A two-step procedure to analyse users' satisfaction

In this paper an integrated use of Nonlinear Principal Component Analysis (NLPCA) and Multilevel Models (MLM) for the analysis of satisfaction data is proposed. The basic hypothesis is that observed ordinal variables describe different aspects of a latent continuous variable that depends on individual and contextual covariates. NLPCA is used to measure the level of a latent variable and MLM are adopted for detecting individual and environmental determinants of the level. By using the Eurobarometer survey data, this approach is applied to analyse the European users' satisfaction with services of general interest after the recent privatisation and liberalisation policies.