plot.ltm {ltm} | R Documentation |
Produces the Item Characteristic Curves or Surfaces for fitted ltm
objects.
## S3 method for class 'ltm': plot(x, items = NULL, legend = FALSE, cx = -3.4, cy = 0.9, ...)
x |
an object inheriting from class ltm . |
items |
a numeric vector denoting which items to plot; if NULL all items are plotted. |
legend |
logical; if TRUE a legend is printed. |
cx |
the x-coordinate to be passed to legend . Used only for the one-factor model. |
cy |
the y-coordinate to be passed to legend . Used only for the one-factor model. |
... |
extra graphical parameters to be passed to lines (one-factor case) or to persp (two-factors case). |
Item characteristic curves (also known as item response functions) show how the probability of a positive response, in each item, increases with the values of the latent variables.
If x
is a linear two-factor model, then the plot of the standardized loadings (see coef.ltm
)
is also produced.
When two latent variables are used, plot.ltm
uses options(graphics.record = TRUE)
, i.e.,
the plots are available for scrolling.
Dimitris Rizopoulos dimitris.rizopoulos@med.kuleuven.be
## Item Characteristic Curves for the two-parameter logistic ## model; plot only items 1, 2, 4 and 6: m <- ltm(Wirs ~ z1) plot(m, items = c(1, 2, 4, 6), lwd = 3, cex = 1.4) ## Not run: ## because of an error during R CMD CHECK; you can run it. ## Item Characteristic Surfaces for the interaction model: m <- ltm(Wirs ~ z1 * z2) plot(m) ## End(Not run)