[R-sig-ME] Error in sprintf

[Alexej Siren]

Hello Ben,

Thanks for the reference.  I checked that thread and others before I sent
the message to the list.  However, MuMIn works fine; I was not able to get a
summary of parameter estimates when not model averaging.

I found a workaround to the problem by removing the subset command and
creating separate csv files.  I haven't taken the time to investigate why
the subset syntax isn't working but I will do that soon as I often subset
data.  I've included the formula below in case you're interested.   

fm1 <- glmer(ua ~ X.regen*X.hardwood + (1|pinky), data_std,
family=binomial(logit), subset=(season=="lof"))

Thanks for your response and I'll post the solution when I find the problem.

Alexej


-----Original Message-----
From: r-sig-mixed-models-bounces at r-project.org
[mailto:r-sig-mixed-models-bounces at r-project.org] On Behalf Of Ben Bolker
Sent: Wednesday, February 26, 2014 10:24 PM
To: r-sig-mixed-models at r-project.org
Subject: Re: [R-sig-ME] Error in sprintf

Ben Bolker <bbolker at ...> writes:

> 
> 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'
> > 
> > 

PS this may have been answered earlier on this list --

http://r.789695.n4.nabble.com/Error-running-MuMIn-dredge-function-
using-glmer-models-td4679991.html
[WARNING, broken URL to make gmane happy]

(I found this by googling "error in sprintf gettext symbol" -- this message
is hits #2 and #3)

The link suggests you can work around the problem by (1) installing MuMIn
version 1.9.14 from R-forge (2) omitting na.action=na.omit.

Your problem *might* be something different, but this looks like a good
place to start.

  Ben Bolker

_______________________________________________
R-sig-mixed-models at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models