[R-sig-ME] Error in sprintf

Ben Bolker bbolker at gmail.com
Wed Feb 26 21:33:18 CET 2014


On 14-02-25 10:33 PM, Alexej Siren wrote:
> Hello,
> 
>  
> 
> I'm reanalyzing binomial GLMM's using the same script and data from last
> year.  I'm using the lme4 and AICcmodavg packages and able to produce an AIC
> table but not able to get a summary of the top model - or any model.
> Instead, I get the error listed below.  
> 
>  
> 
> Error in sprintf(gettext(fmt, domain = domain), ...) : 
> 
>   invalid type of argument[1]: 'symbol'
> 
>  
> 
> However, if I use the MuMIn package I can get a table and a summary of the
> model-averaged coefficients.
> 



> 
> Since I analyzed this data I've installed the newest version of R and have
> updated all packages.  I've checked the data and script which have not been
> modified since last summer when it worked.  Further, I have searched through
> various forums to learn more about this error but I've found scant
> information that relates to my issue.  Are there changes in the newer
> version of R or in a package that would elicit this error?  Perhaps the
> structure of my data needs to be changed?  Any suggestions would be
> welcomed.  
> 
>  
> 
  There have been a lot of changes to R, but with this amount of
information it's almost impossible to say anything -- is there any
chance you could provide (1) a reproducible example (see
http://tinyurl.com/reproducible-000 for hints) (2) the results of
sessionInfo() (to tell us what versions of packages you're using) (3)
the results of "traceback()" after you get the error?  I can't find any
instance of "sprintf(gettext" in the lme4 package code, but that only
suggests (not guarantees) that the proximal problem is in one of the
other packages you're using.

  Ben Bolker



More information about the R-sig-mixed-models mailing list