`se()` not working for merMod with specific random parts

[sjPlot]

I am trying to plot the joint random and fixed effects coefficients for each level of each grouping variable (actually, I am only interested in one grouping variable). I have three grouping vars with 948, 8, and 6 levels. Calling the function produces only one plot (for the grouping var with 948 levels) and the following errors and warnings:

sjp.lmer(mlm3, type="coef") Error in [.data.frame(se.fit, , i) : undefined columns selected

In addition: Warning messages: 1: In sweep(cmode.vars, 2, fixed.vars, "+") : STATS does not recycle exactly across MARGIN 2: In sweep(cmode.vars, 2, fixed.vars, "+") : STATS is longer than the extent of 'dim(x)[MARGIN]' 3: In sweep(cmode.vars, 2, fixed.vars, "+") : STATS does not recycle exactly across MARGIN 4: In sweep(cmode.vars, 2, fixed.vars, "+") : STATS does not recycle exactly across MARGIN 5: In sweep(cmode.vars, 2, fixed.vars, "+") : STATS is longer than the extent of 'dim(x)[MARGIN]' 6: In sweep(cmode.vars, 2, fixed.vars, "+") : STATS does not recycle exactly across MARGIN

(see data_and_model.RData)

[favstats]

So this isn't resolved yet, is it?

[strengejacke]

No, not yet. I think, setting show.ci = FALSE should work, but of course, w/o confidence intervals then. I had not yet found the time to investigate this issue further.

[GitDro]

Is this related to se() not working on icc objects? Because I find that se(icc(fit)) returns an NA even on the example vignette with sjstats 0.17.1, R version 3.5.1

[strengejacke]

Is this related to se() not working on icc objects? Because I find that se(icc(fit)) returns an NA even on the example vignette with sjstats 0.17.1, R version 3.5.1

No, but this issue should be fixed with https://github.com/strengejacke/sjstats/commit/010310f0b6c19c080353d7732292ab7ce4507ddd