[R] standardized beta weights for lmer models

Annett Jorschick annett.jorschick at uni-bielefeld.de
Wed Dec 2 16:31:43 CET 2009


Dear Sir or Madam,

I am using the lme4 package to model the effect of various variables on binomial response data in a repeated measurements design. Besides the significance of the main predictors, we are also interested in the size and direction of the (partial) correlation of the predictors with the dependent variable.  Therefore, we wonder whether the estimates listed under "coefficients" in the model summary are standardized or not. Could you inform us about this? And if they are not standardized, is there a way (or package) to calculate the standardized beta-weights and/or the partial correlations belonging to the predictor variables in logit lmer models? 

Thank you very much for your help.

Best regards,
Annett




More information about the R-help mailing list