Scrutinizing the Smart City Index: a multivariate statistical approach

The smart city represents a frequently elaborated concept which however comes short in delivering a consistent definition. Nevertheless, almost every description has always been oriented to its technological component, sustainable development policies, and enabling high capacities for learning and innovation. Moreover, the smart city aims at connecting people, information and other city elements using state-of-the-art technologies. As a result, it creates a sustainable, greener city, pushes forward competitive and innovative commerce, and increases overall life quality. The integrated view of a smart city underlines it does not operate in isolation, which is why every subsystem of a city needs to develop its smart component. A wide range of rankings is used to determine the smartness of cities by mapping out the pros and cons of each analysed city. As the way to integrate various indicators into one value which will represent the rank, a composite index approach is most frequently used. Still, composite indexes are usually formed using the equal weight approach, which is heavily criticised in current literature. In this paper, we try to provide added value to the Smart City Index by implementing the statistical post hoc I-distance approach. The procedure enables us to shed some additional light on the issue of sensitivity of cities’ rank. The application of post hoc I-distance defines indicators which are most significant for the ranking process. It consequently empowers city decision-makers to improve their performance, with a focus on those particular indicators.