mirt (version 1.17.1)

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. Supports up to three dimensional models.

Usage

itemplot(object, item, type = "trace", degrees = 45, CE = FALSE,
  CEalpha = 0.05, CEdraws = 1000, drop.zeros = FALSE, theta_lim = c(-6,
  6), shiny = FALSE, rot = list(xaxis = -70, yaxis = 30, zaxis = 10),
  par.strip.text = list(cex = 0.7), par.settings = list(strip.background =
  list(col = "#9ECAE1"), strip.border = list(col = "black")),
  auto.key = list(space = "right"), ...)

Arguments

object

a computed model object of class SingleGroupClass 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'), standard errors ('SE'), item trace lines ('trace'), information and standard errors ('infoSE') or information and trace lines ('infotrace'), relative

degrees

the degrees argument to be used if there are two or three factors. See iteminfo for more detail. A new vector will be required for three dimensional models to override the default

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

theta_lim

lower and upper limits of the latent trait (theta) to be evaluated, and is used in conjunction with npts

shiny

logical; run interactive display for item plots using the shiny interface. This primarily is an instructive tool for demonstrating how item response curves behave when adjusting their parameters

rot

a list of rotation coordinates to be used for 3 dimensional plots

par.strip.text

plotting argument passed to lattice

par.settings

plotting argument passed to lattice

auto.key

plotting argument passed to lattice

...

additional arguments to be passed to lattice and coef()

Examples

data(LSAT7)
fulldata <- expand.table(LSAT7)
mod1 <- mirt(fulldata,1,SE=TRUE)
mod2 <- mirt(fulldata,1, itemtype = 'Rasch')
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
itemplot(mod3, 4, type = 'info')
itemplot(mod3, 4, type = 'infocontour')
itemplot(mod3, 4, type = 'tracecontour')

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

# use the directlabels package to put labels on tracelines
library(directlabels)
plt <- itemplot(pmod, 3)
direct.label(plt, 'top.points')

# uncomment to run interactive shiny applet
# itemplot(shiny = TRUE)