lmerTest (version 2.0-3)

lsmeans: Calculates Least Squares Means and Confidence Intervals for the factors of a fixed part of mixed effects model of lmer object.

Description

Produces a data frame which resembles to what SAS software gives in proc mixed statement. The approximation of degrees of freedom is Satterthwate's.

Usage

lsmeans(model, test.effs=NULL, method.grad="simple",...)

Arguments

model

linear mixed effects model (lmer object).

test.effs

charachter vector specyfying the names of terms to be tested. If NULL all the terms are tested.

method.grad

approximation method for the grad function, which is used in calculation of denominator degrees of freedom. Could be "simple" or "Richardson". "simple" is the default one (and the faster one).

...

other potential arguments.

Value

• Produces Least Squares Means (population means) table with p-values and Confidence intervals.

References

doBy package, gplots package

See Also

step, rand, difflsmeans

Examples

#import lme4 package and lmerTest package
library(lmerTest)

#specify lmer model
m1 <- lmer(Informed.liking ~ Gender*Information +(1|Consumer), data=ham)

#calculate least squares means for interaction Gender:Information
lsmeans(m1, test.effs="Gender:Information")

#import TVbo data from lmerTest package
data(TVbo) 

m <- lmer(Coloursaturation ~ TVset*Picture + (1|Assessor), data=TVbo)
plot(lsmeans(m))
lsmeans(m, test.effs="TVset")