could you try this, and then not use factor(age) elsewhere?
sv1 <- update( sv1 , age = factor( age ) )
if that doesn't work, is it possible for you to share a reproducible
example? thanks
On Thu, Sep 10, 2015 at 4:51 PM, Emanuele Mazzola <mazzola at
jimmy.harvard.edu> wrote:
> Hello,
>
> I?m having a weird issue with the function ?svychisq? in package
?survey",
> which would be very helpful for me in this case.
>
> I?m tabulating age categories (a factor variable subdivided into 4
> categories: [18,25), [25, 45), [45,65), [65, 85) ) with respect to
> ethnicity/race (another factor variable subdivided into ?hispanic white?,
> ?non hispanic black?, ?hispanic black?).
>
> I?m perfectly able to get to the ?svytable" object, which looks like
this
>
> > svytable(~age+ETN, design=sv1)
> ETN
> age hisp black hispanic white non hisp black
> [18,25) 26.97019 798.87444 183.61834
> [25,45) 145.19650 4783.47678 854.82748
> [45,65) 104.83682 2537.15021 595.04924
> [65,85] 0.00000 0.00000 0.00000
>
> Since it has last row equal to 0 (which would give me troubles with the
> corresponding chi-square p-value), I try to get rid of it by using
>
> > svytable(~factor(age)+ETN, design=sv1)
> ETN
> factor(age) hisp black hispanic white non hisp black
> [18,25) 26.97019 798.87444 183.61834
> [25,45) 145.19650 4783.47678 854.82748
> [45,65) 104.83682 2537.15021 595.04924
>
> which exactly responds to what I?m looking for and to what I?m expecting.
>
> The design is built by using
>
> sv1 = svydesign(ids=~factor(age)+ETN, weights=~WTFA.n, data=totfor)
>
> Now, if I would like to evaluate the corresponding weighted chi squared
> test, I use
>
> svychisq(~factor(age)+ETN, design=sv1)
>
> but here?s what I get from R:
>
> > svychisq(~factor(age)+ETN, design=sv1)
> Error in `[.data.frame`(design$variables, , as.character(rows)) :
> undefined columns selected
>
> Maybe it is a stupid question but I really can?t figure out where the
> error is.
>
> Could you please help me with this?
> Thanks in advance for any information you will provide me with!
>
> Emanuele
>
> ***********************************************************************
> Emanuele Mazzola, Ph.D.
> Department of Biostatistics & Computational Biology
> Dana-Farber Cancer Institute
> 450 Brookline Ave
> Mail Location: LC1056
> Office Location: Longwood Center, Room 1056
> Boston, MA 02215
> Office phone 617-582-7614
> Fax 617-632-2516
>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.
[[alternative HTML version deleted]]