Table 27.1 Item subtotal summaries on one line

Subtotal specification is: ISUBTOTAL=$S1W1

identifies the columns in the item label to be used for classifying the item by $S1W1 or whatever, using the column selection rules.  

EXTREME AND NON-EXTREME KID SCORES

ALL SCORES ARE NON-EXTREME

NON-EXTREME SCORES ONLY

The items included in this summary table.

Items with non-extreme scores (omits items with 0% and 100% success rates)

ITEM COUNT

count of items. "ITEM" is the name assigned with ITEM=

MEAN SCORE

weighted average item score by the persons

MEAN COUNT

weighted average of the count of responses by the persons

MEAN MEASURE

average measure of items

S.E. MEAN

standard error of the average measure of items. If only one item, then the S.E. of the item estimate

P.SD

population standard deviation of the item measures.

S.SD

sample standard deviation of the item measures.

MEDIAN

the measure of the middle item

REAL/MODEL SEPARATION

the separation coefficient: the "true" adjusted standard deviation / root-mean-square measurement error of the items (REALSE= inflated for misfit).

REAL/MODEL RELIABILITY

the item measure reproducibility = ("True" item measure variance / Observed variance) = Separation ² / (1 + Separation ²)

RMSE

Statistical average of the standard errors of the measures

TRUE SD

Observed population S.D. adjusted for measurement error

MEAN OUTFIT

Average outfit mean-square for the group. Expectation near 1.0

ITEM CODE

the classification code in the item label. The first line, "*", is the total for all items. The remaining codes are those in the item columns specified by $S1W1 or whatever, using the column selection rules.

SUBTOTAL RELIABILITY

the reliability (reproducibility) of the means of the subtotals = true variance / observed variance = (observed variance - error variance) / observed variance.

Observed variance = variance of MEAN MEASURES

Error variance = mean-square of the S.E. MEAN

inestimable = some subtotal counts are too small to estimate Reliability

UMEAN=0 USCALE=1

Current user-scaling

ITEM CODE

the classification code in the item label for subtotal "1"

CODE

the classification code in the item label for subtotal "2"

MEAN DIFFERENCE

difference between the mean measures of the two CODE subtotals, "1" and "2"

MEASURE

size of the difference between "1" and "2"

S.E.

standard error of the difference = sqrt ( (S.E. Mean "1")² + (S.E. Mean "2")² )

t

Student's t = MEASURE / S.E.

Welch2-sided

2-sided t-test using Welch's adaptation of Student's t-test.

d.f.

Welch's degrees of freedom

Prob.

two-sided probability of Student's t. See t-statistics.

Source

the variance component.

@TYPE (the specified ISUBTOTAL= classification)

the variation of the subtotal mean measures around the grand mean.

Error

Error is the part of the total variation of the measures around their grand mean not explained by the @TYPE

Total

total variation of the measures around their grand mean

Sum-of-Squares

the variation around the relevant mean

d.f.

the degrees of freedom corresponding to the variation (= number of measures - 1)

Mean-Squares

Sum-of-Squares divided by d.f.

F-test

@TYPE Mean-Square / Error Mean-Square

Prob>F

the right-tail probability of the F-test value with (@TYPE, Error) d.f.

A probability less than .05 indicates statistically significant differences between the means.

Fixed-Effects Chi-Square (of Homogeneity)

a test of the hypothesis that all the subtotal means are the same, except for sampling error

d.f.

degrees of freedom of chi-square = number of sub-totals - 1

prob.

probability of observing this value of the chi-square or larger if the hypothesis is true. A probability less than .05 indicates statistically significant differences between the means.

inestimable

some item counts are too small and/or some variances are zero.