[R-sig-ME] R package ‘lme4’

[Ben Bolker]

On 14-01-21 05:36 AM, Lorenzo Raggi wrote:
> Dear package curator,
> 
> first of all I apologize if I write to you directly.

 I'm taking the liberty of cc'ing this to the r-sig-mixed-models mailing
list; it's often better to send to the list where a wider range of
people can answer, and where others can read the answers as well.

> I did find some problem in using the lme4 package.
> In particular, using the scrip (TBRT2011Script.txt) on example data
> TBRTQuality.csv from a webinar available at:
> 
> http://www.extension.org/pages/61006/estimating-heritability-and-blups-for-traits-using-tomato-phenotypic-data#.Ut5B2LTSJpi

Just want to point out the note on that page:

IMPORTANT NOTE: The code provided has been tested for 2.x of R. There
are issues with the code for version 3.0.
> 
> when I use the lmer function I get the following error:
> 
> Error in checkNlevels(reTrms$flist, n = n, control) :
>   grouping factors must have > 1 sampled level
> 
> When I tried to find some hints in in how solve the problem in the
> manual for Package ‘lme4’ at page 33, third line, it looks that there is
> a printing problem regarding the "option(...)" command description that
> is interrupted. This was a bit frustrating.

  I see that line is too long, but (1) you can see it on your own
computer, better formatted, by typing ?lmerControl in R (when the lme4
package is loaded); (2) the more relevant point is on the previous page:
you want  control=lmerControl(check.nlev.gtr.1="ignore") (and presumably
control=lmerControl(check.nlev.gtr.1="ignore",check.nlev.gtr.5="ignore"), for
completeness.

> 
> Sorry again for bothering you and
> 
> looking forward to hear fro you
> 
> best regards
> 
> Lorenzo
> 
> 
> 
> -- 
> Lorenzo Raggi, Ph.D.
> University of Perugia
> Dept. of Applied Biology
> 
> Borgo XX Giugno, 74
> 06121 Perugia, Italy
> tel.  +39 0755856212
> fax  +39 0755856224