Residuals / Responses Output File = ""

If a residual/response output filename is specified by Residual file=, a file of residuals from the main analysis is produced. This file is designed for input into other programs with one line per measurable observation. It can be used for calculating other fit statistics and producing specialized diagnostic reports.

Heading lines= control the output of the heading line. CSV= allows tab-delimited and other formats. QM quotation marks, controls whether labels are within quotation marks.

This file can be produced from the Output Files menu by clicking on Residual/Response Output file. This has additional options.

Here is an example of the format. The precise format depends on the number of facets in your data:

Obs Stp Exp Res Var StRes Wt LProb Logit Displ Status 1 2 Child Activity

1 1 1.43 -.43 .38 -.70 1.00 -.8394 1.00 -1.13 1 1 1 1 Watch bir

2 2 1.54 .46 .33 .79 1.00 -.5340 1.31 1.37 1 1 2 1 Read abou

1 1 1.09 -.09 .46 -.13 1.00 -.6252 .19 -.19 1 1 3 1 Read abou

1 1 .52 .48 .36 .80 1.00 -.9015 -1.13 1.34 1 1 4 1 Watch gra

Fixed-format Columns	Meaning
1-10	response as observed in the data file
11-20	scored response as renumbered after omitting structural zeroes
21-30	expected score for this response
31-40	score residual: (observed - expected)
41-50	variance of expected score, the statistical information in this response
51-60	standardized residual: (residual / √ variance) ≈ unit normal deviate
61-70	weighting: (model weight * observation weight)
71-80	natural logarithm of the probability of the observation
81-90	logit sum of the element measures producing the observation
91-100	displacement in logits: (residual / variance)
101-110	status: 1 = used in estimation 2 = in one extreme score -1 = invalid after recounting -2 = with invalid elements -3 = in two extreme scores -4 = only in extreme scores
111-120	element number for facet 1
121-130	element number for facet 2
\|
..1-..0	element number for last facet

The Facets residual file shows two logit values.

1. "Logit" is the sum of the element logit measures hypothesized to generate the observation. These produce the Expected observation.

2. "Displacement" is an estimate of the logit deviance of the observation from the expected observation.

So, an estimate of the "logit score for the student on the station for that examiner" is "Logit" + "Displacement".