|
MNSQ= show mean-square instead of t-standardized fit statistics = Yes |
Top Up Down A A |
|
The mean-square or t standardized fit statistics are shown in Tables 7, 11 to quantify the unexpectedness in the response strings, and in Tables 4, 5, 8, 9 for the fit plots.
MNSQ=N Show standardized (ZSTD) fit statistics. ZSTD (standardized as a z-score) is used of a t-test result when either the t-test value has effectively infinite degrees of freedom (i.e., approximates a unit normal value) or the Student's t-statistic distribution value has been adjusted to a unit normal value.
MNSQ=Y Show mean-square fit statistics. Use LOCAL=L for log scaling.
TABLE 7.1 TABLE OF POORLY FITTING PERSONS ( ITEMS IN ENTRY ORDER) NUMBER - NAME -- POSITION ------ MEASURE - INFIT (MNSQ) OUTFIT
17 Rod M -1.41 2.4 A 2.2 RESPONSE: 1: 0 0 2 4 1 4 3 1 3 3 1 4 3 2 3 3 1 4 2 1 Z-RESIDUAL: -2 2 -2 -2 2 -2
Mean-square: TABLE 9.1 -5 -4 -3 -2 -1 0 1 2 3 ++-------+-------+-------+-------+-------+-------+-------+-------++ 2 + | + 2 I | | | T | | | E | | B A | M | | FDE C | 1 +-------------------------------|-----iIJg-Hj------G--------------+ 1 O | | e d f | U | c| b | T | | a h | F | | | I 0 + | + 0 T ++-------+-------+-------+-------+-------+-------+-------+-------++ -5 -4 -3 -2 -1 0 1 2 3 ITEM MEASURE
t standardized ZSTD: -5 -4 -3 -2 -1 0 1 2 3 ++-------+-------+-------+-------+-------+-------+-------+-------++ 2 | | A | | | B | I | | | T 1 + | C + 1 E | | DE | M | | F | | | H | O 0 +-------------------------------|------IJ---j------G--------------+ 0 U | | i | T | | g h | F | | f | I -1 + | + -1 T | | e d | | c| | Z | | | S -2 +-------------------------------|---------------------------------+ -2 T | | | D | | b | | | | -3 + | + -3 | | a | | | | | | | -4 + | + -4 ++-------+-------+-------+-------+-------+-------+-------+-------++ -5 -4 -3 -2 -1 0 1 2 3 ITEM MEASURE |
Help for WINSTEPS® Rasch Measurement Software: www.winsteps.com.