A clustering method for global optimization based on thekth nearest neighbour

In this paper we describe a stochastic method for global optimization based on a uniform sampling in the search domain. After a reduction of the sample, computing the distance between the remaining points and using the distribution of the kth nearest neighbour enables clusters of points to be built up, hopefully fitting the regions of attraction of significant local optima; from each of these a local search is started. The properties of the method are analysed, and detailed computational results on standard test functions are provided.     

Forthcoming in Sociological Forum

Forthcoming in Sociological Forum     

A comment on “A theoretical basis for perturbation methods” by Krishnamurty Muralidhar and Rathindra Sarathy

Statistics and Computing -     

Maternal Working Hours and the Well-Being of Adolescent Children: Evidence from British Data

This study investigates how maternal working hours are related to various outcomes in children aged 11–15 using a sample of mothers and adolescents in the British Household Panel Survey. Research that examines the effects of maternal employment on children has been motivated by the rapid increase of female participation rates in the labour market and increased shares of children living in female-headed or single-mother households. The existing literature on this issue is very limited, mostly based on American data, and provides conflicting results. Fixed effects have been used in the present analysis to control for characteristics of children and mothers that do not vary over time. The results suggest that full-time maternal employment (as opposed to part-time) has little or no effect on the propensity of adolescents to smoke, their life satisfaction, self-esteem, or intention to leave school at 16. These results are stable and consistent across various specifications of the model and different socio-economic status.     

Representativeness in panel surveys

ABSTRACT

A panel is a set of units recruited and used in successive surveys. When the sample unit is the household, so-called R-indicators together with the comparison of distributions of certain variables to those known in the total population help to measure the representativeness of the panel. The method is applied to Understanding Society, a U.K. household longitudinal study. At each wave, under- and over-represented groups of individuals are identified. This allows the implementation of better survey designs and procedures to reduce the bias of nonresponse.     

Determinants of Joblessness During the Economic Crisis: Impact of Criminality in the Italian Labour Market

After nearly 20 years without any significant changes, the indicators of the European Regional Development Fund (ERDF) have undergone a major overhaul in the current 2014–2020 programming period. This paper reviews the characteristics of the ERDF indicator system evolution and contextualises the changes. Based on applying a logic model and practical examples, we argue that in spite of the good justification for changes, the new ERDF intervention logic and indicator system did not solve some of the well-known problems and brought new challenges for the ERDF-supported programmes. The main challenges that still remain to be solved are: differing indicator concepts between the European Union funds; inconsistency of the common output indicators; difficulties in establishing a programme’s contribution to results; persistent problems in determining the target values for results; and broad result indicators with indirect link to interventions.     

Gambling Disorder: Exploring Pre-treatment and In-treatment Dropout Predictors. A UK Study

The aim of this study was to identify predictors of treatment dropout in a sample of gamblers attending a specialist clinic for gambling disorder. We analysed data on 846 treatment-seeking pathological gamblers. Firstly, we investigated differences in socio-demographic and clinical variables between treatment completers and pre-treatment dropouts, as well as between treatment completers and during-treatment dropouts. Subsequently, variables were entered into a multinomial logistic regression model to identify significant predictors of pre-treatment and in-treatment dropout. Overall, 44.8% of clients did not complete the treatment: 27.4% dropped out before starting it, while 17.4% dropped out during the treatment. Younger age and use of drugs were associated with pre-treatment dropout, while family history of gambling disorder, a lower PGSI score, and being a smoker were related with in-treatment dropout. Our findings suggest that pre-treatment dropouts differ from in-treatment dropouts, and, thus, further research will benefit from considering these groups separately. In addition, this newly gained knowledge will also be helpful in increasing treatment retention in specific subgroups of problem gamblers.     

The Role of Secularism of State on the Relationship Between Catholic Identity,Political Orientation,and Gay Rights Issues

In discussions of regulations governing same-sex marriage and adoption by gays and lesbians, the issue of state secularism is often called into question. This study aims to test the mediating effects of state secularism on the relationship between Catholic identity, political orientation, and gay civil rights. Participants were Catholic Italians who completed a questionnaire measuring the constructs under investigation. Results showed that state secularism mediates the effects of Catholic identity and political orientation on attitudes toward same-sex marriage and adoption by gays and lesbians.