some custom functions based on r::cjoint
Changes:
plot.amce2
based on the original plot.amce
, with the small change that print(p)
is the last argument (similar to older versions), which allows to save the plot as an R object and reuse the data within. Combine results from multiple models, control colors and facets more explicitly. You'll find the data in object$plot$data
.
use.qualtrics
introduced based on read.qualtrics
, which takes an R object as argument to reshape data into cjoint long format. Recode your data before transforming it to cjoint format, or use the same function and setup to transform data not from Qualtrics.
amce.lmer
which applies lme4::lmer
instead of clustered OLS. LMER results are stored in the object with the results unter object$lmer.prof
and object$lmer.full
, everything else is similar to amce. Get random effects per individual instead of clustered errors through data.frame(ranef(object$lmer.prof)$respondentid)
, for example. Define hierarchy as in lmer, e.g., hierarchy = "(1 | respondentid)"
.