Return vector of names where NA is greater than a set value

user_14852823 0
Viewed 113

I want to create a vector of column names for columns that have fewer than 3 NAs. Here is my data and what I have tried so far:

df <- data.frame("A" =  c(1,3,4,6,NA,3,NA),
           "B" = c(2,4,5,6,7,8,9),
           "C" = c(3,4,NA,NA,2,3,NA))

NA.list <- map(df, function (x) sum(is.na(x))>=3)

I want to create an output that is similar to this:

names <- c("A", "B")

Any thoughts?

purrr dataframe iteration r function
3 Answers

a simple solution

colnames(df[colSums(is.na(df)) < 3])

[1] "A" "B"
user_28848590

Using Filter :

names(Filter(function(x) x < 3, colSums(is.na(df))))
#[1] "A" "B"

With the same logic you can also use purrr::keep

names(purrr::keep(df, ~sum(is.na(.x)) < 3))
user_39629140

With dplyr we can use select

library(df)
df %>%
    select(where(~ sum(is.na(.)) < 3)) %>%
    names
#[1] "A" "B"

Or using base R

names(which(sapply(df, function(x) sum(is.na(x))) < 3))
#[1] "A" "B"
user_37322710
Related