This specification identifies to which facet each element number in a data line belongs. If you want persons as facet 1, items as facet 2, and judges as facet 3, but the data has been entered with facets in the order of persons (facet 1), judges (facet 3), items (facet 2), then

Facets = 3

Labels =

1 = Person

|

*

2 = Item

|

*

3 = Judge

|

*

Entry = 1,3,2

Data =

4,6,5,2 ; Person 4 was rated by judge 6 on item 5 with a "2"

 

If there are more facets in the data than you wish to include in this analysis, extra facets may be ignored by specifying them as 0.

e.g., if whatever is in the second data facet is to be ignored,

Facets = 3  ; number of facets in the data

Entry = 2,0,1 ; only look at facets 1,2 in Labels=

 

Then for all other purposes, there are only two facets in each data line, so that

Model = ?,?,D ; so only two facets in the model statement: facet 2, facet 1, Dichotomous

 

A facet may be entered twice for paired comparison analysis.

For example, when a judge (facet 1) compares the performance of person A (facet 2) with that of person B (also facet 2), then

Facets = 3

Entry = 1,2,2

Model = ?,?,-?,B1,0.5 ; "-?" reverses the sign of the second person measure

; enter each data point twice: A vs. B, and then B vs. A. so weighting is 0.5