RaschRatingScaleGroup

java.lang.Object
- com.itemanalysis.psychometrics.irt.model.RaschRatingScaleGroup

```
public class RaschRatingScaleGroup
extends java.lang.Object
```
A rating scale group is a set of test items that share the same rating scale structure. Category frequencies for items in a rating scale group are combined for estimation of threshold parameters. This class is used for combining the frequencies and holding estimates of the thresholds and also their fit statistics.

Author:

J. Patrick Meyer

Constructor Summary

Constructors
Constructor and Description
`RaschRatingScaleGroup(java.lang.String groupId, int nCat)` Creates a rating scael group using an ID and set number of categories.

Method Summary

All Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`void`	`acceptAllProposalValues()` After a complete iteration, the threshold values are replaced by the prposal values.
`void`	`addItem(ItemResponseModel irm, ItemResponseSummary isum, int position)` Called for every item on the test, this method will add any item that has the same groupId and number of score categories.
`void`	`checkForDroppping()` Threshold parameters cannot be estimated if one or more categories do not have any observations.
`void`	`clearCounts()` Resets the category counts.
`void`	`computeCategoryStandardError(double[] theta, int[] extremePersons)` Computes the standard error of the threshold parameters.
`int`	`dropStatus()` Status of the item with response to dropping.
`RaschCategoryFitStatistic`	`getCategoryFitAt(int index)` Gets category fit statistics for a particular category.
`java.lang.String`	`getGroupId()` A groupId uniquely identifies each item in the rating scale group.
`java.util.Iterator<ItemResponseModel>`	`getIterator()` An iterator for item response models for this group.
`int`	`getNumberOfCategories()` The number of score categories for this group is fixed.
`int[]`	`getPositions()` The columns of data for each item in the group is recorded and stored in an array.
`double[]`	`getProposalThresholds()` During each iteration of the joint maximum likelihood routine, preliminary threshold estimates are obtained.
`byte[]`	`getScoreWeights()` Each category has an associated score weight.
`double`	`getThresholdAt(int index)` Gets threshold estimates for a particular category.
`double[]`	`getThresholds()` Threshold parameters represent the point where the probability of scoring in two adjacent categories is equal.
`double[]`	`getThresholdStandardError()` Each threshold parameter has an associated standard error.
`double`	`getThresholdStdErrorAt(int index)` Gets a specific threshold standard error.
`void`	`incrementFitStatistics(ItemResponseModel model, double theta, byte Xni)` Category fit statistics are incrementally updated with repeated calls to this method.
`boolean`	`isFixed()`
`double`	`probabilitySumAt(double theta, int k)` Sums the probability of responding in category at position k over all items in this rating scale group.
`void`	`setProposalThresholds(double[] proposalThresholds)` Sets the array of proposal values.
`void`	`setThresholds(double[] thresholds)` Sets the array of threshold estimates.
`double`	`SpjAt(int index)` Spj is the sum of Sij over all categories.
`java.lang.String`	`toString()` Category frequency summaries are stored in this object.
`double`	`TpjAt(int index)` This count is part of the threshold update in the partial credit and rating scale model.

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

- Constructor Detail
  - RaschRatingScaleGroup
```
public RaschRatingScaleGroup(java.lang.String groupId,
                             int nCat)
```
    Creates a rating scael group using an ID and set number of categories. Any item with the same group ID will be placed into this group.
    
    Parameters:
    
    groupId - A unique identifier for the group.
    
    nCat - number of score categories for this group.
- Method Detail
  - addItem
```
public void addItem(ItemResponseModel irm,
                    ItemResponseSummary isum,
                    int position)
```
    Called for every item on the test, this method will add any item that has the same groupId and number of score categories.
    
    Parameters:
    
    irm - an item response model.
    
    isum - an item summary object.
    
    position - column position of the item in the data array.
  - getPositions
```
public int[] getPositions()
```
    The columns of data for each item in the group is recorded and stored in an array. Gets the array of column positions.
    
    Returns:
    
    column positions for items in this group.
  - getIterator
```
public java.util.Iterator<ItemResponseModel> getIterator()
```
    An iterator for item response models for this group.
    
    Returns:
    
    an iterator.
  - getGroupId
```
public java.lang.String getGroupId()
```
    A groupId uniquely identifies each item in the rating scale group. Gets the groupId value.
    
    Returns:
    
    groupId
  - getNumberOfCategories
```
public int getNumberOfCategories()
```
    The number of score categories for this group is fixed. Gets the number of score categories.
    
    Returns:
    
    number of score categories.
  - TpjAt
```
public double TpjAt(int index)
```
    This count is part of the threshold update in the partial credit and rating scale model.
    
    Parameters:
    
    index -
    
    Returns:
  - SpjAt
```
public double SpjAt(int index)
```
    Spj is the sum of Sij over all categories.
    
    Parameters:
    
    index - an index of the category for which the statistic is sought.
    
    Returns:
    
    Spj for a category.
  - checkForDroppping
```
public void checkForDroppping()
```
    Threshold parameters cannot be estimated if one or more categories do not have any observations. This method checks for empty categories and sets the flag for the item. A code of 0 indicates that all categories have at least one observation, a code of -1 indicates a category with no observations, and a code of +1 indicates a category that contains all of the responses. Do not drop an item if it has fixed parameters.
  - dropStatus
```
public int dropStatus()
```
    Status of the item with response to dropping. See checkForDroppping().
    
    Returns:
  - clearCounts
```
public void clearCounts()
```
    Resets the category counts. This method is used during the data summary in JointMaximumLikelihoodEstimation.summarizeData(double).
  - getThresholds
```
public double[] getThresholds()
```
    Threshold parameters represent the point where the probability of scoring in two adjacent categories is equal. They are estimated in JointMaximumLikelihoodEstimation. Threshold estimates are stored in an array. This method gets the stored estimates.
    
    Returns:
    
    threshold estimates.
  - getThresholdAt
```
public double getThresholdAt(int index)
```
    Gets threshold estimates for a particular category.
    
    Parameters:
    
    index - position of category in threshold array.
    
    Returns:
    
    estimate for the threshold at the position given by index.
  - setThresholds
```
public void setThresholds(double[] thresholds)
```
    Sets the array of threshold estimates. This should only be used for setting initial values.
    
    Parameters:
    
    thresholds - values of thresholds.
  - getProposalThresholds
```
public double[] getProposalThresholds()
```
    During each iteration of the joint maximum likelihood routine, preliminary threshold estimates are obtained. These proposal values are stored in a separate array. After one complete iteration is complete, the threshold values are set to the proposal values. See JointMaximumLikelihoodEstimation.updateThresholds(RaschRatingScaleGroup).
    
    Returns:
  - setProposalThresholds
```
public void setProposalThresholds(double[] proposalThresholds)
```
    Sets the array of proposal values.
    
    Parameters:
    
    proposalThresholds - proposed threshold values.
  - acceptAllProposalValues
```
public void acceptAllProposalValues()
```
    After a complete iteration, the threshold values are replaced by the prposal values.
  - isFixed
```
public boolean isFixed()
```
  - getScoreWeights
```
public byte[] getScoreWeights()
```
    Each category has an associated score weight. They are stored in an array. Gets the score weights.
    
    Returns:
    
    score weights.
  - probabilitySumAt
```
public double probabilitySumAt(double theta,
                               int k)
```
    Sums the probability of responding in category at position k over all items in this rating scale group.
    
    Parameters:
    
    theta - person ability estimate.
    
    k - score category for which the probabilities are summed.
    
    Returns:
  - computeCategoryStandardError
```
public void computeCategoryStandardError(double[] theta,
                                         int[] extremePersons)
```
    Computes the standard error of the threshold parameters.
    
    Parameters:
    
    theta - array of person ability values
    
    extremePersons - array of extreme person flags
  - getThresholdStandardError
```
public double[] getThresholdStandardError()
```
    Each threshold parameter has an associated standard error. The standard errors are not computed during estimation. You must call computeCategoryStandardError(double[], int[]) before calling this method. Otherwise, the returned standard errors will be zero (and will be incorrect).
    
    Returns:
    
    array of threshold standard errors.
  - getThresholdStdErrorAt
```
public double getThresholdStdErrorAt(int index)
```
    Gets a specific threshold standard error.
    
    Parameters:
    
    index - array position of the particular threshold standard error.
    
    Returns:
    
    a threshold standard error.
  - incrementFitStatistics
```
public void incrementFitStatistics(ItemResponseModel model,
                                   double theta,
                                   byte Xni)
```
    Category fit statistics are incrementally updated with repeated calls to this method.
    
    Parameters:
    
    model - and item response model.
    
    theta - person ability estimate.
    
    Xni - an observed item response.
  - getCategoryFitAt
```
public RaschCategoryFitStatistic getCategoryFitAt(int index)
```
    Gets category fit statistics for a particular category.
    
    Parameters:
    
    index - array index of category fit statistics.
    
    Returns:
    
    category fit statistics.
  - toString
```
public java.lang.String toString()
```
    Category frequency summaries are stored in this object. This method provides a listing of the frequencies.
    
    Overrides:
    
    toString in class java.lang.Object
    
    Returns:
    
    output of response frequencies.

Class RaschRatingScaleGroup

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Constructor Detail

RaschRatingScaleGroup

Method Detail

addItem

getPositions

getIterator

getGroupId

getNumberOfCategories

TpjAt

SpjAt

checkForDroppping

dropStatus

clearCounts

getThresholds

getThresholdAt

setThresholds

getProposalThresholds

setProposalThresholds

acceptAllProposalValues

isFixed

getScoreWeights

probabilitySumAt

computeCategoryStandardError

getThresholdStandardError

getThresholdStdErrorAt

incrementFitStatistics

getCategoryFitAt

toString