In this contribution, rating data are modelled using the new class of Nonlinear CUB models, recently introduced to generalize the standard CUB. Both CUB and Nonlinear CUB can be viewed as special cases of a general model explaining the cognitive mechanism driving the individuals' responses on a Likert-type scale. A very interesting feature of the Nonlinear CUB is given by the so-called transition probabilities, useful to understand and describe the individual's mental stance towards the response scale. This and all the other nice features of Nonlinear CUB models are presented by an application to real data coming from a survey aimed at evaluating the parents' satisfaction with the service delivered by public kindergartens in Brescia, Italy.
Nonlinear CUB models
MANISERA, Marica;ZUCCOLOTTO, Paola
2013-01-01
Abstract
In this contribution, rating data are modelled using the new class of Nonlinear CUB models, recently introduced to generalize the standard CUB. Both CUB and Nonlinear CUB can be viewed as special cases of a general model explaining the cognitive mechanism driving the individuals' responses on a Likert-type scale. A very interesting feature of the Nonlinear CUB is given by the so-called transition probabilities, useful to understand and describe the individual's mental stance towards the response scale. This and all the other nice features of Nonlinear CUB models are presented by an application to real data coming from a survey aimed at evaluating the parents' satisfaction with the service delivered by public kindergartens in Brescia, Italy.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.