A Regression Model for the AUC of Clustered Ordinal Test Results and Working Independent Optimal Weights

We study a regression model on the area under the receiver operating characteristic curves (AUC) for clustered (or repeatedly measured) test results. To account for cluster information, we consider a weighted estimating equation for Dodd and Pepe (2003 Dodd , L. , Pepe , M. ( 2003 ). Semiparametric regression for the area under the receiver operating charateristic curve . Journal of the American Statistical Association 98 : 409 – 417 .[Taylor & Francis Online], [Web of Science ®] , [Google Scholar])'s regression model with working independence weights. We find the optimal weight in the given class of working independence weights to minimize the variance (or MSE) of regression estimators. We apply the proposed procedure to analyzing our recent experiment on diagnosing a liver disorder. In this experiment, we investigated MRI images of patients having symptoms of potential liver disorder to compare the performance of different MRI picturing methods in testing for liver disorders.     

Intensive family preservation services in Flanders: an outcome study

Set up as a one-group pre-test–post-test follow-up design, this study examines the outcomes of intensive family preservation services (IFPS) in Flanders addressing the following research questions: Does IFPS make a significant difference concerning (1) the interaction between parent and child? (2) social support as experienced by the parent? (3) stress and empowerment of the parent? and (4) the strengths and difficulties of the minor? Families seem to indicate that their family situation is more manageable, although still, professional help is needed. Parents feel more resilient about the situation and indicate that they can get along with their child better than before the crisis intervention. Parents still feel uncertain about the future and about their role as a parent. There seems to be no indications that their level and quality of social contacts have changed. Further research is suggested to provide more in-depth information about the conditions associated with processes of change during an IFPS intervention.     

Regularized Gaussian belief propagation

Belief propagation (BP) has been applied in a variety of inference problems as an approximation tool. BP does not necessarily converge in loopy graphs, and even if it does, is not guaranteed to provide exact inference. Even so, BP is useful in many applications due to its computational tractability. In this article, we investigate a regularized BP scheme by focusing on loopy Markov graphs (MGs) induced by a multivariate Gaussian distribution in canonical form. There is a rich literature surrounding BP on Gaussian MGs (labelled Gaussian belief propagation or GaBP), and this is known to experience the same problems as general BP on graphs. GaBP is known to provide the correct marginal means if it converges (this is not guaranteed), but it does not provide the exact marginal precisions. We show that our adjusted BP will always converge, with sufficient tuning, while maintaining the exact marginal means. As a further contribution we show, in an empirical study, that our GaBP variant can accelerate GaBP and compares well with other GaBP-type competitors in terms of convergence speed and accuracy of approximate marginal precisions. These improvements suggest that the principle of regularized BP should be investigated in other inference problems. The selection of the degree of regularization is addressed through the use of two heuristics. A by-product of GaBP is that it can be used to solve linear systems of equations; the same is true for our variant and we make an empirical comparison with the conjugate gradient method.     

Can Reporting Heterogeneity Explain Differences in Depressive Symptoms Across Europe?

Depression is one of the leading causes of disability in the developed world. Previous studies have shown varying depression prevalence rates between European countries, and also within countries, between socioeconomic groups. However, it is unclear whether these differences reflect true variations in prevalence or whether they are attributable to systematic differences in reporting styles (reporting heterogeneity) between countries and socioeconomic groups. In this study, we examine the prevalence of three depressive symptoms (mood, sleeping and concentration problems) and their association with educational level in 10 European countries, and examine whether these differences can be explained by differences in reporting styles. We use data from the first and second waves of the COMPARE study, comprising a sub-sample of 9,409 adults aged 50 and over in 10 European countries covered by the Survey of Health, Ageing and Retirement in Europe. We first use ordered probit models to estimate differences in the prevalence of self-reported depressive symptoms by country and education. We then use hierarchical ordered probit models to assess differences controlling for reporting heterogeneity. We find that depressive symptoms are most prevalent in Mediterranean and Eastern European countries, whereas Sweden and Denmark have the lowest prevalence. Lower educational level is associated with higher prevalence of depressive symptoms in all European regions, but this association is weaker in Northern European countries, and strong in Eastern European countries. Reporting heterogeneity does not explain these cross-national differences. Likewise, differences in depressive symptoms by educational level remain and in some regions increase after controlling for reporting heterogeneity. Our findings suggest that variations in depressive symptoms in Europe are not attributable to differences in reporting styles, but are instead likely to result from variations in the causes of depressive symptoms between countries and educational groups.     

Ideology,Practice, and Process? A Review of the Concept of Managerialism in Civil Society Studies

Managerialism is today a frequently applied concept in studies of how ideas and practices related to corporate management are diffused in society. Some assert that managerialism even is what mostly affects the development of contemporary civil society organizations. It is, however, far from clear how the concept of managerialism is used and defined across interest fields. The main conclusion in the present review, involving 105 peer-reviewed articles in civil society studies published between 1990 and 2014, is that the concept of managerialism is so broadly defined that it runs the risk of losing its analytical powers. To avoid this, the paper argues for a more precise conceptual use and suggests that the concept of managerialism should be applied to denote an ideology, the concept of management to capture managerial practices, and the concept of managerialization to describe an organizational change process.     

Unpacking Self-Rated Health and Quality of Life in Older Adults and Elderly in India: A Structural Equation Modelling Approach

The Study on global AGEing and adult health (SAGE) aims at improving empirical understanding of the health and well-being of older adults in low- and middle-income countries. A total of 321 adults aged 50 years and older were interviewed in rural Pune district, India, in 2007. We used Structural Equation Modelling (SEM) to examine the pathways through which social factors, functional disability, risk behaviours, and chronic disease experience influence self-rated health (SRH) and quality of life (QOL) amongst older adults in India. Both SRH and QOL worsened with increased age (indirect effect) and limitations in functional ability (direct effect). QOL, socio-economic status (SES), and social networking had no significant effect on SRH. Smoking was associated with the presence of at least one chronic illness, but this did not have a statistically significant effect on SRH. Higher social networking was seen amongst the better educated and those with regular income, which in turn positively affected the QOL rating. QOL had a direct, but statistically not significant, effect on SRH. In conclusion, the indirect effects of age on SRH mediated through functional ability, and the effects of SES on QOL mediated through social networking, provide new understanding of how age and socio-economic status affect SRH and QOL. By allowing for measurement errors, solving for collinearity in predictor variables by integrating them into measurement models, and specifying causal dependencies between the underlying latent constructs, SEM provides a strong link between theory and empirics.     

Medical care and regional mortality differences within the countries of the European community

This paper addresses the question whether regional mortality differences within developed countries reflect differences in characteristics of medical care service. It adds two new elements to previous studies on this subject: it concentrates on selected ‘avoidable’ causes of death, and it makes parallel analyses for ten EC countries. The results show that levels of medical care supply contribute little to the explanation of regional differences in mortality from the selected conditions. It is concluded that if regional mortality differences are related to health care, factors other than the level of supply are probably involved.     

Voluntary Associations and Net Fertility During the Swedish Demographic Transition

This study investigates the role of changing social relations for fertility decline during the European fertility transition. The growth of voluntary associations at the end of the nineteenth century entailed a radical shift in the landscape of social relations in Sweden. By combining micro-census data from 1890 to 1900 with local-level membership data for three voluntary association groups, this article assesses the effect of parish-level voluntary association size on net fertility in Sweden using mixed-effects Poisson regression models. The results show that the adoption of fertility limitation during the transition period was associated with the creation and diffusion of the idea of respectability within large social network organisations, an idea that has previously been shown to be connected to fertility limitation. Furthermore, by applying a social network perspective, the results show that the strength of the effect was dependent on the structure of the social networks in terms of size, density, and homogeneity. Voluntary association size had the strongest effect for the free churches, which created dense heterogeneous networks through systems of social control, while the size of the temperance association showed no effect on fertility because the connections between nodes were sparse.     

Structure Learning of Contextual Markov Networks using Marginal Pseudo‐likelihood

Markov networks are popular models for discrete multivariate systems where the dependence structure of the variables is specified by an undirected graph. To allow for more expressive dependence structures, several generalizations of Markov networks have been proposed. Here, we consider the class of contextual Markov networks which takes into account possible context‐specific independences among pairs of variables. Structure learning of contextual Markov networks is very challenging due to the extremely large number of possible structures. One of the main challenges has been to design a score, by which a structure can be assessed in terms of model fit related to complexity, without assuming chordality. Here, we introduce the marginal pseudo‐likelihood as an analytically tractable criterion for general contextual Markov networks. Our criterion is shown to yield a consistent structure estimator. Experiments demonstrate the favourable properties of our method in terms of predictive accuracy of the inferred models.