Rasch Analysis of the Power as Knowing Participation in Change Tool--the Brazilian version.

OBJECTIVES the objective of this study was to evaluate the items contained in the Brazilian version of the Power as Knowing Participation in Change Tool (PKPCT). METHOD investigation of the psychometric properties of the mentioned questionnaire through Rasch analysis. RESULTS the data from 952 nursing assistants and 627 baccalaureate nurses were analyzed (average age 44.1 (SD=9.5); 13.0% men). The subscales Choices, Awareness, Freedom and Involvement were tested separately and presented unidimensionality; the categories of the responses given to the items were compiled from 7 to 3 levels and the items fit the model well, except for the following/leading item, in which the infit and outfit values were above 1.4; this item has also presented Differential Item Functioning (DIF) according to the participant's role. The reliability of the items was of 0.99 and the reliability of the participants ranged from 0.80 to 0.84 in the subscales. Items with extremely high levels of difficulty were not identified. CONCLUSIONS the PKPCT should not be viewed as unidimensional, items with extremely high levels of difficulty in the scale need to be created and the differential functioning of some items has to be further investigated.