A Multidimensional Measure of Social Stratification: Composite Indicator of Inequality

In previous works (Fachelli, 2009; Fachelli & López-Roldán, 2010, 2012) we analysed the changes in the social and economic model of Argentina in the period 1997-2010, we observed that the social structure, measured by a variable that typifies stratification, remains basically unchanged during the years analysed in spite of the profound crisis that occurred at the end of 2001 and along 2002. Within this general framework, we developed a multidimensional model where are combined 4 dimensions (work, education, housing and income), 6 variables and 38 categories. With them we obtain a typology with four basic types. The methodology we apply is called "structural and articulated typology": we use factorial analysis, including multiple correspondence analysis (MCA), in order to reduce the initial attribute space into "m" factors. Then we apply cluster analysis to classify individuals into “key” groups or "types" on these factors.

Our goal is to obtain a single measure of social inequality. However, the multidimensionality of the concept of social stratification makes difficult to obtain a complete ordering of households according to “inequality” level. The question that arise is how generate a composite indicator, that we call "social distance", based in a multidimensional measure taking into account the principal factorials components. We follow the work of Asselin (2009) who applies this strategy in the context of the analysis of multidimensional poverty. The communication will presents the results based on the analysis of the Labor Force Survey of Argentina in 2012.