The Similar Faces of Swiss Working Poor - An Empirical Analysis across Swiss Regions using Logistic Regression and Classification Trees

Fabio Losa, Emiliano Soldini
Swiss Journal of Economics and Statistics / Schweizerische Zeitschrift für Volkswirtschaft und Statistik / Revue Suisse d'Economie politique et de Statistique, Volume 147, Issue 1, 2011, Pages 17-44
Download Browse issue

Keywords

working poor, classification trees, logistic regression

Abstract

This paper is an empirical investigation into working poverty across the seven statistical regions in Switzerland. By analysing, in a comparative setting, the determinants of the probability of falling below the poverty line and the resulting social groups at risk, the paper aims at shedding light on the existence of regional patterns and peculiarities which could call for regional policies rather than nationwide ones. By using data from the Swiss Labour Force Survey of 2006, and adopting a definition of working poor as "people aged between 20 and 59 years who are working and who live in a working and poor household" and an absolute poverty line, data are analysed by two complementary methods - logistic regression and classification trees -A in order to better assist policymakers in understanding the issues at stake and in the subsequent decision-making process.