Aktuelle Probleme der Modellierung von Mehrebenen-Daten

Multilevel analyses are applied for many research problems in different research domains. They allow for adequate statistical modelling of individual characteristics embedded in contexts. The contribution reviews the present discussion of the relation between modelling and interpretation of multilevel analyses. Both the problems of data (number of cases and structure of data) and of estimation (selection of estimation procedure, validity and centering) are assessed.     

Financial Inclusion Through Public Works Program: Does Gender-Based Violence Make a Difference?

The study investigates the impact of gender-related violence on financial inclusion. Exploiting the staggered rollout of a major public works program across Indian districts, we find evidence that households in states with gender-related violence are less likely to own a bank account, notwithstanding the fact that districts in these states were early implementers of the public works program. These findings are however, different after the government launched a massive program to promote universal financial inclusion. To be more specific, we find evidence of an increase in account ownership consistent during this period by over 100 percentage points, although the use of bank account by working women declines significantly. These findings of increased access are consistent with the bargaining theory, which suggests by gaining greater resources, women also gain more power. However, low use of accounts highlights the theory of male backlash as a response to undermined economic role at home. Overall, these results suggest that any attempt at ensuring universal financial inclusion needs to adequately take on board the aspect of gender-related violence, which can be the outcome of not only economic, but also social and cultural factors.     

Maternal and Paternal Mental‐state Talk and Executive Function in Preschool Children

The present study examined the relationship between parents’ mental‐state talk and preschoolers’ executive function. Seventy‐two children participated in the present study, as well as their mothers and fathers. When children were enrolled in the second preschool year, mothers’ and fathers’ use of mental‐state references were assessed during a shared picture‐book reading task with the child. Later, four months before admission to the first grade, preschoolers’ executive function was measured. Hierarchical regression analysis revealed that maternal, but not paternal, mental‐state talk was a significant predictor of children's executive function composite, even after accounting for child gender, age, verbal ability, and parental education. When looking at each of the EF components, maternal mental‐state talk proved to be a predictor of set‐shifting whereas no significant relations emerged with inhibitory control or working memory. These findings add to prior research on parenting quality and executive function in preschoolers.     

The Role of Proactive Coping Strategies,Time Perspective,Perceived Efficacy on Affect Regulation,Divergent Thinking and Family Communication in Promoting Social Well-Being in Emerging Adulthood

The study analyzed the influence of time perspective, proactive coping strategies, perceived self-efficacy in affect regulation, divergent thinking and perceived quality of family communication on social well-being in a sample of 232 emerging adults. Social well-being showed positive correlations with proactive coping strategies, future-oriented time perspective, expression of positive emotions and regulation of negative emotions, divergent thinking, open communication with parents. A problematic family communication and a present-oriented time perspective were negatively correlated with social well-being. A forward Ridge step-wise regression model was conducted, evidencing four significant positive predictors of social well-being: proactive coping, perceived efficacy in affect regulation and open communication with parents; on the contrary, present oriented time perspective contributes significantly but negatively to social well-being.     

Variable Selection Methods in High-dimensional Regression—A Simulation Study

A challenging problem in the analysis of high-dimensional data is variable selection. In this study, we describe a bootstrap based technique for selecting predictors in partial least-squares regression (PLSR) and principle component regression (PCR) in high-dimensional data. Using a bootstrap-based technique for significance tests of the regression coefficients, a subset of the original variables can be selected to be included in the regression, thus obtaining a more parsimonious model with smaller prediction errors. We compare the bootstrap approach with several variable selection approaches (jack-knife and sparse formulation-based methods) on PCR and PLSR in simulation and real data.     

Bootstrap and Other Resampling Methodologies in Statistics of Extremes

In Statistics of Extremes, the estimation of parameters of extreme or even rare events is usually done under a semi-parametric framework. The estimators are based on the largest k-ordered statistics in the sample or on the excesses over a high level u. Although showing good asymptotic properties, most of those estimators present a strong dependence on k or u with high bias when the k increases or the level u decreases. The use of resampling methodologies has revealed to be promising in the reduction of the bias and in the choice of k or u. Different approaches for resampling need to be considered depending on whether we are in an independent or in a dependent setup. A great amount of investigation has been performed for the independent situation. The main objective of this article is to use bootstrap and jackknife methods in the context of dependence to obtain more stable estimators of a parameter that appears characterizing the degree of local dependence on extremes, the so-called extremal index. A simulation study illustrates the application of those methods.     

The Pacific Identity and Wellbeing Scale (PIWBS): A Culturally-Appropriate Self-Report Measure for Pacific Peoples in New Zealand

We describe and validate the Pacific Identity and Wellbeing Scale (PIWBS). The PIWBS is a culturally appropriate self-report measure assessing a five-factor model of Pacific identity and wellbeing. Items and construct definitions were developed through qualitative interviews, review of psychological theories, and previous research on Pacific concepts of ethnic identity and wellbeing. Exploratory and confirmatory factor analyses supported the model (Study 1 N = 143; Study 2 N = 443). The proposed five-factor model of Pacific identity and wellbeing includes scales assessing (1) Perceived Familial Wellbeing, (2) Perceived Societal Wellbeing, (3) Pacific Connectedness and Belonging, (4) Religious Centrality and Embeddedness, and (5) Group Membership Evaluation. The PIWBS provides a culturally appropriate valid and reliable assessment tool that can be used for within-cultural research for Pacific peoples from a Pacific perspective. A copy of the PIWBS and scoring instructions for its use are included.     

Social and Family Policies in Italy: Not Totally Frozen but Far from Structural Reforms

This article analyses the main shifts in the political and public discourse about families, children, elderly people, care needs and women‐friendly policies in Italy over the last two decades. It shows that while family and gender relationships have become an ideologically highly charged public issue, policies at the practical level have remained largely stagnant, marginal and fragmentary. At the same time, important institutional changes (such as the constitutional reform of 2001, which introduced a form of federalism) have created new problems of governance. The authors argue that in the face of inadequate policies, the recourse of individuals and families to old (family solidarity) and new (migrant labour) solutions may cause new tensions and inequalities.