[R-sig-ME] KRmodcomp issues

[Henrik Singmann]

Hi Kevin,

Your problem will most likely disappear if you install the development version of pbkrtest (v 0.3-7) from here:
http://people.math.aau.dk/~sorenh/software/pbkrtest/devel/

If not, you have to contact the maintainer of pbkrtest as described on their website with a reproducibel example. See under section "Reporting unexpected behaviours (bugs)" here:
http://people.math.aau.dk/~sorenh/software/pbkrtest/

I hope this helps,
Henrik


Am 05.11.2013 02:39, schrieb Kevin Gould:
> Hi all!  This is my first time posting to the forum.  I am, admittedly, an
> R novice (and statistical almost-novice) with little experience with mixed
> models analysis.  However, I was referred here by someone who is an R whiz,
> if that counts for anything.  Today I was attempting to use KRmodcomp and
> got the following (rather confusing) output:
>
> F-test with Kenward-Roger approximation; computing time: 2.10 sec.
> large : Time ~ Grammar + Syllables + Order + (Grammar + Syllables + Order |
>      Subject)
> small : Time ~ Syllables + Order + (Grammar + Syllables + Order | Subject)
>           stat     ndf     ddf F.scaling p.value
> Ftest -25.141   1.000  39.975         1       1
>
> Any idea what could be causing this?  Here's the summary for ModelA, if
> that helps:
>
> Linear mixed model fit by REML ['lmerMod']
> Formula: Time ~ Grammar + Syllables + Order + (Grammar + Syllables + Order
> |      Subject)
>     Data: data1
>
> REML criterion at convergence: 317.5946
>
> Random effects:
>   Groups   Name              Variance Std.Dev. Corr
>   Subject  (Intercept)       0.021865 0.14787
>            Grammarspeed_slow 0.004639 0.06811  -0.78
>            Syllables         0.000408 0.02020   0.10 -0.69
>            OrderB            0.061705 0.24841  -0.58  0.66 -0.30
>   Residual                   0.085452 0.29232
> Number of obs: 571, groups: Subject, 41
>
> Fixed effects:
>                     Estimate Std. Error t value
> (Intercept)        1.510485   0.122052  12.376
> Grammarspeed_slow -0.123387   0.027608  -4.469
> Syllables          0.033882   0.016742   2.024
> OrderB            -0.003791   0.064584  -0.059
>
> Correlation of Fixed Effects:
>              (Intr) Grmmr_ Syllbl
> Grmmrspd_sl  0.096
> Syllables   -0.923 -0.302
> OrderB      -0.209  0.152 -0.034
>
> The t-value of -4.47 seems promising, so any help you could give me would
> be greatly appreciated!
>
> Again, I'm a novice.  If there's any other information that you require
> from me here, please don't hesitate to ask!
>
> Thanks very much!
>
> -Kevin
>
> 	[[alternative HTML version deleted]]
>

-- 
Dipl. Psych. Henrik Singmann
PhD Student
Albert-Ludwigs-Universität Freiburg, Germany
http://www.psychologie.uni-freiburg.de/Members/singmann