A proposal for measuring global subjective satisfaction

Abstract

In many fields of social studies it would be useful to measure the degree of global satisfaction (or agreement) of a population based on individual opinions about several aspects. Some examples regard, for instance, the measurement of customer satisfaction, of students’ or graduates’ opinions about university teaching, of subjective well-being and so on. This problem concerns the more general framework of measuring complex phenomena that are not directly observable and hence approximated by a set of elementary (one-dimensional) indicators. The problem of synthesizing a set of variables thus arises. A common solution is to construct a composite indicator, which synthesizes all the elementary indicators by means of a weighted average. There is a widely developed statistical literature on this subject. Here we propose an alternative approach. We consider a set of elementary variables that are observed in a population. Each variable represents the individual degree of satisfaction about one of several aspects related to the complex phenomenon. In aiming to synthesize them, starting from their joint distribution, we define a new variable that measures global individual satisfaction. To evaluate the degree of global satisfaction for the entire population, we propose a normalized index, which takes values between 0 (maximum dissatisfaction) and 1 (maximum satisfaction). Specifically, this index measures the distance between the observed distribution of the new variable and the theoretical one, which refers to the situation of maximum dissatisfaction (all individuals are not at all satisfied for every aspect).