[R-sig-ME] New version of afex on CRAN now

Henrik Singmann singmann at psychologie.uzh.ch
Sat Apr 15 01:50:20 CEST 2017

Dear list,

After one year without updates, a new version (0.17-8) of afex (Analysis 
of Factorial Experiments with ANOVA and Mixed Models) has finally been 
accepted on CRAN yesterday:

This new version comes with a series of new features, mostly related to 
mixed models, that might be of interest to some. The full list of 
changes can be found here:

The major news are:
- For LMMs, mixed() now supports the Satterthwaite approximation for 
obtaining p-values (method="S"), via lmerTest.

- mixed() methods "KR" and "S" are now obtained via tests on the full 
model, which can be considerably faster.

- Added a new vignettes containing an extensive mixed model example 

- All arguments and slots with "." in the name have been renamed to use 
"_". This may lead to warnings in existing code.

- Added argument all_fit to mixed(). If TRUE, each model (i.e., full and 
restricted) is fitted with all available optimizers and the best fit in 
each case selected. Can be quite helpful for GLMMs with convergence 
problems and method="LRT".

- All returned objects ("mixed", "afex_aov", and "nice") have a pretty 
print() method. (Thanks to Frederik Aust)

Given that I occasionally receive questions related to afex and the 
existing mailing list can be intimidating I have decided to establish a 
support forum at: http://afex.singmann.science/
Feel free to share the address with students or other interested 
parties. As some questions seem to repeat, I hope this can become a 
valuable resource.


Dr. Henrik Singmann
Universität Zürich, Schweiz

More information about the R-sig-mixed-models mailing list