Skip to content
Prev 3533 / 20628 Next

Very weird lmer results, compared to SAS proc mix

Can you post the summaries of these models, so we can see if there
might be something else odd going on?  (Making the data, or an otherwise
reproducible example, available would be even better ...)
   It might also be useful to post the parameter estimates from PROC
MIXED.  Maybe (it's a long shot) PROC MIXED automatically includes the
fixed effect of age (which would be highly correlated with exposure,
which might explain the non-significance of exposure)?
Yong Wu wrote: