R help - Sep 2015 - urgent question about Lmer models

Dear r-help,

I have a question about the std error of a lmer model with library
(LmerTest).

I have 12 sites with 6 treatments over each site. I measured some response
variable (biomass_of_insects) (with a gaussian distribution).

I did :
library (LmerTest)
model<- lmer((biomass_of_insects) ~ as.factor(treatments) + (1 | sites))

However, the response of the model show always the same std error (see
below):

Fixed effects:
                       Estimate Std. Error      df t value Pr(>|t|)
(Intercept)             501.333     80.656  66.000   6.216 3.91e-08 ***
as.factor(treat)F       126.667    114.065  66.000   1.110    0.271
as.factor(treat)I        -8.333    114.065  66.000  -0.073    0.942
as.factor(treat)I+F1/3  -75.000    114.065  66.000  -0.658    0.513
as.factor(treat)I+F2/3   18.333    114.065  66.000   0.161    0.873
as.factor(treat)I+F3/3   15.917    114.065  66.000   0.140    0.889



Do you know why std.error is always the same ?

Thank you very much,

Aline

	[[alternative HTML version deleted]]

You need to consult a local statistical expert or post on a statistics
list like stats.stackexchange.com .  This is a statistical question
not an R question.

The answer is: Because the design is balanced and the treatment error
is obtained from the (within site) rsd, but you will probably need
more explanation than this.

BTW, if you haven't already done so, try making some informative
trellised plots to understand what is going on. Formal statistical
analysis alone can be very misleading.

Cheers,
Bert

Bert Gunter

"Data is not information. Information is not knowledge. And knowledge
is certainly not wisdom."
   -- Clifford Stoll


On Thu, Sep 3, 2015 at 8:57 AM, Aline Andrey <aline.andrey1 at gmail.com>
wrote:
> Dear r-help,
>
> I have a question about the std error of a lmer model with library
> (LmerTest).
>
> I have 12 sites with 6 treatments over each site. I measured some response
> variable (biomass_of_insects) (with a gaussian distribution).
>
> I did :
> library (LmerTest)
> model<- lmer((biomass_of_insects) ~ as.factor(treatments) + (1 | sites))
>
> However, the response of the model show always the same std error (see
> below):
>
> Fixed effects:
>                        Estimate Std. Error      df t value Pr(>|t|)
> (Intercept)             501.333     80.656  66.000   6.216 3.91e-08 ***
> as.factor(treat)F       126.667    114.065  66.000   1.110    0.271
> as.factor(treat)I        -8.333    114.065  66.000  -0.073    0.942
> as.factor(treat)I+F1/3  -75.000    114.065  66.000  -0.658    0.513
> as.factor(treat)I+F2/3   18.333    114.065  66.000   0.161    0.873
> as.factor(treat)I+F3/3   15.917    114.065  66.000   0.140    0.889
>
>
>
> Do you know why std.error is always the same ?
>
> Thank you very much,
>
> Aline
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

Bert Gunter <bgunter.4567 <at> gmail.com> writes:
> 
> You need to consult a local statistical expert or post on a statistics
> list like stats.stackexchange.com .  This is a statistical question
> not an R question.
> 
> The answer is: Because the design is balanced and the treatment error
> is obtained from the (within site) rsd, but you will probably need
> more explanation than this.
> 
> BTW, if you haven't already done so, try making some informative
> trellised plots to understand what is going on. Formal statistical
> analysis alone can be very misleading.
> 
> Cheers,
> Bert
> 
> Bert Gunter
   That's a good explanation, very similar to the one I posted
at http://stackoverflow.com/questions/32383259/
   lmertest-package-standard-errors  (broken URL for Gmane).

  I know most people who read this message will already know this,
but (to the original poster) *please don't cross-post to R lists and
StackOverflow*; it causes wasted effort, as in this case.

  Ben Bolker