Table 11 summarizes the response modeling in the bias analysis
Bias/Interaction: 1. Student, 2. Format
+--------------------------------------------------+
| Cat Score Exp. Resd StRes| |
|-----------------------------+--------------------|
| 9.60 9.60 9.60 .01 .01 | Mean (Cnt: 296) |
| 1.34 1.34 1.30 .31 .53 | S.D. (Population) |
| 1.34 1.34 1.30 .31 .53 | S.D. (Sample) |
+--------------------------------------------------+
Count of measurable responses = 296.00
Raw-score variance of observations = 1.80 100.00%
Variance explained by Rasch measures = 1.53 85.00%
Variance of residuals = 0.27 15.00%
Variance explained by bias/interactions = 0.18 10.00%
Variance remaining in residuals = 0.09 5.00%
Identification
|
Meaning
|
Cat
|
Observed value of the category in the data file
|
Score
|
Value of category in this analysis. This is less than Cat. if there are structural zeroes.
|
Exp.
|
Expected score based on current measures
|
Resd
|
Residual, the score difference between Score and Exp.
|
StRes
|
The residual divided by its standard error. The Rasch model is now over-parameterized, so the standardized residuals are expected to have an S.D. less than 1.0.
|
Mean
|
average of the observations
|
Cnt:
|
Count is the number of observations
|
S.D. (Population)
|
standard deviation treating this sample as the entire population
|
S.D. (Sample)
|
standard deviation treating this sample as a sample from the population. It is larger than S.D. (Population).
|
Count of measurable responses
|
the number of ratings used in estimating the bias terms. This may be only a small fraction of the entire data set.
|
Raw-score variance of observations
|
square of S.D. (Population) of Score in Table 11
|
Variance explained by Rasch measures
|
from Table 5
|
Variance of residuals
|
from Table 5
|
Variance explained by bias/interactions
|
Variance of Residuals - Variance remaining in residuals
|
Variance remaining in residuals
|
square of S.D. (Population) of Resd in Table 11
|
