[R-sig-ME] Ordinal Package and post hoc test

[fernando barbero]

Hi David, thank you very much, what you are suggesting is a LRT between a
full model, with the four populations (fm1) and a reduced model (fm2) with
only one population? I understand the LRT approach, but I think that the
reduced model (fm2) should have 3 populations (pop2, pop3 and pop4), so I
can asses if there are differences between pop1 and the other three (then
with another reduced model, with pop1,pop3 and pop4 I could asses
differences between pop2 and the other ones, and like that I would test all
the populations), what do you think about this?
Fernando

-----Mensaje original-----
De: David Duffy [mailto:David.Duffy at qimr.edu.au] 
Enviado el: lunes, 28 de mayo de 2012 08:07 p.m.
Para: fernando barbero
CC: r-sig-mixed-models at r-project.org
Asunto: Re: [R-sig-ME] Ordinal Package and post hoc test

On Mon, 28 May 2012, fernando barbero wrote:

>
fm1<-clmm(leaf~population+(1|population:family),data=d,threshold"flexible",
<            link="probit",Hess=TRUE)

> now I want to asses differences within populations, as visual inspection
of
> the data suggests that one population has got more seedlings in the A
> category than the other 3 populations. I am thinking in something like the
> glht function of the multcomp package (tukey<-glht(fm1,

How about
pop1 <- d$population == "pop1"
fm2  <- clmm(leaf~ pop1 +(1|pop1:family), data=d,
              threshold"flexible", link="probit",Hess=TRUE)
anova(fm1, fm2), with or without a Bonferroni correction, or use AIC.

-- 
| 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
-----
Se certificó que el correo no contiene virus.
Comprobada por AVG - www.avg.es
Versión: 2012.0.1913 / Base de datos de virus: 2425/5030 - Fecha de la
versión: 29/05/2012