Good evening, all:
Back again. I have now found, set up, and
run glmmPQL. I need some help in reading the output.
I am trying to find whether there is
meaningful residual center-to-center variability among dialysis centers in the
odds that a patient will have a type of diagnostic test, corrected for the
patients’ baseline characteristics. The focus here is on whether there
is real center-to-center variability, not on the baseline predictors. I ran:
test.glmm <- glmmPQL(fixed = [two sided
formula of fixed effects], random = ~1 | Center, family=binomial(link=logit),
data="" dispersion = NULL,…)
This ran and converged.
The output included a lot of stuff that I
understand, but also the following:
Random effects:
Formula: ~1 | Center
(Intercept) (Residual)
StdDev: 0.4090162 0.9714077
This is the part that I don’t quite
get.
a. The intercept, I assume, is the StdDev
of the random effect. How do I test whether the StdDev is significantly
different from 0? What is its scale? (The naïve median of the chance of the
test at each center is about 19%, with a range from 0 to 100%.)
b: What is the Residual? Is this on the
same scale as the Intercept? I.e., can one say that 0.4090162/(0.4090162+0.9714077)
of the variability is “explained” by including the random effect? Does
the fact that the Residual is very close to 1.0 mean that there is essentially
no over dispersion once I include the random Center effect?
c. Should I be testing the significance
of the random effect by using a likelihood ratio test comparing with the same
model above, but setting:
random = ~ 1 (without the “|
Center”)? This gives me a huge difference in log likelihood.
d. Is there somewhere that I can read up
to understand this function?
Many thanks in advance for any help you
can give me.
Larry Hunsicker
From: Jimenez-Leal
William [mailto:william.jimenezleal@lsc.gov.uk]
Sent: Tuesday, April 10, 2007
10:32 AM
To: Hunsicker, Lawrence;
s-news@lists.biostat.wustl.edu
Subject: RE: [S] Logistic
regression with random effects
Did you try glmmPQL() ?
It actually works by calling the lme.
William
From:
s-news-owner@lists.biostat.wustl.edu
[mailto:s-news-owner@lists.biostat.wustl.edu] On
Behalf Of Hunsicker, Lawrence
Sent: 10 April 2007 16:15
To: s-news@lists.biostat.wustl.edu
Subject: [S] Logistic regression
with random effects
Good
morning, all:
I
suppose that this must be the thousandth time someone has asked this question,
and I apologize that I don’t know how to look up past questions and
answers. I am trying to study whether there is a significant difference
in outcomes among centers providing a kind of medical service. The
outcome
is binary. There are a batch of individually varying covariates of
importance, but the real focus is on whether, after correcting for these
covariates, there is a meaningful variability among
centers in outcome. I am not interested in the specific centers, but
rather in the overall distribution of underlying center effect. It would
seem that the appropriate statistical method is logistic regression
with a random center effect. I can do this with Egret, but I am
wondering whether it is possible to do this in S-Plus. In S-Plus we have
lme for linear random effects, and we have glm for estimating logistic
regression. But is there something that combines these two?
As
always, thanks in advance to anyone that can help me.
Larry
Hunsicker