[R-sig-ME] New to LMER with 2 (easy?) questions...

[Hank Stevens]

HI Rolf,
The BLOCK numbers should reflect their "true" underlying nature.
These data would suggest that Site and Block are crossed because e.g.
BLOCK 1 exists at two sites:
SITE  BLOCK
1    1
1    2
1    3
2    1
2    2
2    3

These data would suggest that Blocks are unique (and therefore
necessarily nested within Sites):
SITE  BLOCK
1    1
1    2
1    3
2    4
2    5
2    6

Hank

On Thu, May 21, 2009 at 5:05 PM, Rolf Turner <r.turner at auckland.ac.nz> wrote:
>
> On 21/05/2009, at 3:12 PM, Ben Bolker wrote:
>
>        <snip>
>
>>   The only thing I would check for is that your BLOCK numbers
>> are truly "nested" within SITE, i.e. that your blocks are numbered
>> 1..n within each site, not 1:(n*N) (where n = # blocks per site,
>> N = # of sites).
>
>        <snip>
>
> I was under the impression that lmer() required that nested effects
> have *distinct* subscripts for their levels, e.g. block 1 in site 1
> should *not* have the same index as block 1 in site 2 --- they being
> after all, different blocks.  I thought that this was one of Doug
> Bates' particular pet peeves about the (irrational) way that other
> packages (e.g. The-Package-That-Must-Not-Be-Named) do things.
>
> Have I been suffering from a misapprehension?  Wouldn't be the first time
> ....
>
>        cheers,
>
>                Rolf Turner
>
> ######################################################################
> Attention:\ This e-mail message is privileged and confid...{{dropped:9}}
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>



-- 
Hank Stevens
http://www.cas.muohio.edu/~stevenmh/
513-529-4206
E pluribus unum