[R] lmer (LME4) and survey standard error

[Ben Bolker]

David Epstein <davideps <at> umich.edu> writes:

[snip]

> I am using lmer (LME4) to build a model from data for 19 different
> neighborhoods drawn, in part, from the American Communities Survey
> (ACS). The ACS data is static while other variables change over the
> five years under investigation. I am new to mixed effects models and
> was hoping that someone could suggest a way to include the ACS
> standard error for each covariate into the model. I currently call
> lmer as follows:
> 
> #Y = dependent variable (numeric, vector)
> #G = neighborhood names (string, vector)
> #X = independent variables (numeric, matrix)
> 
> m = lmer(Y ~ (1|G) + X)
> 

  [snip]

  I suggest (1) taking a look at the weights= argument to
see if it does what you need
and (2) asking this question on the r-sig-mixed-models list,
where there may be more expertise on this subject.

  Ben Bolker