0

I am able to use lmer on an older computer without this error, but the new version (downloaded yesterday) gives me this error message:

Error in [[<-.data.frame(*tmp*, i, value = integer(0)) : replacement has 0 rows, data has 117

in response to this function:

model1<-lmer(threatened~as.numeric(focal.chel)+as.numeric(opponent.chel)+(1|as.factor(males)),na.action=na.omit,data=spiders)

I would really like to be able to use the newer version, as I've read that it has less problems with false convergence than older versions. Any help anyone can offer as to how to get this to run without the above error would be greatly appreciated.

My data can be found here: https://dl.dropboxusercontent.com/u/16881915/Rfile.txt

apomene
  • 14,282
  • 9
  • 46
  • 72
user3060658
  • 1
  • Looking at the News page for the package - there's reference to a bug on index variables `+(1|as.factor(males))` so it could is likely that the behaviour has changed for this section of your code. How you go about resolving it, I'm not sure and hopefully someone else will be able to answer. Knowing the problem area may help in the interim though. http://cran.r-project.org/web/packages/lme4/news.html – Steph Locke Dec 03 '13 at 09:54
  • by the way, when asking questions of this sort it's extremely useful to specify the versions you're using (`sessionInfo()` or `packageVersion("lme4")`) on both computers – Ben Bolker Dec 03 '13 at 12:37
  • Also see http://stackoverflow.com/questions/19749313/lme4-upgrade-produces-error-message-even-after-grouping-variables-within-the-dat/20353535#20353535 – Ben Bolker Dec 03 '13 at 14:20
  • and cross-posted discussion at r-help: http://article.gmane.org/gmane.comp.lang.r.general/303372 – Ben Bolker Dec 03 '13 at 16:34

0 Answers0