[R] NA vs. <NA>

[Peter Dalgaard]

Muenchen, Robert A (Bob) wrote:
> Dear R-Helpers,
>
> Why does R show character missing values in vectors as NA and when
> stored in a data frame as <NA>? I've searched but did not find an
> explanation.
>
> Thanks,
> Bob
>
>   
>> gender <- c("f","f","f",NA,"m","m","m","m")
>> gender
>>     
> [1] "f" "f" "f" NA  "m" "m" "m" "m"  #here it lacks brackets.
>   
>> q1 <- c(1,2,2,3,4,5,5,4)
>> q1
>>     
> [1] 1 2 2 3 4 5 5 4
>   
>> myDF <- data.frame(q1,gender)
>> myDF
>>     
>   q1 gender
> 1  1      f
> 2  2      f
> 3  2      f
> 4  3   <NA>  #here it has brackets.
> 5  4      m
> 6  5      m
> 7  5      m
> 8  4      m
>   
It is actually a factor in the latter case

 > data.frame(gender)$gender
[1] f    f    f    <NA> m    m    m    m  
Levels: f m

However, you have the same effect with

 > data.frame(gender,stringsAsFactors=FALSE)
  gender
1      f
2      f
3      f
4   <NA>
5      m
6      m
7      m
8      m

The thing to notice is that the printing is without the quote character. 
We also have

 > noquote(gender)
[1] f    f    f    <NA> m    m    m    m  

And the point in either case is that we need some way to distinguish 
between NA (missing) and "NA" (New Alliance, Noradrenalin, North 
America, Neil Armstrong, etc.)

-- 
   O__  ---- Peter Dalgaard             Øster Farimagsgade 5, Entr.B
  c/ /'_ --- Dept. of Biostatistics     PO Box 2099, 1014 Cph. K
 (*) \(*) -- University of Copenhagen   Denmark      Ph:  (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)              FAX: (+45) 35327907