Analysing the Relationship Between Immigrant Status and the Severity of Offending Behaviour in Terms of Individual and Contextual Factors.

Background: Social inclusion is a context for both risk and protective factors of migrant youth delinquency. This study aims to shed light on the issue by comparing delinquency amongst native, first-generation, and second-generation immigrant youths in Portugal, a country located in the south of Europe, an area where research in this field is still scarce. Methods: The research is based on the International Self-Reported Delinquency (ISRD-3) dataset, which includes information on over 4,000 adolescents, who self-reported on their socio-demographic status, leisure activities, school and neighbourhood environment, family bonds, and self-control. Results: Nested Logistic Regression analyses showed that a young first-generation immigrant is twice as likely to commit a crime, with or without violence, as a young native born in Portugal. However, no differences were found regarding the prevalence of delinquency amongst second-generation immigrants and natives, which is likely due to the integration and cultural assimilation of the immigrant over time. Regarding the analysed risk factors, it was found that both structural and individual factors, identified by the theories of control, stress, as well as situational action theory, have a direct effect on the commission of juvenile crimes (both non-violent and violent). Moreover, this effect is significant in adolescents living in Portugal in general, both immigrants and natives. The most influential variable for both types of delinquent behaviour, with and without violence, is peer delinquency, followed by low morality and self-control. Conclusion: These findings have relevant policy implications and are useful for evidence-based interventions aimed at promoting migrant adolescent well-being and targeting host countries' performance.

Cross Cultural Analysis of Direct Employee Participation: Dealing With Gender and Cultural Values.

The goal of this study is analyze the influence of perceived supervisor support (PSS) by employees at a micro level and the role of the cultural values of "power distance" and "masculinity" at a macro level on direct employee participation in decision-making (PDM). Furthermore, the influence of the gender of managers and employees is taken into account. The analysis is based upon the Sixth European Working Conditions Survey carried out by Eurofound in 2016. The results of a Hierarchical linear model indicate that all predictors significantly influenced PDM; PSS positively and cultural values negatively. When the gender of managers and employees is considered, the findings suggest that PSS has a larger impact on PDM when male managers address female employees. Regarding the moderating effect of PSS on cultural values, it is shown that masculinity and power distance lose importance when employees have the support of their supervisors.

Global Sensitivity Estimates for Neural Network Classifiers.

Artificial neural networks (ANNs) have traditionally been seen as black-box models, because, although they are able to find ``hidden'' relations between inputs and outputs with a high approximation capacity, their structure seldom provides any insights on the structure of the functions being approximated. Several research papers have tried to debunk the black-box nature of ANNs, since it limits the potential use of ANNs in many research areas. This paper is framed in this context and proposes a methodology to determine the individual and collective effects of the input variables on the outputs for classification problems based on the ANOVA-functional decomposition. The method is applied after the training phase of the ANN and allows researchers to rank the input variables according to their importance in the variance of the ANN output. The computation of the sensitivity indices for product unit neural networks is straightforward as those indices can be calculated analytically by evaluating the integrals in the ANOVA decomposition. Unfortunately, the sensitivity indices associated with ANNs based on sigmoidal basis functions or radial basis functions cannot be calculated analytically. In this paper, the indices for those kinds of ANNs are proposed to be estimated by the (quasi-) Monte Carlo method.