Learn R Programming

mirt (version 0.8.0)

itemplot: Displays item surface and information plots

Description

itemplot displays various item based IRT plots, with special options for plotting items that contain several 0 slope parameters.

Usage

itemplot(object, item, type = "trace", degrees = 45,
    CE = FALSE, CEalpha = 0.05, CEdraws = 1000,
    drop.zeros = FALSE, ...)

Arguments

object
a computed model object of class ExploratoryClass, ConfirmatoryClass, or MultipleGroupClass. Input may also be a list for comparing similar item types (e.g., 1PL vs 2PL)
item
a single numeric value, or the item name, indicating which item to plot
type
plot type to use, information ('info'), item trace lines ('trace'), relative efficiency lines ('RE'), expected score 'score', or information contours ('infocontour') (not for
degrees
the degrees argument to be used if there are exactly two factors. See iteminfo for more detail
CE
logical; plot confidence envelope?
CEalpha
area remaining in the tail for confidence envelope. Default gives 95% confidence region
CEdraws
draws number of draws to use for confidence envelope
drop.zeros
logical; drop slope values that are numerically close to zero to reduce dimensionality? Useful in objects returned from bfactor or other confirmatory models that contain several zero slopes
...
additional arguments to be passed to lattice

Examples

Run this code
data(LSAT7)
fulldata <- expand.table(LSAT7)
mod1 <- mirt(fulldata,1,SE=TRUE)
mod2 <- mirt(fulldata,1, itemtype = '1PL')
mod3 <- mirt(fulldata,2)

itemplot(mod1, 2)
itemplot(mod1, 2, CE = TRUE)
itemplot(mod1, 2, type = 'info')
itemplot(mod1, 2, type = 'info', CE = TRUE)

mods <- list(twoPL = mod1, onePL = mod2)
itemplot(mods, 1, type = 'RE')

#multidimensional info
itemplot(mod3, 3, type = 'info')

#polytomous items
pmod <- mirt(Science, 1, SE=TRUE, SE.type = 'MHRM')
itemplot(pmod, 3)
itemplot(pmod, 3, CE = TRUE)

Run the code above in your browser using DataLab