It isn't (and never was) indicative of a problem. You can make the warning
go away by using quasibinomial() instead of binomial(), as in the example
on the help page.
Incidentally, it's survey::svyglm (::, not :::)
-thomas
On Wed, 6 Aug 2008, Ben Domingue wrote:
> Howdy,
> Referencing the below exchange:
> https://stat.ethz.ch/pipermail/r-help/2006-April/103862.html
> I am still getting the same warning ("non-integer #successes in a
> binomial glm!") when using svyglm:::survey. Using the API data:
>
> library(survey)
> data(api)
> #stratified sample
> dstrat<-svydesign(id=~1,strata=~stype, weights=~pw, data=apistrat,
fpc=~fpc)
> svyglm(sch.wide~meals+acs.core+hsg,dstrat,family="binomial")
>
> I get:
>
> Stratified Independent Sampling design
> ...
> Warning message:
> In eval(expr, envir, enclos) : non-integer #successes in a binomial glm!
>
> Is this still indicative of a problem? Thanks,
> Ben
>
> ______________________________________________
> 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.
>
Thomas Lumley Assoc. Professor, Biostatistics
tlumley at u.washington.edu University of Washington, Seattle