D=c(0, 0.5, 1))

  A B     C   D
1 1 1 0.100 0.0
2 2 0 0.011 0.5
3 3 1 0.300 1.0

Ho can I remove all binary variables (= B) from this data.frame?

score 3 · Answer 1 · answered Jun 10 '22 at 11:57

3

You could do

library(dplyr)

df %>% select(!where(~all(.x %in% 0:1)))
  A     C   D
1 1 0.100 0.0
2 2 0.011 0.5
3 3 0.300 1.0

answered Jun 10 '22 at 11:57

Allan Cameron

Quinten · Accepted Answer · 2022-06-10T12:13:26.920

3

You can use this:

Filter(function(x) !all(x %in% c(0, 1)), df)

Output:

  A     C   D
1 1 0.100 0.0
2 2 0.011 0.5
3 3 0.300 1.0

Note: you could also add NA to c(0, 1) if needed.

edited Jun 10 '22 at 12:13

answered Jun 10 '22 at 11:58

Quinten

Sotos · Answer 3 · 2022-06-10T12:14:49.520

2

Another option for fun, We can also explicity search for each value (as @Allan Cameron mentions), i.e.

df[colSums(df ==1 | df == 0) != nrow(df)]

#  A     C   D
#1 1 0.100 0.0
#2 2 0.011 0.5
#3 3 0.300 1.0

edited Jun 10 '22 at 12:14

answered Jun 10 '22 at 12:01

Sotos

maybe `%in%` instead of `==` or `df == 1 | df == 0` – Allan Cameron Jun 10 '22 at 12:08
@AllanCameron `%in%` does not work. The `or` operator does on the other hand – Sotos Jun 10 '22 at 12:10
@Sotos, Great option! Is it also possible if `NA` is needed? – Quinten Jun 10 '22 at 12:10
3

This is incorrect. it works because of vector recycling. And you know that. for example, try `df1 <- rbind(df, 1)` and now try `df1 == c(0,1)` and see what happens – Onyambu Jun 10 '22 at 12:12

3 Answers3