[R] 'trim' must be numeric of length one？

[Gavin Simpson]

On Mon, 2007-05-28 at 10:58 +0800, Ruixin ZHU wrote:
> Hi everybody,
>  
> When I followed a practice example, I got an error as follows:
> ########################################################################
> #######################################
> > cc<-read.table('example5_2.dat',header=TRUE)
> > cc
>   EXAM1 EXAM2 EXAM3 EXAM4 EXAM5
> 1    45    34    23    35    50
> 2    23    36    66    66    34
> 3    67    59    72    80    69
> 4    56    43    31    34    40
> 5    74    66    57    32    66
> > mean(cc)
> EXAM1 EXAM2 EXAM3 EXAM4 EXAM5 
>  53.0  47.6  49.8  49.4  51.8 
> > attach(cc)
> > mean(EXAM1,EXAM2,EXAM3,EXAM4,EXAM5)
> Error in mean.default(EXAM1, EXAM2, EXAM3, EXAM4, EXAM5) : 
>         'trim' must be numeric of length one

Why did you think that mean would work in the way you used it?

Reading ?mean shows that the default method for mean has a first
argument 'x', and second argument 'trim', plus some others. So in your
2nd example, you passed EXAM1 as argument 'x' and then EXAM2 as 'trim',
and the other EXAMx variables as other arguments. R was not expecting a
vector as argument 'trim' and rightly complained.

The reason the first example worked is that there is a method for data
frames (see the first entry in the usage section of ?mean) - where you
correctly passed cc as argument 'x' as the function/method requires.

> In addition: Warning message:
> the condition has length > 1 and only the first element will be used in:
> if (na.rm) x <- x[!is.na(x)] 
> Would anybody explain which caused this error, and how to modify it?

What is wrong with the first example you used? Why do you need to get
the means by specifying all the variables explicitly?

There are various ways of getting means other than mean():

lapply(cc, mean)
sapply(cc, mean)
colMeans(cc)

If you want specific columns, either subset the returned object:

mean(cc)[c("EXAM1", "EXAM4")]

or subset the object before calculating the means:

mean(cc[, c("EXAM1", "EXAM4")])
       ^^^

note the extra "," as we need to specify the columns here.

You will need to explain more clearly what you want to do if the above
is not sufficient to solve your problem.

Also, be wary of overly using attach. It can be a handy little tool,
until it bites you in the ass because you forgot to detach/reattach the
object after making some really critical change to the underlying
data/object.

HTH

G

>  
> Thanks!
> _____________________________________________
> Dr.Ruixin ZHU
> Shanghai Center for Bioinformation Technology
> rxzhu at scbit.org
> zhurx at mail.sioc.ac.cn
> 86-21-13040647832
>  
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at stat.math.ethz.ch 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.
-- 
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%
Gavin Simpson                     [t] +44 (0)20 7679 0522
ECRC                              [f] +44 (0)20 7679 0565
UCL Department of Geography
Pearson Building                  [e] gavin.simpsonATNOSPAMucl.ac.uk
Gower Street
London, UK                        [w] http://www.ucl.ac.uk/~ucfagls/
WC1E 6BT                          [w] http://www.freshwaters.org.uk/
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%