[R-sig-ME] 'Missing' tests for lmer models

[John Fox]

Dear Henric,

Thanks for this.

I think that what confused me was that the original message appears in a linked URL, not directly in the list archive.

Anyway, I believe that I effectively answered the question: test="chisq" is equivalent to the previous test="Wald", and test="F" will produce a Kenward-Roger F test. Both tests are Wald tests, hence the change in options. And, as I said, ?Anova could be clearer (and will be in the next release).

Best,
 John

On Thu, 11 Oct 2012 15:15:12 +0200
 "Henric (Nilsson) Winell" <nilsson.henric at gmail.com> wrote:
> Dear John,
> 
> On 2012-10-10 04:52, John Fox wrote:
> 
> > Dear David and Luke,
> >
> > I missed the beginning of this thread and couldn't find it in the R-sig-ME list archive,
> 
> The thread started here:
> https://stat.ethz.ch/pipermail/r-sig-mixed-models/2012q3/019141.html
> 
> and then continued here:
> https://stat.ethz.ch/pipermail/r-sig-mixed-models/2012q4/019236.html
> 
> 
> Best regards,
> Henric
> 
> 
> 
>   so it's not clear to me what the issue is -- sorry -- but I should 
> clarify that Anova() for mer objects only does Wald tests -- either Wald chisquare tests or F-tests, the latter with Kenward-Roger df computed via the pbkrtest package. I think that this could be clearer in ?Anova.
> >
> > Best,
> >   John
> >
> > ------------------------------------------------
> > John Fox
> > Sen. William McMaster Prof. of Social Statistics
> > Department of Sociology
> > McMaster University
> > Hamilton, Ontario, Canada
> > http://socserv.mcmaster.ca/jfox/
> > 	
> > On Wed, 10 Oct 2012 11:14:13 +1000
> >   David Duffy <David.Duffy at qimr.edu.au> wrote:
> >> On Tue, 9 Oct 2012, Luke Duncan wrote:
> >>
> >>> New computer/32bit R:
> >>>
> >>> R version 2.15.1 (2012-06-22)
> >>> Platform: i386-pc-mingw32/i386 (32-bit)
> >>>
> >>> other attached packages:
> >>> [1] lme4_0.999999-0  Matrix_1.0-6  car_2.0-13
> >>>
> >>> Old computer/32bit R:
> >>>
> >>> R version 2.13.1 (2011-07-08)
> >>> Platform: i386-pc-mingw32/i386 (32-bit)
> >>>
> >>> other attached packages:
> >>> [1] lme4_0.999375-42   Matrix_0.9996875-3  car_2.0-11
> >>
> >> Yes, these straddle a moderately large version change in lme4:
> >> see the intro at http://lme4.r-forge.r-project.org/
> >>
> >> The most recent versions of Anova.II.mer() only supports c("chisq", "F"). John Fox may reintroduce Wald tests, but you can always do them yourself.
> >>
> >> Cheers, David Duffy.
> >>
> >> --
> >> | David Duffy (MBBS PhD)                                         ,-_|\
> >> | email: davidD at qimr.edu.au  ph: INT+61+7+3362-0217 fax: -0101  /     *
> >> | Epidemiology Unit, Queensland Institute of Medical Research   \_,-._/
> >> | 300 Herston Rd, Brisbane, Queensland 4029, Australia  GPG 4D0B994A v
> >>
> >> _______________________________________________
> >> R-sig-mixed-models at r-project.org mailing list
> >> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
> >
> > _______________________________________________
> > R-sig-mixed-models at r-project.org mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
> >
> 

------------------------------------------------
John Fox
Sen. William McMaster Prof. of Social Statistics
Department of Sociology
McMaster University
Hamilton, Ontario, Canada
http://socserv.mcmaster.ca/jfox/