ParComparePlot: Plot comparing item parameters

In CambridgeAssessmentResearch/unimirt: Applications of Unidimensional IRT using the R Package 'mirt'

Plot comparing item parameters

Description

Common items are identified automatically on the basis of shared variable names. This function differs from "ItemParameterCompare" in that more than two IRT objects can be compared. In addition, if all models have been estimated with 'SE=TRUE' then this function include 95 per cent confidence intervals for item parameters making it easier to assess whether differences between item parameters are statistically significant.

Usage

ParComparePlot(
  mirtobjlist,
  mirtnames = NULL,
  commononly = TRUE,
  compare = "Difficulties"
)

Arguments

mirtobjlist	A list of estimated IRT models. For this function to make any sense at all these should all be of the same type.
mirtnames	A character vector of the same length as "mirtobjlist" use to add colour labels to the chart. Dy default they are just labelled obj1, obj2, ...
commononly	Logical value denoting whether only items that occur in more than one object should be included in the chart. Default TRUE.
compare	Character string equal to "Difficulties" (the default), "Slopes" or "Guessing" defining which item parameters should be compared.

Examples

## Not run: 
mirt1=unimirt(mathsdata[1:250,1:10],SE=TRUE)
mirt2=unimirt(mathsdata[251:500,6:15],SE=TRUE)

ParComparePlot(list(mirt1,mirt2),c("first250","next250"), compare = "Difficulties")
ParComparePlot(list(mirt1,mirt2),c("first250","next250"), compare = "Slopes")

## End(Not run)

CambridgeAssessmentResearch/unimirt documentation built on June 10, 2025, 6:03 a.m.