[R-sig-ME] how to extract the degrees of freedom in lmer

[Ben Bolker]

  This is really a special case of R FAQ 7.35,

http://cran.r-project.org/doc/FAQ/R-FAQ.html#Why-are-p_002dvalues-not-displayed-when-using-lmer_0028_0029_003f

  The very short answer is that Doug Bates, the author
of lmer, no longer believes in the dominant paradigm of computing F
statistics with a given numerator and a denominator df extracted by some
approximation (preferably one which works correctly when applied
to classical balanced orthogonal nested designs, where the
paradigm is appropriate).  If you need to work in this paradigm
you should probably continue to use lme ...

  (and yes, that *was* "very short" compared to the discussion
referenced in the above FAQ item ...)

  Ben Bolker


Anne Dubois wrote:
> Dear all,
> 
> Previously I was using "lme" and I could extract the denominator degrees
> of freedom   with : summary(fit)$tTable[,3]
> 
> Now, I am trying to do the same in "lmer" but I do not know if it is
> possible because the degrees of freedom does not appear in the summary.
> Can you help me, please ?
> 
> Thank you for your time.
> Sincerely,
> 
> Anne Dubois (anne.dubois at inserm.fr)
> 
> PS : To illustrate my problem, I use the dataset ergoStool (see below)
> 
> 
>> Stool.lme<-lme(effort~Type,random=~1|Subject,data=ergoStool)
>> summary(Stool.lme)
> Linear mixed-effects model fit by REML
> Data: ergoStool
>       AIC      BIC   logLik
>  133.1308 141.9252 -60.5654
> 
> Random effects:
> Formula: ~1 | Subject
>        (Intercept) Residual
> StdDev:    1.332465 1.100295
> 
> Fixed effects: effort ~ Type
>               Value Std.Error DF   t-value p-value
> (Intercept) 8.555556 0.5760123 24 14.853079  0.0000
> TypeT2      3.888889 0.5186838 24  7.497610  0.0000
> TypeT3      2.222222 0.5186838 24  4.284348  0.0003
> TypeT4      0.666667 0.5186838 24  1.285304  0.2110
> Correlation:
>       (Intr) TypeT2 TypeT3
> TypeT2 -0.45              TypeT3 -0.45   0.50       TypeT4 -0.45  
> 0.50   0.50
> 
> Standardized Within-Group Residuals:
>        Min          Q1         Med          Q3         Max
> -1.80200345 -0.64316591  0.05783115  0.70099706  1.63142054
> 
> Number of Observations: 36
> Number of Groups: 9
>> summary(Stool.lme)$tTable[,3]
> (Intercept)      TypeT2      TypeT3      TypeT4
>         24          24          24          24
> 
> 
> 
>> Stool.lmer<-lmer(effort~Type+(1|Subject),data=ergoStool)
>> summary(Stool.lmer)
> Linear mixed model fit by REML
> Formula: effort ~ Type + (1 | Subject)
>   Data: ergoStool
>   AIC   BIC logLik deviance REMLdev
> 133.1 142.6 -60.57    122.1   121.1
> Random effects:
> Groups   Name        Variance Std.Dev.
> Subject  (Intercept) 1.7753   1.3324 Residual             1.2107  
> 1.1003 Number of obs: 36, groups: Subject, 9
> 
> Fixed effects:
>            Estimate Std. Error t value
> (Intercept)   8.5556     0.5760  14.853
> TypeT2        3.8889     0.5187   7.498
> TypeT3        2.2222     0.5187   4.284
> TypeT4        0.6667     0.5187   1.285
> 
> Correlation of Fixed Effects:
>       (Intr) TypeT2 TypeT3
> TypeT2 -0.450             TypeT3 -0.450  0.500      TypeT4 -0.450 
> 0.500  0.500
> 
> 
>