I'm using R2.8.1 and trying to do some nested modelling using lmer and weights. When I multiply/divide weights with a constant, variance of groups intercept changes, but variance of the residual doesn't (using unconditional model). Shouldn't they change equally? Are there any specifications on how to set weights properly? Thanks for reply, ___________________ Gasper Cankar. PhD researcher National Examinations Centre (Ric) gasper.cankar at ric.si tel. +386 1 54 84 682 fax. +386 1 54 84 601