[R-sig-ME] degree of freedom is non-integer from lmerTest

[Jake Westfall]

Yuqi,

By default, the DFs in lmerTest are based on the Satterthwaite
approximation:
https://en.wikipedia.org/wiki/Welch%E2%80%93Satterthwaite_equation

Also see ?lmerTest::summary.merModLmerTest

Jake

On Thu, Oct 5, 2017 at 4:59 PM, Yuqi Liu <yliu at psych.udel.edu> wrote:

> Dear R-experts,
>
>   I am using lmerTest to do mixed model analysis on one of my experiments.
> There are two factors, each with 2 levels, resulting in 4 experiment
> conditions.  It was repeated measures design, and each experiment condition
> was conducted once for each individual. The model is thus very simple, with
> each factor as a fixed effect and subjects as a random effect: model <-
> lmerTest (DV~IV1*IV2+(1|subject),data).
>
>  When I looked at the result using summary(model), the Dfs are
> non-integers, is it because we had to exclude some of the trials, so that
> some participants had missing data? In general, I am wondering how t-tests
> are performed in lmerTest, and how degree of freedom was calculated.
>
> Thank you for your help!
> Yuqi
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>

	[[alternative HTML version deleted]]