R help - Sep 2014 - "missings=function(x) x[x==998|x==999]<-NA" doesn't work...

Hi!

I want to have a function that assigns NAs to certain values of my 
variable "var" in the dataset "d". This doesn't work:
> missings=function(x) x[x==998|x==999]<-NA
> missings(d$var)
> table(d$var, useNA="always")
    0      1  999 <NA> 
 220  752  321 5264 

I don't get any error messages, but "d$var" remains unchanged. The
function:
> missings=function(x) x[x==90|x==99]<<-NA
doesn't work either, and I read that "<<-" is
"dangerous" anyway?

It is important for me to work with variable names (and therefore with 
functions instead loops) because the number and order of variables in my 
dataset changes regularly.

Thank you,
Doreen
	[[alternative HTML version deleted]]

You need to assign the output of missings() to something. For that
matter, missings() needs some output.

d <- data.frame(a=1:5, b=6:10, var=c(1, 1, 998, 999, 2))

missings <- function(x) {
    x[x==998|x==999]<-NA
    x
}

d$var <- missings(d$var)

> d
  a  b var
1 1  6   1
2 2  7   1
3 3  8  NA
4 4  9  NA
5 5 10   2


Sarah

On Thu, Sep 18, 2014 at 10:13 AM, Doreen Mueller <Doreen.Mueller at
dza.de> wrote:
> Hi!
>
> I want to have a function that assigns NAs to certain values of my
> variable "var" in the dataset "d". This doesn't
work:
>
>> missings=function(x) x[x==998|x==999]<-NA
>> missings(d$var)
>> table(d$var, useNA="always")
>
>     0      1  999 <NA>
>  220  752  321 5264
>
> I don't get any error messages, but "d$var" remains
unchanged. The
> function:
>> missings=function(x) x[x==90|x==99]<<-NA
> doesn't work either, and I read that "<<-" is
"dangerous" anyway?
>
> It is important for me to work with variable names (and therefore with
> functions instead loops) because the number and order of variables in my
> dataset changes regularly.
>
> Thank you,
> Doreen


-- 
Sarah Goslee
http://www.functionaldiversity.org

Hi,

On Sep 18, 2014, at 10:13 AM, Doreen Mueller <Doreen.Mueller at dza.de>
wrote:
> Hi!
> 
> I want to have a function that assigns NAs to certain values of my 
> variable "var" in the dataset "d". This doesn't
work:
> 
>> missings=function(x) x[x==998|x==999]<-NA
>> missings(d$var)
>> table(d$var, useNA="always")
> 
>    0      1  999 <NA> 
> 220  752  321 5264 
> 
> I don't get any error messages, but "d$var" remains
unchanged. The
> function:
>> missings=function(x) x[x==90|x==99]<<-NA
> doesn't work either, and I read that "<<-" is
"dangerous" anyway?
> 
You are so close.  R returns the value of the last thing evaluated in your
function.  In this case, the *copy* of your input argument was modified within
the function, but you didn't return the value of the copy to the calling
environment.  You need to explicitly return the modified value.
> missings <- function(x) { x[ (x==998) | (x==999) ] <- NA ; return(x)
}
> missings(990:1010)
 [1]  990  991  992  993  994  995  996  997   NA   NA 1000 1001 1002 1003 1004
1005 1006 1007
[19] 1008 1009 1010

By the way, don't forget to switch your email client to use text instead of
html when sending a message to the list.

Cheers,
Ben




> It is important for me to work with variable names (and therefore with 
> functions instead loops) because the number and order of variables in my 
> dataset changes regularly.
> 
> Thank you,
> Doreen
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.