Responses:          Diagnosis             INFIT OUTFIT

 Easy--Items--Hard   Pattern               MnSq    MnSq

--------------------------------------------------------

 111|0110110100|000  Modelled/Ideal         1.1     1.0

 000|0000011111|111  Miscode                4.3    12.6

 011|1111110000|000  Carelessness/Sleeping  1.0     3.8

 111|1111000000|001  Lucky Guessing         1.0     3.8

 101|0101010101|010  Response set/Miskey    2.3     4.0

 111|1000011110|000  Special knowledge      1.3     0.9

 111|1111100000|000  Guttman/Deterministic  0.5     0.3

 111|1010110010|000  Imputed outliers *     1.0     0.6

--------------------------------------------------------

Right|Transition|Wrong

 high - low - high   OUTFIT sensitive to outlying observations

                     >>1.0 unexpected outliers

                     <<1.0 overly predictable outliers

 

  low - high - low   INFIT sensitive to pattern of inlying observations

                     >>1.0 disturbed pattern

                     <<1.0 Guttman  pattern

 

* as when a tailored test is filled out by imputing all "right" response to easier items and all "wrong" to harder items.

 

The exact details of these computations have been lost, but the items appear to be uniformly distributed about 0.4 logits apart, from Linacre, Wright (1994) Rasch Measurement Transactions 8:2 p. 360

 

The Z-score standardized statistics report, as unit normal deviates, how likely it is to observe the reported mean-square values, when the data fit the model.