A Cluster Analysis of Multidimensional Poverty in Switzerland
24 Pages Posted: 25 Jul 2006
Date Written: July 2006
The measurement of poverty has often been criticized for relying solely on measures of financial deprivation. Poverty being a multidimensional state, related to health, schooling, living environment, psychological state as well as social tides, care should be taken to integrate these various components to have a proper picture of poverty. This is especially true for rich countries where poor financial conditions are often alleviated by social policies like minimum income, unemployment or housing benefits. Social exclusion and poor health can therefore dominate the poverty feeling.
We illustrate how some descriptive statistical tools can offer new insights in the context of multidimensional poverty. Factor analysis is used in a first step to construct poverty indicators based on many possible dimensions without posing too many a priori restrictions. The base variables are thus combined to produce common factors which convey some aspect of multidimensional poverty. By ascribing individual scores on each factor, we then use cluster analysis to determine population's subgroups that are unevenly affected by the various dimensions of poverty, what allows us to identify the poor. Finally, a logit regression is run to find the determinants of poverty.
Keywords: Multidimensional poverty, Factor analysis, Cluster analysis, Switzerland
JEL Classification: I32, C25, C23
Suggested Citation: Suggested Citation