Dummy facets and group-anchoring for interactions

Dummy facets are facets intended only for investigating interactions, not for measuring main effects. All the elements of a dummy facet are anchored at 0.

 

This feature is useful for dummy facets removing classification elements from the measurement. These may be used to partition fit, detect bias or select rating scale (or partial credit) models.

 

Specifying dummy facets:

 

Labels=

2=Classifier,A ; anchoring wanted

1=Type A, 0 ; anchor at 0, so doesn't affect measurement

2=Type B, 0

 

or, and easier,

Labels=

2=Classifier, D ; dummy-facet anchoring wanted

1=Type A ; no anchor value required

2=Type B

 

When Umean= is used, use the Umean= value for these "dummy" elements

Umean=50,10 ; User mean is 50, user scaling is 10 per logit.

Labels=

2=Classifier,A ; anchoring wanted

1=Type A, 50 ; anchor at umean value, so doesn't affect measurement

2=Type B, 50

 

Or, easier

Labels=

2=Classifier,D ; dummy-facet anchoring wanted

1=Type A ; no anchor value required; automatically set at umean value.

2=Type B

 

Modeling with dummy facets:

Facets = 3 ; persons, classifiers, items

Models = ?, ?B, ?B, R ; interactions between classifiers and items

 

Example: Let's suppose we have three facets, "Rater leniency", "Rater experience", and "Examinees"

 

Scenario 1: measure the raters, report the groups

 

Facets=3

Models=?,?,?,R

Noncenter =1 ; noncentered facet cannot be anchored or dummy

Labels=

1, Rater leniency

1=Jose , , 1     ; novice

2=Maria , , 2   ; expert

....

*

2, Rater experience, D   ; dummy facet, all elements automatically anchored at 0

1=Novice

2=Expert

*

3, Examinees

...

*

 

Scenario 2: Measure the experience. Rater measures are relative to experience.

 

Facets=2

Models=?,?,R

Noncenter=2

Labels=

1, Rater leniency, A  ; group anchored by group

1=Jose, 0 , 1  

2=Maria, 0 , 2

....

*

2, Rater experience  ; standard facet

1=Novice

2=Expert

*

3, Examinees

...

*