mirt (version 1.17.1)

plot-method: Plot various test-implied functions from models

Description

Plot various test implied response functions from models estimated in the mirt package.

Usage

## S3 method for class 'SingleGroupClass,missing':
plot(x, y, type = "score", npts = 50,
  degrees = 45, theta_lim = c(-6, 6), which.items = 1:extract.mirt(x,
  "nitems"), MI = 0, CI = 0.95, rot = list(xaxis = -70, yaxis = 30, zaxis
  = 10), facet_items = TRUE, main = NULL, drape = TRUE, colorkey = TRUE,
  ehist.cut = 1e-10, add.ylab2 = TRUE, 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"), profile = FALSE,
  ...)

Arguments

x

an object of class SingleGroupClass, MultipleGroupClass, or DiscreteClass

y

an arbitrary missing argument required for R CMD check

type

type of plot to view; can be 'info' to show the test information function, 'rxx' for the reliability function, 'infocontour' for the test information contours, 'SE' for the test standard error function,

npts

number of quadrature points to be used for plotting features. Larger values make plots look smoother

degrees

numeric value ranging from 0 to 90 used in plot to compute angle for information-based plots with respect to the first dimension. If a vector is used then a bubble plot is created with the summed information across the angles specified (e.g.,

theta_lim

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

which.items

numeric vector indicating which items to be used when plotting. Default is to use all available items

MI

a single number indicating how many imputations to draw to form bootstrapped confidence intervals for the selected test statistic. If greater than 0 a plot will be drawn with a shaded region for the interval

CI

a number from 0 to 1 indicating the confidence interval to select when MI input is used. Default uses the 95% confidence (CI = .95)

rot

allows rotation of the 3D graphics

facet_items

logical; apply grid of plots across items? If FALSE, items will be placed in one plot for each group

main

argument passed to lattice. Default generated automatically

drape

logical argument passed to lattice. Default generated automatically

colorkey

logical argument passed to lattice. Default generated automatically

ehist.cut

a probability value indicating a threshold for excluding cases in empirical histogram plots. Values larger than the default will include more points in the tails of the plot, potentially squishing the 'meat' of the plot to take up less area than visually

add.ylab2

logical argument passed to lattice. Default generated automatically

par.strip.text

plotting argument passed to lattice

par.settings

plotting argument passed to lattice

auto.key

plotting argument passed to lattice

profile

logical; provide a profile plot of response probabilities (objects returned from mdirt only)

...

additional arguments to be passed to lattice

Examples

x <- mirt(Science, 1, SE=TRUE)
plot(x)
plot(x, type = 'info')
plot(x, type = 'infotrace')
plot(x, type = 'infotrace', facet_items = FALSE)
plot(x, type = 'infoSE')
plot(x, type = 'rxx')

# confidence interval plots when information matrix computed
plot(x)
plot(x, MI=100)
plot(x, type='info', MI=100)
plot(x, type='SE', MI=100)
plot(x, type='rxx', MI=100)

# use the directlabels package to put labels on tracelines
library(directlabels)
plt <- plot(x, type = 'trace')
direct.label(plt, 'top.points')

set.seed(1234)
group <- sample(c('g1','g2'), nrow(Science), TRUE)
x2 <- multipleGroup(Science, 1, group)
plot(x2)
plot(x2, type = 'trace')
plot(x2, type = 'trace', which.items = 1:2)
plot(x2, type = 'trace', which.items = 1, facet_items = FALSE) #facet by group
plot(x2, type = 'info')

x3 <- mirt(Science, 2)
plot(x3, type = 'info')
plot(x3, type = 'SE', theta_lim = c(-3,3))