Heteroscedastic Global Tests that the Regression Parameters for Two or More Independent Groups are Identical

Simulation results are reported on methods that allow both within group and between group heteroscedasticity when testing the hypothesis that independent groups have identical regression parameters. The methods are based on a combination of extant techniques, but their finite-sample properties have not been studied. Included are results on the impact of removing all leverage points or just bad leverage points. The method used to identify leverage points can be important and can improve control over the Type I error probability. Results are illustrated using data from the Well Elderly II study.     

The relationship between presenteeism and job satisfaction: A mediated moderation model using work engagement and perceived organizational support

Research on presenteeism has largely focused on its prevalence, determinants and consequences on the health of both employees and organizations. To date, few studies have examined the influence of presenteeism on workers’ attitudinal and motivational responses. Based on the Effort–Recovery Theory, this study evaluates the mediating effect of work engagement in the relationship between presenteeism and job satisfaction. Building on previous work, this study also proposes to consider perceived organizational support as a moderator of the work engagement–job satisfaction relationship. Presented in the form of a mediated moderation model, the results of bootstrapped regression analyses show three main results. First, presenteeism is negatively associated with work engagement and job satisfaction. Secondly, work engagement is one of the factors through which presenteeism influences job satisfaction. Thirdly, perceived organizational support moderates the association between work engagement and job satisfaction so that, at a low level of work engagement, feeling supported by the organization makes a difference on job satisfaction. Overall, this research is part of the limited number of studies that have focused on the interplay existing between presenteeism and its consequences in terms of job attitudes.     

Keeping the "care" in managed care: the importance of the therapeutic relationship and psychosocial factors in the managed care setting

There is much truth in the adage that "the more things change, the more they stay the same." Nowhere does this seem more apparent than in health care where, amidst monumental reconfiguration, basic foundations of physician-patient relationships and attention to the impact of psychosocial factors on health and health care delivery remain as critical influences. While the importance of the therapeutic relationship and the influence of psychosocial factors in medical care has been clear in traditional systems of delivery, these factors may be even more critical in managed care systems. These emphases must be incorporated by design, however, and not left to default.     

Dynamic reliability models with conditional proportional hazards

A dynamic approach to the stochastic modelling of reliability systems is further explored. This modelling approach is particularly appropriate for load-sharing, software reliability, and multivariate failure-time models, where component failure characteristics are affected by their degree of use, amount of load, or extent of stresses experienced. This approach incorporates the intuitive notion that when a set of components in a coherent system fail at a certain time, there is a jump from one structure function to another which governs the residual lifetimes of the remaining functioning components, and since the component lifetimes are intrinsically affected by the structure function which they constitute, then at such a failure time there should also be a jump in the stochastic structure of the lifetimes of the remaining components. For such dynamically-modelled systems, the stochastic characteristics of their jump times are studied. These properties of the jump times allow us to obtain the properties of the lifetime of the system. In particular, for a Markov dynamic model, specific expressions for the exact distribution functions of the jump times are obtained for a general coherent system, a parallel system, and a series-parallel system. We derive a new family of distribution functions which describes the distributions of the jump times for a dynamically-modelled system.     

A model for short-term group therapy on a children's inpatient unit

The paper describes the group therapy model which has evolved over the past two years on the Child Psychiatry Inpatient Unit of North Carolina Memorial Hospital. It begins with a brief review of the literature concerning group therapy with latency-aged children. The setting and patient population are described. Vignettes from group sessions are used to illustrate techniques which have evolved to deal with this particular setting and population. These include a leadership style characterized by much clarification, modeling, and limit setting; extensive focus on beginning and terminating; clearly defined behavioral limits; the use of time out; and use of play materials requiring little skill or attention as anxiety binders as well as projective media; and observation and supervisory discussions open to all members of the treatment team. The paper concludes with a discussion of the potentials and limitations of group therapy in such a setting. The thesis is that this model is effective in diagnosing developmental levels, in developing interpersonal skills, in substituting verbalization for action, and in mastering separation anxieties, though the setting limits the extent to which controlled regression, transference, and group pressures can develop and be used to produce change.     

Mini-batch learning of exponential family finite mixture models

Mini-batch algorithms have become increasingly popular due to the requirement for solving optimization problems, based on large-scale data sets. Using an existing online expectation–maximization (EM) algorithm framework, we demonstrate how mini-batch (MB) algorithms may be constructed, and propose a scheme for the stochastic stabilization of the constructed mini-batch algorithms. Theoretical results regarding the convergence of the mini-batch EM algorithms are presented. We then demonstrate how the mini-batch framework may be applied to conduct maximum likelihood (ML) estimation of mixtures of exponential family distributions, with emphasis on ML estimation for mixtures of normal distributions. Via a simulation study, we demonstrate that the mini-batch algorithm for mixtures of normal distributions can outperform the standard EM algorithm. Further evidence of the performance of the mini-batch framework is provided via an application to the famous MNIST data set.     

The French Regions and Their Social Health

In this article, a new indicator designed to capture the multidimensionality of the social health of the French regions is put to the test. Drawing on regional data for 2004, this indicator of social health (ISH) sheds new light on the social performance of the French regions. The worst performers are the highly urbanised regions, whereas others, such as Limousin, turn out to perform well in social terms. Two or three regions remain stuck at the bottom of the table regardless of the indicator used. Nord-Pas-de-Calais and Languedoc-Roussillon, which bring up the rear in terms of gross disposable income (GDI) and gross domestic product (GDP) per capita, also record very mediocre scores for social health as measured by our ISH.

Individuals and Communities: the Importance of Neighbors Volunteering

In this analysis, I examine the effects of community-level volunteering on an individual’s choices regarding time – whether to work and whether to volunteer. In order to better explain the decision to volunteer, a classic pure public goods structure is contrasted with a less restrictive impure public goods model that admits other possible private motivations. The results of this study undermine the neoclassical notion that volunteering can be understood solely as a pure public good that is provided less when others are seen to be contributing. In fact, individuals are found to be more, not less, likely to volunteer when others in their communities do so. An innovative instrumental variables strategy is used to account for reflection bias and the possible endogeneity caused by selective sorting of individuals into neighborhoods, which allows for a causal interpretation of these results. Employment regressions provide preliminary evidence that average volunteering relates, to some extent, with the decision of whether to participate in the labor force. Variations in the effect of average volunteering across age and gender are also explored. The present work is unique by virtue of its use of a large and representative dataset, along with rigorous statistical testing. I use United States Census 2000 Summary File 3 and Current Population Survey (CPS) 2004–2007 September Supplement file data and control for various individual and community-level characteristics.