As we have already pointed out, in other section of this course, getting good quality results basically means three things: having good data, having good indicator, and having a good statistical economic framework that somehow fits in their theoretical concept that wants to be or that needs to be measured. This last issue in our criteria is the one that is going to be analyzed now and that is going to be developed. Of course, we cannot imagine only one situation when composing or when creating a composite indicator. Sometimes, the aim of information we have or the need of measurement is related to an indicator that these already available in terms of data, in terms of a variable. And, it exists also a theoretical framework that deals with it. Let me give you an example. For example, if you want to analyze the production of goods and services generated by the economy of a country during a year. Of course, for these type of indicator, we already have the Gross National Product or the Gross Domestic Product and their theoretical framework where we can frame it, are the national accounts. So, well-established and well-developed all over and around the world. But unfortunately, this is not always the case. It can be for example, that we want to measure some variable that is not observable. For example, the well-being or this objective well-being or some environmentally variable. In this case, we might have a framework to work with economic framework, statistical framework, but we need to define what is the target we need or what is really the variable that we want to approximate. Then will be of course, a third case, which is the case where we have no, at all. Neither a statistical, nor economic framework or theoretical framework. In this case, it's hard to recommend you from here that you undertake the work of constructing a composite indicator. Why? Because basically, in the absence or in the lackness of theoretical framework, it would be rather hard for you to understand what is the outcome of the composite indicator. Just to try to be clear about this, how to- In this section, we have trying to choose two examples that we believe that are very helpful for our purposes. This examples are related to the analysis of subjective human well-being and to the analysis and the development of the index of economic well-being. Let us start with the first of then, which is basically the analysis of individual or subjective well-being. We are going to assume in our framework of the following: We have a General Satisfaction index that we do not, of course, observe of individuals. But according to psychology and other disciplines, this General Satisfaction index can be explained, can be determined in different dimensions. The possibility that enables us that theoretical framework to establish different dimensions. And, the definition of a criteria that enables us to put any of these indicators in the corresponding dimension is very important and it's one of the key issues when elaborating composite indicators. But in this concrete case, I'm explaining about, we might think that this General Satisfaction index is a function of what? Is a function for example of health satisfaction, job satisfaction, satisfaction with the house, financial satisfaction, leisure, environment. All these are different dimensions of this General Satisfaction model or in short, this General Satisfaction Indicator. And up to certain extent, our interest is to try to look for variables, for indicators that are place in any of these different dimensions. Of course, at some point, there might be individual characteristics non-observable. And they might affect at the same point or at the same time, the General Satisfaction of the individual. And also, the different dimensions of these satisfactions. These un-observable variables need to be reflected in their statistical model that wants to analyze this General Satisfaction. But however, it is a more complicated problem and we believe it's far from the scope of this course. What we really think is interesting for you, is that the theoretical framework should enable you to decide whether a particular indicator related to health for example, is going to be a part of that health dimension or for example, the unemployment rate is going to be a part as an indicator of the job satisfaction. Okay. So then, this individualization characteristics that affect both the General Satisfaction index and the different dimension this General Satisfaction, health, financial, job are not of- how to introduce them in the model? Are not of our interest in this course because this would rather take us quite a lot of time to explain to you and we would need to introduce an additional statistical methods, statistical and econometric methods that what it is of interest for you, is the fact that the statistic, the theoretical framework should help you to assign any of these indicators to the great dimension. For example, the unemployment rate should be placed in the dimension or in the job dimension of satisfaction or some health, some variable related to the health system should be introduced in the health part of the model. In this way, we will have as a final result, a theoretical framework that would be a degree of satisfaction explained by several dimensions that at the same time are explained by some partial indicators and some heterogeneous unknown of separate variables that would be analyzed in future sessions.
