CODEWITHSUNDEEP
r
Adrian
Adrian

Reputation: 9793

R: is.na() does not pick up NA value

So I have a dataset and just by looking at it there are clear NA's in the dataset. 

 > dput(bmi.cig)
structure(list(MSI.subset.BMI = structure(c(4L, 4L, 4L, 4L, 4L, 
4L, 4L, 4L, 4L, 4L, 4L, 1L, 2L, 3L, 3L, 1L, 3L, 3L, 1L, 4L, 4L, 
4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L), .Label = c("0", "1", "2", 
"NA"), class = "factor"), MSI.subset.Cigarette = structure(c(3L, 
3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 2L, 2L, 1L, 2L, 1L, 2L, 
2L, 2L, 2L, 1L, 2L, 1L, 2L, 2L, 2L, 1L, 1L, 1L, 1L), .Label = c("1", 
"2", "NA"), class = "factor")), .Names = c("MSI.subset.BMI", 
"MSI.subset.Cigarette"), row.names = c(NA, 30L), class = "data.frame")


> head(bmi.cig)
  MSI.subset.BMI MSI.subset.Cigarette
1             NA                   NA
2             NA                   NA
3             NA                   NA
4             NA                   NA
5             NA                   NA
6             NA                   NA

I want to remove any row that contains an NA in either column, so I'm using the list-wise deletion function ld in the ForImp package. However, R is not recognizing the NA values. 

is.na(bmi.cig$MSI.subset.BMI)

I get

    > is.na(bmi.cig$MSI.subset.BMI)
 [1] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
[26] FALSE FALSE FALSE FALSE FALSE

So once I use the ld function I just get an empty dataset in return.

Upvotes: 3

Views: 19899

Answers (2)

Rich Scriven
Rich Scriven

Reputation: 99331

As @rbatt mentions, you have character NA values as factor levels. You can remove them and get the NA entries to register as real NA values for the entire data set with

df[] <- lapply(df, function(x) {
    is.na(levels(x)) <- levels(x) == "NA"
    x
})

where df is your data set. And now test with

is.na(df)

Upvotes: 2

rbatt
rbatt

Reputation: 4807

It's b/c the columns are factors, and the levels are "NA". I.e., try

data <- structure(list(MSI.subset.BMI = structure(c(4L, 4L, 4L, 4L, 4L, 
+ 4L, 4L, 4L, 4L, 4L, 4L, 1L, 2L, 3L, 3L, 1L, 3L, 3L, 1L, 4L, 4L, 
+ 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L), .Label = c("0", "1", "2", 
+ "NA"), class = "factor"), MSI.subset.Cigarette = structure(c(3L, 
+ 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 2L, 2L, 1L, 2L, 1L, 2L, 
+ 2L, 2L, 2L, 1L, 2L, 1L, 2L, 2L, 2L, 1L, 1L, 1L, 1L), .Label = c("1", 
+ "2", "NA"), class = "factor")), .Names = c("MSI.subset.BMI", 
+ "MSI.subset.Cigarette"), row.names = c(NA, 30L), class = "data.frame")
> class(blah[,1])

data[,1]=="NA"

The NA's are actually characters (class("NA")), not class logical like class(NA).

Upvotes: 5

Related Questions