ScoreDistFromMirt: Function to estimate sum score distribution for a selection...

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

Function to estimate sum score distribution for a selection of items in a mirt object

Description

This function can be used to help facilitate observed score IRT equating. Early trials suggest this function works best with unidimensional models.

Usage

ScoreDistFromMirt(mirtobj, which.items = NULL, theta.mean.sd = NULL)

Arguments

mirtobj	An estimated IRT model (of class SingleGroupClass) estimated either using mirt or unimirt. Must be fitted using the (default) "EM" method. Will work with any form of model (graded response, Rasch,...). Will even work with multidimensional models but may be innaccurate unless the object incorporates a large number of quadrature points.
which.items	A vector of denoting which items should be included in calculating the score.
theta.mean.sd	A vector of length 2 giving the mean and standard deviation of the ability distribution. If not supplied then these are derived directly from the mirt object. This parameter may only be used for unidimensional models.

Value

The function returns a data.frame with columns: score (a vector of possible scores) ,prob (the proportion of candidates expected to achieve each scores), expectedtheta (expected ability given the total score), and sd theta the standard deviation of abilities for candidates with the given total score.

Examples

## Not run: 
mirt1=unimirt(mathsdata,"2")
ScoreDistFromMirt(mirt1)

## End(Not run)

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