Reconciling internal and external satisfaction through probabilistic graphical models: An empirical study
International Journal of Quality and Service Sciences
ISSN: 1756-669X
Article publication date: 18 September 2017
Abstract
Purpose
This paper aims to holistically reconcile internal and external customer satisfaction using probabilistic graphical models. The models are useful not only in the identification of the most sensitive factors for the creation of both internal and external customer satisfaction but also in the generation of improvement scenarios in a probabilistic way.
Design/methodology/approach
Standard Bayesian networks and object-oriented Bayesian networks are used to build probabilistic graphical models for internal and external customers. For each ward, the model is used to evaluate satisfaction drivers by category, and scenarios for the improvement of overall satisfaction variables are developed. A global model that is based on an object-oriented network is modularly built to provide a holistic view of internal and external satisfaction. The linkage is created by building a global index of internal and external satisfaction based on a linear combination. The model parameters are derived from survey data from an Italian hospital.
Findings
The results that were achieved with the Bayesian networks are consistent with the results of previous research, and they were obtained by using a partial least squares path modelling tool. The variable ‘Experience’ is the most relevant internal factor for the improvement of overall patient satisfaction. To improve overall employee satisfaction, the variable ‘Product/service results’ is the most important. Finally, for a given target of overall internal and external satisfaction, external satisfaction is more sensitive to improvement than internal satisfaction.
Originality/value
The novelty of the paper lies in the efforts to link internal and external satisfaction based on a probabilistic expert system that can generate improvement scenarios. From an academic viewpoint, this study moves the service profit chain theory (Heskett et al., 1994) forward by delivering operational guidelines for jointly managing the factors that affect internal and external customer satisfaction in service organizations using a holistic approach.
Keywords
Citation
Musella, F., Guglielmetti Mugion, R., Raharjo, H. and Di Pietro, L. (2017), "Reconciling internal and external satisfaction through probabilistic graphical models: An empirical study", International Journal of Quality and Service Sciences, Vol. 9 No. 3/4, pp. 347-370. https://doi.org/10.1108/IJQSS-02-2017-0007
Publisher
:Emerald Publishing Limited
Copyright © 2017, Emerald Publishing Limited