1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
|
\name{piplot}
\alias{piplot}
\title{Person-Item Plots for IRT Models}
\description{
Base graphics plotting function for person-item plot visualization of IRT models.
}
\usage{
piplot(object, pcol = NULL, histogram = TRUE, ref = NULL, items = NULL,
xlim = NULL, names = NULL, labels = TRUE, main = "Person-Item Plot",
xlab = "Latent trait", abbreviate = FALSE, cex.axis = 0.8, cex.text = 0.5,
cex.points = 1.5, grid = TRUE, \dots)
}
\arguments{
\item{object}{a fitted model object of class \code{"raschmodel"},
\code{"rsmodel"}, \code{"pcmodel"}, \code{"plmodel"} or \code{"gpcmodel"}.}
\item{pcol}{optional character (vector), specifying the color(s) used for the
person parameter plot.}
\item{histogram}{logical. For models estimated via MML (\code{plmodel}s and
\code{gpcmodel}s), should a histogram of the person-wise (individual) person
parameters be drawn additionally to the normal distribution density of the
person parameters?}
\item{ref}{argument passed over to internal calls of \code{\link{threshpar}}
and \code{\link{itempar}}.}
\item{items}{character or numeric, specifying the items which should be
visualized in the person-item plot.}
\item{xlim}{numeric, specifying the x axis limits.}
\item{names}{character, specifying labels for the items.}
\item{labels}{logical, whether to draw the number of the threshold as text
below the threshold.}
\item{main}{character, specifying the overall title of the plot.}
\item{xlab}{character, specifying the x axis labels.}
\item{abbreviate}{logical or numeric, specifying whether object names are to
be abbreviated. If numeric, this controls the length of the abbreviation.}
\item{cex.axis}{numeric, the magnification to be used for the axis notation
relative to the current setting of \code{cex}.}
\item{cex.text}{numeric, the magnification to be used for the symbols
relative to the current setting of \code{cex}.}
\item{cex.points}{numeric, the magnification to be used for the points
relative to the current setting of \code{cex}.}
\item{grid}{logical or color specification of horizontal grid lines. If set to
\code{FALSE} or \code{"transparent"} grid lines can be suppressed.}
\item{\dots}{further arguments passed to internal calls of
\code{\link{lines}}, \code{\link{points}} and \code{\link{text}}.}
}
\details{
The person-item plot visualization illustrates the distribution of the person
parameters against the absolute item threshold parameters under a certain data
set and IRT model. For models estimated via MML (\code{plmodel}s and
\code{gpcmodel}s), the normal distribution density of the person parameters is
drawn. If \code{histogram} is set to \code{TRUE} (the default), a histogram of
the person-wise (individual) person parameters is drawn additionally. If a
multiple group model has been fitted by supplying an \code{impact} variable,
multiple person parameter plots are drawn, each corresponding to a specific
level of this variable.
}
\seealso{\code{\link{curveplot}}, \code{\link{regionplot}},
\code{\link{profileplot}}, \code{\link{infoplot}}}
\examples{
## load verbal agression data
data("VerbalAggression", package = "psychotools")
## fit partial credit model to verbal aggression data
pcmod <- pcmodel(VerbalAggression$resp)
## create a person-item plot visualization of the fitted PCM
plot(pcmod, type = "piplot")
## just visualize the first six items and the person parameter plot
plot(pcmod, type = "piplot", items = 1:6, pcol = "lightblue")
\donttest{
if(requireNamespace("mirt")) {
## fit generalized partial credit model to verbal aggression data
gpcmod <- gpcmodel(VerbalAggression$resp)
## create a person-item plot visualization of the fitted GPCM
plot(gpcmod, type = "piplot")
## turn off the histogram and grid
plot(gpcmod, type = "piplot", histogram = FALSE, grid = FALSE)
## fit GPCM to verbal aggression data accounting for gender impact
mgpcmod <- gpcmodel(VerbalAggression$resp, impact = VerbalAggression$gender)
## create a person-item plot visualization of the fitted GPCM
plot(mgpcmod, type = "piplot", pcol = c("darkgreen", "darkorange"))
}
}
}
\keyword{aplot}
|
