A better plot may result from copying Facets graph coordinates, see "Copy Plot Data to Clipboard".

 

Provides the values plotted in Table 8 for the probability curves and expected score ogive in fixed column format. If comma-separated values are preferred, specify "Graphfile=filename" and CSV=Y. For Tab-separated values, specify "Graphfile=filename" and CSV=Tab

 

filenames ending .xls and .xlsx are written as Excel workbooks

filenames ending .sav are written as SPSS save files

filenames ending .rda and .rdata are written as R Statistics data files

filenames ending with any other suffix or no suffix are written as DOS text files.

 

Heading lines= controls display of heading lines. QM= quotation marks, controls whether labels are within quotation marks.

 

From the Output Tables menu, this Graph Output file controls the production of the Graph output file

 

 

    Scale   Measure  Expected    ExpCat    Prob:0    Prob:1    Prob:2

        1     -3.00       .10         0       .90       .10       .00

        1     -2.91       .11         0       .89       .11       .00

        1     -2.82       .12         0       .88       .11       .00

 

Example 1: Provide coordinates so I can plot beautiful probability curves using my graphics package:

Graphfile=plot.dat,CSV; the name of an output file with comma-separated values, produces, for a 3 category rating scale (or partial credit):

 

1,-2.00,.23,0,.79,.20,.01

1,-1.92,.24,0,.77,.21,.02

1,-1.85,.26,0,.76,.22,.02

 

Example 2: You want a computed raw rating (score) on each criterion for each essay. The computed raw rating matches the observed raw rating (score). The observed rating is awarded by Rater X. The computed rating is awarded by Rater Z, who has the mean rater leniency, which is usually zero.

 

1. Output the Facets Graph file.

2. Compute the difference between the measure for Rater X and the measure for Rater Z.

3. Find the Measure for the observed rating (in the Expected Score column) in the Graph file

4. Adjust this measure by the measure difference from 2.

5. The computed rating is the Expected Score at the adjusted Measure.

 

If the rater is lenient, then the computed rating will be lower than the observed rating (except for the bottom category).

If the rater is severe, then the computed rating will be higher than the observed rating (except for the top category).