com.itemanalysis.psychometrics.irt.estimation

Class RaschCategoryFitStatistic

java.lang.Object

com.itemanalysis.psychometrics.irt.estimation.RaschCategoryFitStatistic

public class RaschCategoryFitStatistic
extends java.lang.Object

Storeless computation of Rasch category INFIT and OUTFIT statistics. Statistics are incrementally updated with each observation and then computed before the result is returned.

Constructor Summary

Constructors

Constructor and Description
RaschCategoryFitStatistic(java.lang.Integer k, byte[] scoreWeight)

Method Summary

Modifier and Type	Method and Description
double	getUnweightedMeanSquare() Computes and the returns the OUTFIT mean square fit statistic.
double	getWeightedMeanSquare() Computes and then returns the INFIT mean square fit statistic.
void	increment(double Xni, double theta, ItemResponseModel model)
double	kurtosisOfResponse(ItemResponseModel irm, double theta) Kurtosis of response Xni.
double	varianceOfResponse(ItemResponseModel irm, double theta) Variance of response Xni.

Methods inherited from class java.lang.Object

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

Constructor Detail

RaschCategoryFitStatistic

public RaschCategoryFitStatistic(java.lang.Integer k,
                                 byte[] scoreWeight)

Method Detail

increment

public void increment(double Xni,
                      double theta,
                      ItemResponseModel model)

getUnweightedMeanSquare

public double getUnweightedMeanSquare()

Computes and the returns the OUTFIT mean square fit statistic.

Returns:

OUTFIT mean squares fit statistic.

getWeightedMeanSquare

public double getWeightedMeanSquare()

Computes and then returns the INFIT mean square fit statistic.

Returns:

INFIT mean square fit statistic.

varianceOfResponse

public double varianceOfResponse(ItemResponseModel irm,
                                 double theta)

Variance of response Xni. This method is needed for computation of fit statistics.

Parameters:

theta - examinee ability

Returns:

kurtosisOfResponse

public double kurtosisOfResponse(ItemResponseModel irm,
                                 double theta)

Kurtosis of response Xni. This method is needed for computation of fit statistics.

Parameters:

theta -

Returns: