The goal of this article consists of describing the calibration of an instrument to assess quality of life-related personal outcomes using Rasch analysis. The sample was composed of 3.029 recipients of social services from Catalonia (Spain) and was selected using a probabilistic polietapic sample design. Results related to unidimensionality, item and person separation reliability, calibration, items’ level of difficulty, response categories and differential item functioning by gender and type of collective are provided and point out the general fit of the data to the model. However, it is recommended to include more difficult items and eliminate one that seems not to be adequate
Resumen realizado por los autores recogido de la publicación