Stata 15 help for irtgraph icc

[IRT] irtgraph icc -- Item characteristic curve plot

Syntax

Basic syntax

irtgraph icc [varlist] [, options]

Full syntax

irtgraph icc (varlist [, plot_options]) (varlist [, plot_options]) [...] [, options]

varlist is a list of items from the currently fitted IRT model.

options Description ------------------------------------------------------------------------- Plots blocation[(line_options)] add vertical lines for estimated item difficulties plocation[(line_options)] add horizontal lines for midpoint probabilities bcc plot boundary characteristic curves for categorical items ccc plot category characteristic curves range(# #) plot over theta = # to #

Line line_options affect rendition of the plotted curves

Add plots addplot(plot) add other plots to the ICC plot

Y axis, X axis, Titles, Legend, Overall twoway_options any options other than by() documented in [G-3] twoway_options

Data n(#) evaluate curves at # points; default is n(300) data(filename[, replace]) save plot data to a file -------------------------------------------------------------------------

plot_options Description ------------------------------------------------------------------------- blocation[(line_options)] add vertical lines for estimated item difficulties plocation[(line_options)] add horizontal lines for midpoint probabilities bcc plot boundary characteristic curves for categorical items ccc plot category characteristic curves line_options affect rendition of the plotted curves -------------------------------------------------------------------------

varlist may use factor-variable notation; see fvvarlist. line_options in plot_options override the same options specified in options.

Menu

Statistics > IRT (item response theory)

Description

irtgraph icc plots item characteristic curves (ICCs) for binary items and category characteristic curves (CCCs) for categorical items for the currently fitted IRT model.

Options

+-------+ ----+ Plots +------------------------------------------------------------

blocation[(line_options)] specifies that for each ICC, a vertical line be drawn from the estimated difficulty parameter on the x axis to the curve. The optional line_options specify how the vertical lines are rendered; see [G-3] line_options. This option implies option bcc.

plocation[(line_options)] specifies that for each ICC, a horizontal line be drawn from the midpoint probability on the y axis to the curve. The optional line_options specify how the horizontal lines are rendered; see [G-3] line_options. This option implies option bcc.

bcc specifies that boundary characteristic curves (BCCs) be plotted for categorical items. The ICCs for the individual item categories are plotted by default. This option has no effect on binary items.

ccc specifies that category characteristic curves (CCCs) be plotted for all items. This is the default behavior for categorical items. For binary items, this option will plot ICCs for both outcomes.

range(# #) specifies the range of values for theta. This option requires a pair of numbers identifying the minimum and maximum. The default is range(-4 4) unless the estimated difficulty parameters exceed these values, in which case the range is extended.

+------+ ----+ Line +-------------------------------------------------------------

line_options affect the rendition of the plotted ICCs; see [G-3] line_options.

+-----------+ ----+ Add plots +--------------------------------------------------------

addplot(plot) allows adding more graph twoway plots to the graph; see [G-3] addplot_option.

+-----------------------------------------+ ----+ Y axis, X axis, Titles, Legend, Overall +--------------------------

twoway_options are any of the options documented in [G-3] twoway_options, excluding by(). These include options for titling the graph (see [G-3] title_options) and for saving the graph to disk (see [G-3] saving_option).

+------+ ----+ Data +-------------------------------------------------------------

n(#) specifies the number of points at which the ICCs, CCCs, and BCCs are to be evaluated. The default is n(300).

data(filename[, replace]) saves the plot data to a Stata data file.

Examples

Setup . webuse masc1

Fit a 1PL model to binary items q1-q9 . irt 1pl q1-q9

Replay the table of estimated IRT parameters, sorting the output by parameter instead of by item and in ascending order of difficulty . estat report, byparm sort(b)

Use the 1PL parameters to plot the item characteristic curves for all items in the model . irtgraph icc

Same as above, but plot only specified items and shrink the legend and move it inside the plot region . irtgraph icc q8 q3 q9 q1 q2 q4 q6 q7 q5, legend(pos(4) col(1) ring(0) size(small))

Use the 1PL parameters to plot the item characteristic curves, placing the highest and lowest difficulty items in a one plotting group with their difficulty locations plotted and the remaining items in another plotting group . irtgraph icc (q5 q8, blocation) (q1-q4 q6 q7 q9), legend(off)

Same as above, but let's add options to make items q5 and q8 stand out . irtgraph icc (q8 q5, lcolor(black) lwidth(thick) bloc(lcolor(black))) (q1-q4 q6 q7 q9, lpattern(dash)), range(-5 5) xlabel(-5 -2.41 0 1.65 5) legend(off) lcolor(red)

Stored results

irtgraph icc stores the following in r():

Macros r(xvals) values used to label the x axis r(yvals) values used to label the y axis


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index