Search

Italian sociology has largely overlooked gambling, even if it is a widespreadphenomenon with important socio-economic redistributive consequences. Inthis paper, we aim to fill this gap exploring the relationship between socio-economicposition and propensity to gamble among Italian households in the last decade. Inthe first part, we describe the main features of legal gambling in Italy and its importancefor the State's revenues. In the second part, we review the main researchfindings from other countries, we discuss the main theories about social stratificationof gambling and elaborate our research hypotheses. In the last part, we analyse datafrom three waves of the Istat's survey on Consumption of Italian Households (1999,2003, 2008), focusing on participation and monthly expenditure in «Lotto, lotteriesand soccer gambling». Results of non-parametric (Lorentz curves, Suits index) andparametric analyses (probit and OLS regression) indicate that gambling in Italy canbe considered as a form of (voluntary) regressive taxation, because the less affluenthouseholds spend on gambling a higher proportion of their income than richerfamilies. Furthermore, the effect of income on gambling expenditure is larger amongthe lower educated families. At the end, there is evidence of a higher propensityto gamble among households of the working class, with fewer women and elderly,and from the South of Italy.

Previous research on Italian higher education showed that social origin affected students' academic progression and results in the 20th century. In this paper we examine the role of student employment - i.e. working during university - in the reproduction of social inequality in academic outcomes. In the first part, we review previous research results in the US, UK and Italy and discuss several competing hypotheses. In the second part, we use data from the Italian Longitudinal Household Survey (ILFI) to study a) the relation between student employment and academic outcomes; b) the relation between social origin and student employment, and c) the mediating effect of student employment in the relation between social origin and academic outcomes. Bivariate analysis and multinomial logistic regression models show that full-time students are more likely to graduate on time than working-students, but only high-intensity work has a detrimental effect on dropping out. Social origin affects the probability of being a high-intensity worker, but not the likelihood of being a low-intensity worker. Finally, results from a non-linear decomposition analysis suggest that the overall role of student employment in the reproduction of inequality in higher education is low, while the most important variable is the type of highschool attended (especially lyceum vs non-academic).