[R-sig-ME] Testing significance of random effect in lme () and lmer()

[David Duffy]

On Tue, 24 Aug 2010, R.S. Cotter wrote:
> Is this the right way to state the results: The random effect ID (likelihood
> ratio test; c2=12.26, d.f.=1, p=0.0005)?
>
> Is this the right way to state the results: The random effect ID (likelihood
> ratio test; p<0.0001)? Or should I also state the value L.Ratio?

Why would you give the value of the likelihood ratio test for lme but not 
lmer?  And the P-value, as mentioned recently here recently (and see Ben 
Bolker's http://glmm.wikidot.com/faq, and Fabian Scheipl's RLRsim package) 
is too conservative, though it won't matter here.

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