fitlmer: Run mixed-effects model using ('lmer')

In dalejbarr/simgen: Monte Carlo Simulations of Simultaneous Generalization

Description

Runs either a random-intercepts only model or a maximal random-effects model (by-subject and by-item random intercepts, by-subject random slope, and by-item random slope for within-item design); NB: this function is for one-factor design only.

Usage

fitlmer(mcr.data, ri.only = FALSE, wsbi = FALSE)

Arguments

mcr.data	A dataframe formatted as described in mkDf.
ri.only	Whether the random effects specification is to be random-intercepts only (TRUE) or maximal random-effects (FALSE).
wsbi	Whether the design is between-items (TRUE) or within-items (FALSE).

Details

fitlmer will attempt to fit the model specified by the user, and will progressively simplify the model as needed to get it to converge. If no model converges, it returns NAs. fitlmer performs a likelihood ratio test for the treatment effect, as well as returns a p-value for the t-statistic using an approximation from the normal distribution.

Value

A vector with elements:

fm	Code for the model that converged: (1) dropped one slope; (2) dropped two slopes; (3) main model did not converge; (4) comparision model for likelihood ratio test did not converge.
t	t-statistic for the treatment effect
chi	chi-square statistic for the likelihood ratio test (1 df)
pt	p-value for the t-statistic (normal distribution)
pchi	p-value for the chi-square statistic

See Also

mkDf, fitlmer.mcmc

Examples

nmc <- 10
pmx <- cbind(randParams(genParamRanges(), nmc, 1001), seed=mkSeeds(nmc, 1001))

# between-items dataset
x.bi <- mkDf(nsubj=24, nitem=24, mcr.params=pmx[1,], wsbi=TRUE)

# within-items dataset
x.wi <- mkDf(nsubj=24, nitem=24, mcr.params=pmx[1,], wsbi=FALSE)

# maximal model
fitlmer(x.bi, wsbi=TRUE)

fitlmer(x.wi, wsbi=FALSE, ri.only=FALSE) # maximal
fitlmer(x.wi, wsbi=FALSE, ri.only=TRUE) # random intercepts only

dalejbarr/simgen documentation built on May 14, 2019, 3:32 p.m.