[R-sig-ME] lmertest - pairwise comparison p-value correction

[Phillip Alday]

Ever lmerTest model is also a valid lme4 model (class merModLmerTest extends class lmerMod), so that shouldn't be a problem. You can also force the issue with something like:

lsmeans.command( as("lmerMod",lmertest.model))

Or going the other way:

summary(as("merModLmerTest",lme4.model),ddf="Satterthwaite")

Both examples assume that you have the necessary packages loaded.

Best,
Phillip 

PS: Keep the list in CC, please. I often forget to do this myself, so feel free to remind me when I forget ...


> On 17 Oct 2016, at 15:18, S Ellison <S.Ellison at lgcgroup.com> wrote:
> 
> Thanks for the suggestion, but I was also looking for the dreaded p-values for lmer models.  Hence lmerTest. That does not play nicely with lsmeans, so I was hoping to stick to one package.
> 
> But maybe not ...
> 
> Steve E
> 
> 
> ________________________________________
> From: Phillip Alday [Phillip.Alday at unisa.edu.au]
> Sent: 17 October 2016 03:42
> To: S Ellison
> Cc: r-sig-mixed-models at r-project.org
> Subject: Re: [R-sig-ME] lmertest - pairwise comparison p-value correction
> 
> Without taking the time to look that up, you could also just use the lsmeans package, which has several different options for multiple-comparisons corrections and also uses the least-square means approach.
> 
> Best,
> Phillip
> 
> 
>> On 17 Oct 2016, at 10:09, S Ellison <S.Ellison at lgcgroup.com> wrote:
>> 
>> A hopefully quick question: Can anyone point me to the p-value correction method used for multiple pairwise testing - if any - by difflsmeans() in the lmerTest package?
>> 
>> It's not obvious from the package documentation and the code I've scanned doesn't seem to include a correction.
>> 
>> 
>> S Ellison
>> 
>> 
>> 
>> *******************************************************************
>> This email and any attachments are confidential. Any use...{{dropped:8}}
>> 
>> _______________________________________________
>> R-sig-mixed-models at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
> 
> 
> 
> *******************************************************************
> This email and any attachments are confidential. Any u...{{dropped:8}}