From the Graphs menu:
A dialog box displays. In this example, using Essays.txt, we want a separate graph for each of the 12 Readers, showing their bias on the three Essays.
One graph for each, DIF/Bias/Interaction curves for the essays.
Click on Non-Uniform DIF.
Here I have selected the model curve for the Readers, and for Reader 1. A the non-uniform (empirical) and uniform DIF curves.
The model (no DIF) curve is the red curve. Uniform DIF is the smooth blue curve. Non-uniform DIF is the jagged blue curve.
Click on "Non-Uniform DIF" and make other selections. Multiple facets can be selected.
Adding DIF/Bias/Interaction facets to the analysis.
We need DIF/Bias classifiers to be elements of a facet. If the facet does not exist, but the classification codes are in labels, then we can easily add the DIF facet.
Example: Here is KCT.txt. We want Boy and Girl to be elements of facet 3 so we can use facet 3 for item bias by child's sex.
Changes and additions in bold:
; Kct.txt: c:\facets\examples\kct.txt
TITLE='Knox Cube Test (Best Test Design p.31)'
Facets = 3 ; two facets: children and items + facet sex
...
Model = ?,?,?,D ; elements of the two facets interact to produce dichotomous responses
Labels =
1,Children ; Children are facet 1
1-17=Boy,,1 ; Pretend boys, in group 1, are numbered 1 through 17.
18-35=Girl,,2 ; Pretend girls, in group 2, are numbered 18 through 35.
* ; end of child labels for facet 1
2,Tapping items ; Items are facet 2
1=1-4 ; Items labelled by the order in which the four blocks are tapped
....
18=4-1-3-4-2-1-4
* ; end of item labels
3, Sex
1= B ; Boy
2= G ; Girl
*
Dvalues = 3,1,1,1 ; the element number for facet 3 is the first character of the label in facet 1
Delements = LN ; facet elements can be indicated by labels
; no change to the data
Data = ; no data file name, so data follows immediately in this file
1 ,1 ,1 ; child 1 on item 1 scored 1...
.....