com.itemanalysis.psychometrics.irt.estimation

Class IrtObservedScoreDistribution

java.lang.Object

com.itemanalysis.psychometrics.irt.estimation.IrtObservedScoreDistribution

public class IrtObservedScoreDistribution
extends java.lang.Object

This class computes the distribution of summed scores for a given set of item response models. It uses the recursive algorithm developed by Thissen, et al. It assumes that test scores range from 0 to the sum of the maximum points awarded by each item. The value of the test score is also the index of the score in the array. See: Thissen, D., Pommerich, M., Billeaud, K., and Williams, V. S. L. (1995). Item response theory for scores on tests including polytomous items with ordered responses. Applied Psychological Measurement, 19 (1), 39-49. Thissen, D., and Wainer, H. (2001). Test scoring. Mahwah, NJ: Lawrence Erlbuam.

Constructor Summary

Constructors

Constructor and Description
IrtObservedScoreDistribution(ItemResponseModel[] irm, DistributionApproximation latentDistribution)

Method Summary

Modifier and Type	Method and Description
void	compute() Computes the IRT observed score distribution for a test that contains binary items, polytomous items, or binary and polytomous items.
void	computeAllBinaryItems() Computes the IRT observed score distribution for a test that contains only binary items.
double	getEAP(int summedScore) Computes the EAP estimate of examinee ability (theta) for a given summed score.
DistributionApproximation	getLatentDistribution()
int	getNumberOfScores()
double	getProbabilityAt(int thetaIndex, int summedScore) Provides the probability of a summed score given a value of theta.
java.lang.String	toString()

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Detail

IrtObservedScoreDistribution

public IrtObservedScoreDistribution(ItemResponseModel[] irm,
                                    DistributionApproximation latentDistribution)

Method Detail

computeAllBinaryItems

public void computeAllBinaryItems()

Computes the IRT observed score distribution for a test that contains only binary items. This recursive algorithm is described in the book Test Scoring. Thissen, D., and Wainer, H. (2001). Test scoring. Mahwah, NJ: Lawrence Erlbuam.

compute

public void compute()

Computes the IRT observed score distribution for a test that contains binary items, polytomous items, or binary and polytomous items. This method is preferred over computeAllBinaryItems() because it is more general and can handle binary item, polytomous item, and mixed format tests. It is an implementation of the recursive algorithm described in: Thissen, D., Pommerich, M., Billeaud, K., and Williams, V. S. L. (1995). Item response theory for scores on tests including polytomous items with ordered responses. Applied Psychological Measurement, 19 (1), 39-49. See also: Thissen, D., and Wainer, H. (2001). Test scoring. Mahwah, NJ: Lawrence Erlbuam.

getEAP

public double getEAP(int summedScore)

Computes the EAP estimate of examinee ability (theta) for a given summed score.

Returns:

EAP estimate of theta.

getNumberOfScores

public int getNumberOfScores()

getLatentDistribution

public DistributionApproximation getLatentDistribution()

getProbabilityAt

public double getProbabilityAt(int thetaIndex,
                               int summedScore)

Provides the probability of a summed score given a value of theta.

Parameters:

thetaIndex - position of the quadrature point (latent distribution value) in the array.

summedScore - a summed score.

Returns:

probability of a summed score.

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object