Bayesian networks of customer satisfaction survey data

A Bayesian network (BN) is a probabilistic graphical model that represents a set of variables and their probabilistic dependencies. Formally, BNs are directed acyclic graphs whose nodes represent variables, and whose arcs encode the conditional dependencies among the variables. Nodes can represent any kind of variable, be it a measured parameter, a latent variable, or a hypothesis. They are not restricted to represent random variables, which form the “Bayesian” aspect of a BN. Efficient algorithms exist that perform inference and learning in BNs. BNs that model sequences of variables are called dynamic BNs. In this context, [A. Harel, R. Kenett, and F. Ruggeri, Modeling web usability diagnostics on the basis of usage statistics, in Statistical Methods in eCommerce Research, W. Jank and G. Shmueli, eds., Wiley, 2008] provide a comparison between Markov Chains and BNs in the analysis of web usability from e-commerce data. A comparison of regression models, structural equation models, and BNs is presented in Anderson et al. [R.D. Anderson, R.D. Mackoy, V.B. Thompson, and G. Harrell, A bayesian network estimation of the service–profit Chain for transport service satisfaction, Decision Sciences 35(4), (2004), pp. 665–689]. In this article we apply BNs to the analysis of customer satisfaction surveys and demonstrate the potential of the approach. In particular, BNs offer advantages in implementing models of cause and effect over other statistical techniques designed primarily for testing hypotheses. Other advantages include the ability to conduct probabilistic inference for prediction and diagnostic purposes with an output that can be intuitively understood by managers.     

An Empirical Test of the Strategic-Grid Model of Information Systems Planning*

Contingency models of information systems planning predict that no single planning approach will suit all organizations' needs. Little empirical research has been undertaken, however, to evaluate this prediction. Accordingly, we used McFarlan, McKenney, and Pyburn's (1983) strategic-grid model to study the information systems planning problems encountered by 49 governmental agencies. Twenty-seven agencies were required to follow a planning approach best suited to organizations that had a high level of dependence on both their existing and proposed systems. We predicted that agencies not having these characteristics would encounter the most problems with the approach. The remaining 22 agencies could choose their own planning approach. We studied this latter group to determine whether the problems encountered by the first group could be attributed to the mandated approach. Overall, the empirical results obtained were equivocal. Some results indicated that more planning problems were encountered by agencies in which the mandated approach was not appropriate to their position in the strategic grid. Other results were not supportive of this proposition. More work needs to be undertaken, therefore, to evaluate the predictive and explanatory power of contingency models of information systems planning. In addition, our research indicates a need to develop more rigorous theories of information systems planning behaviors, to improve the instruments needed to measure these behaviors, to explore the relationship between information systems planning behaviours and organizational effectiveness, to investigate how organizational learning impacts planning behaviors, and to determine the types of information systems planning problems that diffuse through organizations and those that remain localized.     

An Explanation of the Occupational Structure of Sectors of Industry

Abstract. This study has been carried out by the Research Centre for Education and Labour Market (ROA) as part of a long-range research project commissioned by the Ministry of Education and Sciences. The objective of the project is to develop an education/labour-market information system to help secondary and university students choose the right type of school and/or occupation (ROA 1988). The study reported on here sets out to disclose the factors underlying the development of the occupational pattern. The shares of occupational classes in each economic sector are taken as points of departure. The explanatory variables are technological progress and cyclical and structural sectoral developments. For each occupational class, an estimation is made with the help of a linear model, the estimation method being that of Weighted Least Squares. To permit several variables to be included in the equation, the data have been pooled over sectors and time. In two fifths of all cases, a model with explanatory variables has been chosen; in the remaining instances a sector-dummy model was found to suffice.