
Computes the LS-means for the final backward reduced model and passes these
to plot.ls_means
.
# S3 method for step_list
plot(x, y = NULL, which = NULL, pairwise = FALSE,
mult = TRUE, level = 0.95, ddf = c("Satterthwaite",
"Kenward-Roger"), ...)
a step_list
object; the result of running
step
.
not used and ignored with a warning.
optional character vector naming factors for which LS-means should
be plotted. If NULL
(default) plots for all LS-means are generated.
pairwise differences of LS-means?
if TRUE
and there is more than one term for which to plot
LS-means the plots are organized in panels with facet_wrap
.
confidence level.
denominator degree of freedom method.
currently not used.
Error bars are confidence intervals - the default is 95 level can be changed.
# NOT RUN {
# }
# NOT RUN {
# Fit example model:
tv <- lmer(Sharpnessofmovement ~ TVset * Picture +
(1 | Assessor:TVset) + (1 | Assessor:Picture) +
(1 | Assessor:Picture:TVset) + (1 | Repeat) + (1 | Repeat:Picture) +
(1 | Repeat:TVset) + (1 | Repeat:TVset:Picture) + (1 | Assessor),
data = TVbo)
# Backward reduce the model:
(st <- step(tv)) # takes ~10 sec to run
# Pairwise comparisons of LS-means for Picture and TVset:
plot(st, which=c("Picture", "TVset"), pairwise = TRUE)
# }
# NOT RUN {
# }
Run the code above in your browser using DataLab