[R] tables package - error message of "duplicate values"

[David Winsemius]

On Apr 19, 2013, at 2:03 PM, Santosh wrote:

> Rounding was done to replicate the problem I faced in the original data set...

I got an error every time I tried to use digits=0. My basic rule is that if a parameter choice consistently errors out ... stop using it. I'm fundamentally a pragmatist.

> 2 independent identifiers in a group (shown only for 1 of 7 groups) may have same value of b or c.

I don't understand what that means  ....  or why it should affect the ability of the program to deliver 'a'-level-specific estimates of the mean/sd values of 'b' & 'c' in the environment of data.frame-'a'. I showed that the problem was in not using factor() on the LHS of the formula when the grouping variable is numeric.

-- 
David.

> 
> 
> On Fri, Apr 19, 2013 at 1:53 PM, David Winsemius <dwinsemius at comcast.net> wrote:
> 
> On Apr 19, 2013, at 1:29 PM, Santosh wrote:
> 
> > Dear expeRts!
> > when I try summarize data using "tabular" of tables package..
> >
> > here is an example..
> >> a <-
> > data.frame(a=rep(10,10),b=round(runif(10,10,20)),c=round(runif(10,40,50)))
> >> a
> >    a  b  c
> > 1  10 14 44
> > 2  10 17 40
> > 3  10 13 40
> > 4  10 18 40
> > 5  10 11 49
> > 6  10 16 46
> > 7  10 15 43
> > 8  10 19 49
> > 9  10 19 49
> > 10 10 13 45
> >> tabular((a+1)~(n=1)+Format(dig=0)*(b + c) * (mean + sd),data=a)
> > Error in term2table(rows[[i]], cols[[j]], data, n) :
> >  Duplicate values: a and b
> >
> > Is there a way to let the function  accept the "apparently" duplicate
> > values?
> 
> I don't consider myself an expert, just one who reads the manuals and makes modest adjustments to the help-page examples.  Perhaps if you explained what you thought the correct answer would be ( and can you please explain why you are setting digits=0?
> 
> tabular( 1~Format(digits=2) *(b+c)* (mean + sd),data=a)
> #-------------------------
>      b         c
>      mean sd   mean sd
>  All 14.9  3.0 44.3  2.9
> #-----------------
> # Putting a numeric variable on the LHS appears to fail but wrapping it in factor() allows success:
> 
> > a<-data.frame(a=rep(c(1,2),10),b=round(runif(10,10,20)),c=round(runif(10,40,50)))
> >
> > tabular( (factor(a) +1)~Format(digits=2) *(b+c)* (mean + sd),data=a)
> #----------
>            b         c
>  factor(a) mean sd   mean sd
>  1         15.2  3.7 45.0  2.6
>  2         13.0  3.5 48.2  1.2
>  All       14.1  3.7 46.6  2.6
> 
> 
> --
> 
> David Winsemius
> Alameda, CA, USA
> 
> 

David Winsemius
Alameda, CA, USA