How to select columns in R without missing values?

There are two easy methods to select columns of an R data frame without missing values, first one results in a vector and other returns a matrix. For example, if we have a data frame called df then the first method can be used as df[,colSums(is.na(df))==0] and the second method will be used as t(na.omit(t(df))).

Example

Consider the below data frame −

 Live Demo

df1<-data.frame(x1=sample(c(NA,5),20,replace=TRUE),x2=rpois(20,5)) df1

Output

   x1   x2 1  NA  10 2   5   3 3  NA   6 4  NA   9 5  NA   2 6   5   1 7  NA   3 8  NA   4 9  NA   4 10  5   0 11 NA   4 12 NA   1 13  5   7 14  5   6 15 NA   4 16  5   7 17  5   8 18 NA   1 19 NA   5 20  5   5

Selecting columns of df1 that do not contain missing values using first method −

Example

df1[ , colSums(is.na(df1)) == 0]

Output

[1] 10 3 6 9 2 1 3 4 4 0 4 1 7 6 4 7 8 1 5 5

Selecting columns of df1 that do not contain missing values using second method −

Example

t(na.omit(t(df1))) 

Output

       x2 [1,]  10 [2,]   3 [3,]   6 [4,]   9 [5,]   2 [6,]   1 [7,]   3 [8,]   4 [9,]   4 [10,]  0 [11,]  4 [12,]  1 [13,]  7 [14,]  6 [15,]  4 [16,]  7 [17,]  8 [18,]  1 [19,]  5 [20,]  5

attr(,"na.action") x1 1 attr(,"class") [1] "omit"

Example

 Live Demo

df2<-data.frame(y1=sample(c(NA,2),20,replace=TRUE),y2=rnorm(20,5,0.23)) df2

Output

    y1     y2 1  NA   4.926885 2  NA   4.880143 3   2   4.999657 4   2   5.101575 5  NA   4.778465 6  NA   4.534625 7  NA   5.341219 8   2   5.462156 9   2   4.553130 10 NA   5.282899 11 NA   4.884406 12 NA   4.960496 13  2   4.913396 14 NA   4.780874 15 NA   5.222774 16  2   5.007299 17 NA   4.900481 18  2   4.839670 19  2   5.107357 20 NA   4.874243

Selecting columns of df2 that do not contain missing values using second method −

Example

t(na.omit(t(df2)))

Output

       y2 [1,]  4.926885 [2,]  4.880143 [3,]  4.999657 [4,]  5.101575 [5,]  4.778465 [6,]  4.534625 [7,]  5.341219 [8,]  5.462156 [9,]  4.553130 [10,] 5.282899 [11,] 4.884406 [12,] 4.960496 [13,] 4.913396 [14,] 4.780874 [15,] 5.222774 [16,] 5.007299 [17,] 4.900481 [18,] 4.839670 [19,] 5.107357 [20,] 4.874243

attr(,"na.action") y1 1 attr(,"class") [1] "omit"