Forum: help

RE: Multiple random effects [ Reply ] By: Chris Street on 2014-05-11 12:19	[forum:41031]
Hi Richard, I hope you're well. Sorry for the delay in responding, the email notification ended up in my junk mail. Thanks so much for your help, that's very clear =] Best C

RE: Multiple random effects [ Reply ]
By: Richard Morey on 2014-05-10 23:23

[forum:41018]

Hi Chris,

You need to simply concatenate the names of the random effects with R's c() function:

whichRandom = c("RandEff1", "RandEff2").

As for neverExclude, you probably want the same thing:

neverExclude = c("RandEff1", "RandEff2"),

assuming that these are nuisance effects. If you want random slopes, you enter them in as an interaction with the corresponding covariate. Just entering "+ RandEff1" will yield a random intercept only. If "b" is a covariate, you can get a random slope and intercept model with:

"DV ~ b + b:RandEff1 + RandEff1"

Multiple random effects [ Reply ]
By: Chris Street on 2014-05-08 00:44

[forum:41017]

Hi,

I am looking to enter random effects for my participants (level 2) as well as my stimuli (level 1) to model for individual and stimulus differences. What is the syntax when using generalTestBF? I imagine something like this:

generalTestBF(DV ~ IVs + RandEff1 + RandEff2, data, whichRandom = {"RandEff1", "RandEff2"}, neverExclude)

I am also wondering whether the random effect creates random slopes for all the variables entered as fixed effects, or does the random effect only model an intercept?

Thanks!
Chris