Dimension reduction for model-based clustering

We introduce a dimension reduction method for visualizing the clustering structure obtained from a finite mixture of Gaussian densities. Information on the dimension reduction subspace is obtained from the variation on group means and, depending on the estimated mixture model, on the variation on group covariances. The proposed method aims at reducing the dimensionality by identifying a set of linear combinations, ordered by importance as quantified by the associated eigenvalues, of the original features which capture most of the cluster structure contained in the data. Observations may then be projected onto such a reduced subspace, thus providing summary plots which help to visualize the clustering structure. These plots can be particularly appealing in the case of high-dimensional data and noisy structure. The new constructed variables capture most of the clustering information available in the data, and they can be further reduced to improve clustering performance. We illustrate the approach on both simulated and real data sets.     

An Assessment of the Psychometric Properties of Italian Version of CPGI

The aim of this study was to adapt to the Italian context a very commonly used international instrument to detect problem gambling, the canadian problem gambling index (CPGI), and assess its psychometric properties. Cross-cultural adaptation of CPGI was performed in several steps and the questionnaire was administered as a survey among Italian general population (n = 5,292). Cronbach’s alpha reliability coefficient was 0.87 and can be considered to be highly reliable. Construct validity was assessed first by means of a principal component analysis and then by means of confirmatory factor analysis, showing that only one factor, problem gambling, was extracted from the CPGI questionnaire (an eigenvalues of 4,684 with percentage of variance 52 %). As far as convergent validity is concerned, CPGI was compared with Lie/Bet questionnaire, a two-item screening tool for detecting problem gamblers, and with both depression and stress scales. A short form DSM-IV CIDI questionnaire was used for depression and VRS scale, a rating scale, was used for rapid stress evaluation. A strong convergent validity with these instruments was found and these findings are consistent with past research on problem gambling, where another way to confirm the validity is to determine the extent to which it correlates with other qualities or measures known to be directly related to problem gambling. In sum, despite the lack of a direct comparison with a classic gold-standard such as DSM-IV, the Italian version of CPGI exhibits good psychometric properties and can be used among the Italian general population to identify at-risk problem gamblers.     

The Geography of Economics and Happiness: Spatial Patterns in the Effects of Economic Conditions on Well-Being

This paper investigates the cross-country distribution of the relationship between economic conditions and well-being. Using a large sample of individuals from 94 countries worldwide, we find that the effect of income on well-being is larger in countries with lower GDP per capita, while the negative effect of being unemployed is stronger in countries with higher unemployment rate or higher GDP per capita. Interestingly, the effect of being unemployed displays positive spatial dependence across countries that is not accounted for by aggregate socio-economic conditions. Overall, the results indicate that geography, culture and institutions must be explicitly taken into account in order to understand the relationship between economic conditions and well-being.     

Rethinking and Revising Goode’s Contribution to Global Family Change

Abstract

This essay provides a comprehensive overview of William J. Goode’s contribution to the study of global family and social change. I begin by describing Goode’s theoretical perspectives and outlining his theses dating back to the 1960s. I then provide an assessment of where and why some of his predictions proved wrong and elaborate on what we have learnt on changes in families at the global level over the past half century. Lastly, I speculate on how Goode would rethink his arguments nowadays in light of fifty years of new evidence and scholarly developments – both theoretical and methodological. In so doing, I highlight shortcomings of current approaches and outline directions for future family research and theorizing.     

Complex Factors and Behaviors in the Gambling Population of Italy

Gambling has seen significant growth globally, and particularly in Italy: it has rapidly evolved from a simple recreational activity to represent 4% of Italian GDP in 2010.A sample of 4.494 gamblers was drawn from IPSAD-Italia^®2007-2008 (Italian Population Survey on Alcohol and Drugs) in order to examine different gambling patterns (assessed using the Canadian Problem Gambling Index Short form scale).Separate analysis was performed on young adults, age 15–24 (n = 1,241; male 56.2%), and adults, age 25–64 (n = 3,253; male 53.8%): compared with adults, Italian youth, although they gambled less (35.7% vs. 45.3%), appeared to have higher prevalence of low risk gambling (6.9% vs. 5.8%) and moderate risk or problem gambling (2.3% vs. 2.2%). Males are more likely to be moderate-risk or problem gamblers. Those with only a primary education are more likely to be moderate-risk or problem gamblers (young adults: RRR = 5.22; adults: RRR = 3.23) than those with a university education, just like those youth who use depressants, but only among younger (RRR = 3.38).A fundamental issue, “do not disapprove of gambling”, seems to relate to problematic gambling: a specific Italian legislation, the Abruzzi Decree Law, could have influenced the perception that gambling may contribute positively to provide additional funds to the government for social good as well as to add needed jobs. Regardless of such potential social benefits, gambling is a social epidemic and if this association should be confirmed by more focused studies, policy makers should evaluate ways to affect this perception as soon as possible.     

Approximating distributions of random functionals of Ferguson-Dirichlet priors

We explore the possibility of approximating the Ferguson-Dirichlet prior and the distributions of its random functionals through the simulation of random probability measures. The proposed procedure is based on the constructive definition illustrated in Sethuraman (1994) in conjunction with the use of a random stopping rule. This allows us to set in advance the closeness to the distributions of interest. The distribution of the stopping rule is derived, and the practicability of the simulating procedure is discussed. Sufficient conditions for convergence of random functionals are provided. The numerical applications provided just sketch the idea of the variety of nonparametric procedures that can be easily and safely implemented in a Bayesian setting.     

Can we declare military Keynesianism dead?

This paper empirically tests the Keynesian hypothesis that government defence spending positively impacts on aggregate output, by using a long-run equilibrium model for the US and the UK.Our contribution, with respect to previous works, is twofold. First, our inferences are adjusted for structural breaks exhibited by the data concerning fiscal and monetary variables. Second, we take into account different dynamics between defence spending on aggregate output, showing that the results are sensitive to sub-sample choices. Though the estimated elasticities in both countries show a lack of significance in the more recent years of the sample, defence-spending priorities addressed to international security may revitalize pro-cyclical effects in the UK, by an industrial policy of defence shared with the EU members.     

The Internetization of International Migration

The Internet has revolutionized our economies, societies, and everyday lives. Many social phenomena are no longer the same as they were in the pre‐Internet era: they have been “Internetized.” We define the Internetization of international migration, and we investigate it by exploring the links between the Internet and migration outcomes all along the migration path, from migration intentions to actual migration. Our analyses leverage a number of sources, both at the micro‐ and the macro‐level, including the Gallup World Poll, the Arab Barometer, data from the International Telecommunication Union, the Italian population register, and unique register data from a migrant reception center in Southern Italy. We also distinguish between economic migrants—those who leave their country of origin with the aim of seeking better economic opportunities elsewhere—and political migrants—those who are forced to leave their countries of origin for political or conflict‐related reasons. Our findings point to a consistently positive relationship between the diffusion of the Internet, migration intentions, and migration behaviors, supporting the idea that the Internet is not necessarily a driving force of migration per se, but rather an enabling “supportive agent.” These associations are particularly relevant for economic migrants, at least for migration intentions. Further analyses underscore the importance of the Internet in providing a key informational channel which helps to define clearer migration trajectories.     

Factor Structure and Psychometric Properties of the Italian Version of the Homosexuality Scale of the Trueblood Sexual Attitudes Questionnaire

The aim of the present study was to translate the Homosexuality scale of the Trueblood Sexual Attitudes Questionnaire into the Italian language and to assess its factor structure and psychometric properties in Italian psychology students. The questionnaire was originally developed and validated in U.S. college students, and later in Turkish social work students, showing high internal consistency. It measures attitudes toward several sexual practices and behaviors, regarding self and others. Particularly, the Homosexuality scale measures attitudes toward different sexual and romantic practices with people of the same sex. A total of 199 Italian psychology students participated to the study, and they were administered the Italian translation of the scale. We applied exploratory factor analysis and confirmatory factor analysis. Results showed that the scale has high internal consistency, and that the original two-factor model accounting for attitudes toward self and others fits the data well. Implications for education and assessment in student populations are discussed.     

Ranking and Prioritization of Emergency Departments Based on Multi-indicator Systems

In this work we discuss how Emergency Departments (EDs) can be ranked on the basis of multiple indicators. This problem is of absolute relevance due to the increasing importance of EDs in regional healthcare systems and it is also complex as the number of indicators that have been proposed in the literature to measure ED performance is very high. Current literature faces this problem using synthetic (or numerically aggregated) indicators of a set of performance measures but, although simple, this solution has a number of drawbacks that make this choice inefficient: a compensation effect among the indicators; a high degree of subjectivism in the indicators weighting; opacity in the decision making; all the EDs are considered to be comparable. Indeed, the situations in which EDs are comparable (i.e. when all the performance of one ED are not lower than the performance indicators of the other) are a minority and incomparability is by itself a source of information that should be used to identify situations for which different policy actions should be designed. In this work we propose to use non compensatory composite indicators and partial ordering theory to rank and compare EDs giving value to the reasons of such an incomparability. These methods are applied on a case study of 19 EDs in an administrative region in Italy.