[R-sig-ME] modeling fixed effects with different spatial scales in glmm using nlme

Sharada Ramadass @h@r@d@@r@m@d@@@ @ending from gm@il@com
Fri May 25 13:28:17 CEST 2018

  This is my first time working with glmm and the associated software in R
and I am using nlme to model my research data. I have multiple queries
relating to the model design and results interpretation and will post them
as separate questions for the sake of clarity.

Here is my first query:

I am looking at predictors of growth (response), and have fixed effects at
different spatial scales - individual level measurement (for an organism)
called I, and measurements at two different spatial scales for abiotic (A)
and biotic (B) factors.

Do I need to do anything different to incorporate the different scales of
the fixed effects or could I use them in a normal glmm model, e.g.
(this is just a notation and not the exact syntax)
lme(growth ~ I + A + B +<random effects>)

In some ways it mimics a split plot design which has explanatory variables
at different spatial scales, though this is not an experimental setup but
field data.

Any inputs will be appreciated.
Thanks and Regards,

	[[alternative HTML version deleted]]

More information about the R-sig-mixed-models mailing list