Delete columns based on average

Question

I have this matrix:

row c1  c2  c3  c4  c5
R1  27  38  94  40  4
R2  69  16  85  2   15
R3  30  35  64  95  6
R4  20  33  77  98  55
R5  20  44  60  33  89
R6  12  88  87  44  38

I want to delete all columns in the matrix that have an average of less than 30. How is this done with colMeans?

Answer 1

We can use colMeans (assuming the row is the row.names of the matrix)

m1[, colMeans(m1, na.rm = TRUE) >= 30, drop = FALSE]

-output

   c2 c3 c4 c5
R1 38 94 40  4
R2 16 85  2 15
R3 35 64 95  6
R4 33 77 98 55
R5 44 60 33 89
R6 88 87 44 38

if row is a column, then it is not making sense as a matrix because a matrix can have only a single type and thus any character element changes the whole matrix to character. Probably a data.frame would be better i.e.

df1[, c(TRUE, colMeans(df1[-1], na.rm = TRUE) >= 30), drop = FALSE]

-output

 row c2 c3 c4 c5
1  R1 38 94 40  4
2  R2 16 85  2 15
3  R3 35 64 95  6
4  R4 33 77 98 55
5  R5 44 60 33 89
6  R6 88 87 44 38

data

m1 <- structure(c(27L, 69L, 30L, 20L, 20L, 12L, 38L, 16L, 35L, 33L, 
44L, 88L, 94L, 85L, 64L, 77L, 60L, 87L, 40L, 2L, 95L, 98L, 33L, 
44L, 4L, 15L, 6L, 55L, 89L, 38L), .Dim = 6:5, .Dimnames = list(
    c("R1", "R2", "R3", "R4", "R5", "R6"), c("c1", "c2", "c3", 
    "c4", "c5")))

df1 <- structure(list(row = c("R1", "R2", "R3", "R4", "R5", "R6"), c1 = c(27L, 
69L, 30L, 20L, 20L, 12L), c2 = c(38L, 16L, 35L, 33L, 44L, 88L
), c3 = c(94L, 85L, 64L, 77L, 60L, 87L), c4 = c(40L, 2L, 95L, 
98L, 33L, 44L), c5 = c(4L, 15L, 6L, 55L, 89L, 38L)), 
class = "data.frame", row.names = c(NA, 
-6L))

Answer 2

If we change matrix to dataframe we could use select with condition: (data borrowed from akrun)

df1 %>% 
    select(which(colMeans(.) >= 30))

output:

   c2 c3 c4 c5
R1 38 94 40  4
R2 16 85  2 15
R3 35 64 95  6
R4 33 77 98 55
R5 44 60 33 89
R6 88 87 44 38