Hi João, That makes sense! I tried what Shadiya suggested but I just got 1 as the degree of freedom for all of my factors.
Without degrees of freedom, is there a way to compute effect size? Thank you, Simritpal On Sep 22, 2016 9:30 AM, "João Veríssimo" <jl.veriss...@gmail.com> wrote: > Hi Simritpal, > > The glmer output gives you Z-tests for each estimate. So far as I know, > these are not associated with degrees of freedom. > > Best, > João > > On Wed, 2016-09-21 at 00:51 -0400, Simritpal Malhi wrote: > > Thanks again Dr. Kingston! Another question - how would you obtain or > > calculate degrees of freedom for glmer as it is not provided in the > > output? > > > > > > On Sun, Sep 18, 2016 at 1:54 PM, John Kingston > > <jkings...@linguist.umass.edu> wrote: > > Dear Simirtpal, > > > > You need to use "glmer" not "lmer" to run such a model. > > > > Best, > > John > > > > > > On 2016-09-18 13:41, Simritpal Malhi wrote: > > Hello, > > > > I am trying to run my error analysis using a logit > > mixed model. I have > > installed the lmer test package because I would like > > to report > > p-values. I have attached a sample of how my data file > > is set-up. > > > > e$correct <- as.factor(e$correct) > > > > erranalysis=lmer(correct~task*concrete > > +task*embodied+ > > > > task*symbolic+(1|item)+(1|subject),data=e, > > > > family=binomial) > > > > I put the dependent variable as "correct" and I added > > "family=binomial" to specify that the "correct" > > dependent variable was > > coded as 1 if the participant had an error and 0 if > > the participant > > was correct. However, I am getting the following error > > message: > > > > Error in as(model, "merModLmerTest") : no method or > > > > default for coercing “glmerMod” to “merModLmerTes” > > > > In addition: Warning messages: > > > > 1: In lme4::lmer(formula = correct ~ task * concrete > > + > > > > task * embodied + : calling lmer with 'family' is > > > > deprecated; please use glmer() instead > > > > 2: In checkConv(attr(opt, "derivs"), opt$par, ctrl = > > > > control$checkConv, : Model failed to converge with > > > > max|grad| = 0.00205133 (tol= 0.001, component 1) > > > > Any help would be appreciated. Thank you for your > > time. > > > > > > > > > >
