using multcomp to update glmm emmeans

[Joseph M. Orr]

In the vignette, “Mixed Model Reanalysis of RT data” multcomp is used for a emmeans object,
summary(as.glht(update(pairs(emm_i1), by = NULL)), test = adjusted("free"))

I ran a glmm using mixed and generated an emm object
emm_i3 <- emmeans(vsr_rew_m4lrt, "rewCond", by = c("prevRew","version")) but when I try to use the above example I get an error:

summary(as.glht(update(pairs(emm_i3), by = NULL)), test = adjusted(“free”))
Note: adjust = “tukey” was changed to “sidak”
because “tukey” is only appropriate for one set of pairwise comparisons
Error in UseMethod(“vcov”) :
no applicable method for ‘vcov’ applied to an object of class “c(’emmwrap’, ‘list’)”
Error in h(simpleError(msg, call)) :
error in evaluating the argument ‘object’ in selecting a method for function ‘summary’: no ‘vcov’ method for ‘model’ found!

Do I need to provide some additional options to glht to get it to work for a glmm?

[henrik]

Solving specific problems without a reproducible example is generally difficult. At least for the data oin the vignette, it still works:

library("afex") # needed for mixed() and attaches lme4 automatically.
library("emmeans") # emmeans is needed for follow-up tests 
library("multcomp") # for advanced control for multiple testing/Type 1 errors.
data("fhch2010") # load 
fhch <- droplevels(fhch2010[ fhch2010$correct,]) # remove errors
str(fhch2010) # structure of the data

m9s <- mixed(log_rt ~ task*stimulus*density*frequency + 
               (stimulus+frequency||id)+
               (task|item), fhch, 
             control = lmerControl(optCtrl = list(maxfun = 1e6)), expand_re = TRUE)
emm_options(lmer.df = "asymptotic")
emm_i1 <- emmeans(m9s, "frequency", by = c("stimulus", "task"))
update(pairs(emm_i1), by = NULL, adjust = "holm")
summary(as.glht(update(pairs(emm_i1), by = NULL)), test = adjusted("free"))

This produces:

Note: adjust = "tukey" was changed to "sidak"
because "tukey" is only appropriate for one set of pairwise comparisons

	 Simultaneous Tests for General Linear Hypotheses

Linear Hypotheses:
                                 Estimate Std. Error z value Pr(>|z|)    
low - high, word, naming == 0     0.05877    0.01491   3.941 0.000162 ***
low - high, nonword, naming == 0 -0.20699    0.01497 -13.823  < 2e-16 ***
low - high, word, lexdec == 0     0.06353    0.01669   3.807 0.000162 ***
low - high, nonword, lexdec == 0  0.11086    0.01675   6.619 1.08e-10 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Adjusted p values reported -- free method)

[Author]

Posts