Statistical techniques for continuous improvement: a citizen's satisfaction survey

Abstract

PurposeThe purpose of this paper is to propose a path analysis of data coming from a citizen's satisfaction survey to support decision makers in quality service improvement. In detail, the survey aims to measure citizen's satisfaction of an Italian local Public Administration regarding the “infant school (0‐6 years) enrollment service”.Design/methodology/approachThe survey represents an experimentation of an original model measuring customers' satisfaction toward on‐line services. Some statistical methods to analyse a given dataset from different points of view are selected.FindingsOutcomes of descriptive statistics as well as of multivariate data analysis to summarize information variables are presented. A new multivariate statistical technique, Probabilistic Expert Systems (PES) (Cowell et al.), is proposed to simulate corrective actions (scenarios) and to suggest the best one for the service quality improvement.Originality/valueThe paper shows that statistical methods are able to support the decisional process because they allow the development of information (gathered from survey) into know‐how. However, managers need to join together both statistical information and experience by means of a systematic method, in order to take effective decisions.