[R-sig-ME] Using R, how to present mixed models vs. regular linear regression models?

[Ben Bolker]

Michael <comtech.usa at ...> writes:

> 
> And what's the difference between these two models?
> 
> M0 <- lmer (y ~ 1 + (1 | county))
> 
> M1 <- lmer (y ~ -1 + (1 | county))
> 
> On Thu, May 31, 2012 at 9:52 AM, Michael <comtech.usa at ...> wrote:
> 
> > What are the key differences between the following two models?
> >
> >
> >
> > lmefit = lmer(MathAch ~ SES + (1 |School) , MathScores)
> >
> > lmfit = lm(MathAch ~ SES + School -1 , MathScores)
> >
> > To me, they seem to be the same, except that lmefit takes less parameters
> > (because it used Normal distribution to model the levels at the group
> > level...)
> >

  If you don't get a satisfactory answer here you might try on 
http://stats.stackexchange.com ; this seems like a very well-suited
question for that site.

  Your second question is easier because it only involves the fixed
effects; the second model doesn't use an intercept, and so assumes that
the average y across all counties is exactly zero (this is a pretty
weird model; most of the time it doesn't make sense to include a random
effect whose corresponding population-level effect is missing/forced to zero).

  In the second model in your first question I would be careful;
I don't think the -1 does what you think it does/want it to do
(i.e., set the average effect of school to zero).
  
  Ben Bolker