Plotting residuals for GLMER model and zero counts
Dear R-ers, although a number of options exist out there to plot logit models, I can't seem to find one that works for glmer. What I would like to do is, in light of having found no interaction between the two fixed effects (which answers my research question), look at a plot to tell how well the model fits the data. The package LMERconveniencefunctions ver 2.0 has a nice plot for lmer but this won't work on my model because R automatically asks me to use glmer. It has a factor DV (Correct), two fixed factor IVs (Group and Syntax), and two random effect (ID and item). Here's the output of calling: model<- glmer(Correct ~ Group * Syntax + (Syntax + 1 | ID) + (Group + 1 | item), data=...., family=binomial) #Generalized linear mixed model fit by maximum likelihood ['glmerMod'] Family: binomial ( logit ) Formula: Correct ~ Group * Syntax + (Syntax + 1 | ID) + (Group + 1 | item) Data: ..... AIC BIC logLik deviance 235.4060 276.5207 -107.7030 215.4060 Random effects: Groups Name Variance Std.Dev. Corr ID (Intercept) 44.919 6.702 Syntaxof 45.850 6.771 -0.99 item (Intercept) 6.282 2.506 Groupns 38.105 6.173 -0.63 Number of obs: 451, groups: ID, 38; item, 16 Fixed effects: Estimate Std. Error z value Pr(>|z|) (Intercept) -9.236 2.192 -4.214 2.51e-05 Groupns -2.815 4.743 -0.594 0.5528 Syntax of 5.546 2.458 2.256 0.0241 Groupns:Syntax of -1.457 6.405 -0.228 0.8200 (Intercept) *** Groupns Syntaxof * Groupns:Syntaxof --- Signif. codes: 0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.? 0.1 ? ? 1 Correlation of Fixed Effects: (Intr) Gropns Syntxf Groupns -0.499 Syntaxof -0.892 0.445 Grpns:Syntx 0.370 -0.741 -0.443 I have tried messing around with Mosaic but here I keep running into the error
Error: length(dim(observed)) == 2 is not TRUE
which I suspect is due to one cell in the 2x2x2 crosstab of results below being lower than 2 (please correct me here if I'm mistaken): Group Syntax Cor InC nns 's 118 14 nns of 128 19 ns 's 62 6 ns of 53 1 I've tried using xtabs and formulae as the x argument for the mosaic function but neither seems to work. On a side note, how do I cope with cases where participants produced 0 counts to a cell? I understand Chi-square does not cope well with these. Should I leave them as is, or turn these into NAs, or eliminate as outliers? Thank you in advance for reading. Francesco