[R-sig-ME] significance level in lme4 and lmerTest

[Ben Bolker]

    These packages (lme4, lmerTest, car) follow the general R convention 
of reporting p-values but not specifying an accept/reject threshold for 
you.  When you say "the effects were significant", it is *you* making 
the decision about what threshold to use (unless you're using some other 
package, such as 'report', which might apply a cut-off for you). Either 
you or some other code you used has also done the rounding of p-values 
(to <.001, <.0001); for very small values (<2e-16) R doesn't show the 
exact p-value by default, but otherwise it does.

   (Since you are reporting an effect with a p-value of slightly greater 
than 0.01 as significant, *presumably* you are using an alpha/rejection 
threshold of 0.05, but it's not R doing it ...)

   If you can give us more detail about what you did, we might be able 
to help further, but that's all we (or at least I) can say for now.

On 7/5/22 8:14 AM, Hasibe Kahraman wrote:
> Hello,
> 
> I used the lme4 package for modelling my data and ANOVA Type III (Version 3.0-12; Fox & Weisberg, 2019). I reported the results of LME analyses in my manuscript (see below):
> The main effects of group and prime type were significant (F(1,139.02)=6.5005, p=.011; F(2,197.66)=8.5271, p<.001, respectively), with a robust two-way interaction between Group and Prime Type (F(2,190.45)=11.9593, p<.0001)
> 
> 
> However, one reviewer asked that I should specify which p-value is my "significance" level. Therefore, could I possibly learn whether it is set to .05 or .01?
> 
> Thanks,
> Hasibe
> 
> 	[[alternative HTML version deleted]]
> 
> _______________________________________________
> R-sig-mixed-models using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models

-- 
Dr. Benjamin Bolker
Professor, Mathematics & Statistics and Biology, McMaster University
Director, School of Computational Science and Engineering
Graduate chair, Mathematics & Statistics