Skip to content

cdermont/r_cjoint_cd

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 

Repository files navigation

cjoint_cd

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)".

About

custom functions based on r::cjoint

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages