Forum: open-discussion
Monitor Forum | Start New Threadlogit mixed models using lme4 [ Reply ] By: Ruben van Eupen on 2010-06-09 21:49 | [forum:2968] |
hello I have some questions about fitting logit mixed models using the package lme4. Fitting the following model using R package lme4: final_model = lmer(correct ~ TMD + (1 | subject), data_temp, family = binomial, REML = FALSE) summary(final_model) How is it possible to test the significance of your random effect? the anova (model 1, model2) does not work, at least not if you take glm() to fit your other model without the random effect... . Just compare AIC's and Deviances between models? And is it possible to get your residual variances, the so called estimate for the variance of eta? Becausse then you can compare how much variance if taken by your random effect compared to the residual variance... thx ruben |