Learn R Programming

psychotools (version 0.4-0)

profileplot: Profile Plots for IRT Models

Description

Base graphics plotting function for profile plot visualization of IRT models.

Usage

profileplot(object, what = c("items", "thresholds", "discriminations"), parg = list(type = NULL, ref = NULL, alias = TRUE), index = TRUE, names = TRUE, main = NULL, abbreviate = FALSE, ref = TRUE, col = "lightgray", border = "black", pch = NULL, cex = 1, refcol = "lightgray", linecol = "black", lty = 2, ylim = NULL, xlab = NULL, ylab = NULL, add = FALSE, srt = 45, adj = c(1.1, 1.1), axes = TRUE, ...)

Arguments

object
a fitted model object of class "raschmodel", "rsmodel" or "pcmodel".
what
character, specifying the type of parameters to be plotted.
parg
list of arguments passed over to internal calls of itempar, threshpar or discrpar.
index
logical, should different indexes for different items be used?
names
logical or character. If TRUE, the names of the items are displayed on the x-axis. If FALSE, numbers of items are shown. Alternatively a character vector of the same length as the number of items can be supplied.
main
character, specifying the overall title of the plot.
abbreviate
logical or numeric, specifying whether object names are to be abbreviated. If numeric this controls the length of the abbreviation.
ref
logical, whether to draw a horizontal line for the reference level. Only takes effect if argument what is "items" or "discriminations".
col, border, pch, cex
Graphical appearance of plotting symbols. Can be of the same length as the number of items, i.e., a different graphical appearance is used for each item. If what = "thresholds", col and pch can be matrices with a number of columns equal to the number of threshold parameters per item resulting in different symbols and colors used for different threshold parameter profiles.
refcol
character, specifying the line color for the reference line (if ref).
linecol
character or numeric, specifying the line color to be used for the profiles.
lty
numeric, specifying the line type for the profiles.
ylim
numeric, specifying the y axis limits.
xlab, ylab
character, specifying the x and y axis labels.
add
logical. If TRUE, new plotted profiles are added to an existing plot.
srt, adj
numeric. Angle (srt) and adjustment (adj) in case names (rather than numbers) are used as x-axis labels. These are passed to text.
axes
logical. Should axes be drawn?
...
further arguments passed over to plot.

Details

The profile plot visualization illustrates profiles of specific estimated parameters under a certain IRT model.

See Also

curveplot, regionplot, infoplot, piplot

Examples

Run this code
## Load Verbal Aggression data
data("VerbalAggression", package = "psychotools")

## Fit Rasch, rating scale and partial credit
## model to VerbalAggression data
rmmod <- raschmodel(VerbalAggression$resp2)
rsmod <- rsmodel(VerbalAggression$resp)
pcmod <- pcmodel(VerbalAggression$resp)

## Profile plots of the item parameters of
## the three fitted IRT models
plot(rmmod, type = "profile", what = "items", col = 4)
plot(rsmod, type = "profile", what = "items", col = 2, add = TRUE)
plot(pcmod, type = "profile", what = "items", col = 3, add = TRUE)
legend(x = "topleft", legend = c("RM", "RSM", "PCM"),
  col = 1, bg = c(4, 2, 3), pch = 21, bty = "n")

## Profile plots of the threshold parameters of type 'mode'
plot(rmmod, type = "profile", what = "thresholds",
  parg = list(type = "mode"))
plot(rsmod, type = "profile", what = "thresholds",
  parg = list(type = "mode"))
plot(pcmod, type = "profile", what = "thresholds",
  parg = list(type = "mode"))

## Profile plot of the discrimination parameters
## of a dichotomous Rasch model (all zero by definition)
plot(rmmod, type = "profile", what = "discrimination")

Run the code above in your browser using DataLab