Bayesian Analysis of Multiple Hypothesis Testing with Applications to Microarray Experiments

Recently, the field of multiple hypothesis testing has experienced a great expansion, basically because of the new methods developed in the field of genomics. These new methods allow scientists to simultaneously process thousands of hypothesis tests. The frequentist approach to this problem is made by using different testing error measures that allow to control the Type I error rate at a certain desired level. Alternatively, in this article, a Bayesian hierarchical model based on mixture distributions and an empirical Bayes approach are proposed in order to produce a list of rejected hypotheses that will be declared significant and interesting for a more detailed posterior analysis. In particular, we develop a straightforward implementation of a Gibbs sampling scheme where all the conditional posterior distributions are explicit. The results are compared with the frequentist False Discovery Rate (FDR) methodology. Simulation examples show that our model improves the FDR procedure in the sense that it diminishes the percentage of false negatives keeping an acceptable percentage of false positives.     

On-Line Process Control using Attributes with Misclassification Errors: An Economical Design for Short-Run Production

On-line process control consists of inspecting a single item for every m (integer and m ≥ 2) produced items. Based on the results of the inspection, it is decided whether the process is in-control (the fraction of conforming items is p ₁; State I) or out-of-control (the fraction of conforming items is p ₂ < p ₁; State II). If the inspected item is non conforming, it is determined that the process is out-of-control, and the production process is stopped for an adjustment; otherwise, production continues. As most designs of on-line process control assume a long-run production, this study can be viewed as an extension because it is concerned with short-run production and the decision regarding the process is subject to misclassification errors. The probabilistic model of the control system employs properties of an ergodic Markov chain to obtain the expression of the average cost of the system per unit produced, which can be minimised as a function of the sampling interval, m. The procedure is illustrated by a numerical example.     

Estimation of P(Y < X) for Weibull Distribution Under Progressive Type-II Censoring

Based on progressively Type II censored samples, we consider the estimation of R = P(Y < X) when X and Y are two independent Weibull distributions with different shape parameters, but having the same scale parameter. The maximum likelihood estimator, approximate maximum likelihood estimator, and Bayes estimator of R are obtained. Based on the asymptotic distribution of R, the confidence interval of R are obtained. Two bootstrap confidence intervals are also proposed. Analysis of a real data set is given for illustrative purposes. Monte Carlo simulations are also performed to compare the different proposed methods.     

Beyond the Disciplinary Role of Governance: How Boards Add Value to Spanish Foundations

Due to the significant role of non‐profit organizations in the development of modern societies, these organizations must adjust their boardrooms. Since traditional theories of corporate governance appear to be limited in explaining the changing non‐profit world, we propose an extended model of governance that integrates the traditional arguments of agency theory with a cognitive dimension. Based on a sample of Spanish foundations, we present evidence on the effect of the board composition on foundations' organizational efficiency. We show that board size and independence do not have a definitive effect on the entity's efficiency. Instead, the knowledge diversity inside the boardroom and the active character of trustees have a positive influence on resource allocation. The cognitive dimension of the extended model of governance is critical to explaining how boards impact on organizational performance.     

Integrated Approaches for the Public Health Prioritization of Foodborne and Zoonotic Pathogens

To address the persistent problems of foodborne and zoonotic disease, public health officials worldwide face difficult choices about how to best allocate limited resources and target interventions to reduce morbidity and mortality. Data‐driven approaches to informing these decisions have been developed in a number of countries. Integrated comparative frameworks generally share three methodological components: estimating incidence of acute illnesses, chronic sequelae, and mortality; attributing pathogen‐specific illnesses to foods; and calculating integrated measures of disease burden such as cost of illness, willingness to pay, and health‐adjusted life years (HALYs). To discuss the similarities and differences in these approaches, to seek consensus on principles, and to improve international collaboration, the E.U. MED‐VET‐NET and the U.S.‐based Food Safety Research Consortium organized an international conference convened in Berlin, Germany, on July 19–21, 2006. This article draws in part on the deliberations of the conference and discusses general principles, data needs, methodological issues and challenges, and future research needs pertinent to objective data‐driven analyses and their potential use for priority setting of foodborne and zoonotic pathogens in public health policy.     

Evaluation of Scenarios for Reducing Human Salmonellosis Through Household Consumption of Fresh Minced Pork Meat

Nontyphoidal salmonellosis is the second most frequently reported zoonotic disease in the European Union (EU) and is considered to be a major threat to human health worldwide. The most reported Salmonella serovar in the EU is S. Enteritidis, mainly associated with egg contamination, followed by S. Typhimurium, with the latter being the most predominant serovar isolated from pork. These findings suggest that reducing the Salmonella contamination in the pork production might be a good strategy to prevent and control human salmonellosis in the EU. Recently, a quantitative microbial risk assessment (QMRA) has been developed to assess the risks for human salmonellosis due to home consumption of fresh minced pork meat in Belgium.⁽ 1 ⁾ The newly developed risk model is called the METZOON model. In the current study, the METZOON model was used to evaluate the effectiveness of different hypothetical Salmonella mitigation strategies implemented at different stages of the minced pork production and consumption chain by means of a scenario analysis. To efficiently evaluate the mitigation strategies, model results were obtained by running simulations using the randomized complete block design. The effectiveness of a mitigation strategy is expressed using point and interval estimates of the effect size for dependent observations, expressed as the standardized difference in population means. The results indicate that the most effective strategies are taken during the slaughter processes of polishing, evisceration, and chilling, and during postprocessing, whereas interventions in the primary production and at the beginning of the slaughter process seem to have only a limited effect. Improving consumer awareness is found to be effective as well.     

Emotional intelligence and leadership emergence in small groups

We report the findings from two studies that examine the association between emotional intelligence and leadership emergence in small groups. In both studies, members of groups completed measures of emotional intelligence and other individual differences prior to working on a group project. Their peers rated their leadership emergence at the conclusion of the project. Overall emotional intelligence and a number of its dimensions were associated with leadership emergence over and above cognitive intelligence, personality traits, and gender. These findings were observed when emotional intelligence was measured with an ability test but not when it was measured with a self-report scale. Among the dimensions of emotional intelligence, the ability to understand emotions was most consistently associated with leadership emergence.     

Sleep disturbances in shift-workers: A marker for maladaptation syndrome

Abstract

This study aims to examine whether the relationship between overtime and well-being is influenced by the voluntary vs. involuntary (i.e., compulsory) nature of overtime work and by the presence or absence of rewards for overtime. We also explored the prevalence of these types of overtime and how they were related to work and personal characteristics. A survey was conducted among a representative sample of Dutch full-time employees (N=1612). AN(C)OVA was used to compare rewarded and unrewarded, voluntary and involuntary overtime workers on personal and work characteristics, fatigue, and work satisfaction. Most overtime workers were rewarded (62%). About half of the sample (n=814) could be classified as either voluntary or involuntary overtime workers, or as having “mixed reasons” to work overtime. Voluntary and unrewarded overtime workers had a relatively high income and favourable job characteristics. Involuntary overtime work was associated with relatively high fatigue and low satisfaction, especially for involuntary overtime workers without rewards who can be considered a burnout risk group. Voluntary overtime workers were non-fatigued and satisfied, even without rewards. It can be concluded that control over overtime and rewards for overtime are important for well-being. Moderate overtime work may not be a problem if it is done voluntarily. Moreover, the negative effects of compulsory overtime work may be partly offset by fair compensation for the extra work.     

Social support at work and its relationship to absenteeism

Abstract

In this study, the impact of psychosocial work characteristics on illness and work absenteeism has been examined. The study group consisted of 133 civil servants (33 men and 100 women) with a mean age of 43 years (range 21-65 years). Psychosocial work characteristics, social support, quality of life, work absenteeism, psychosomatic and gastrointestinal symptoms were assessed by means of questionnaires. Five different work support or functions were identified by factor analysis: appraisal support, belonging support, instrumental support, emotional support and supportive atmosphere. The analysis showed that work absenteeism was associated with lower belonging support and a less supportive atmosphere. Furthermore, psychosomatic symptoms were associated with lack of support at work, measured as belonging support, instrumental support and supportive atmosphere. Gastrointestinal symptoms were associated with lack of belonging support and lack of social integration outside work. A bad general health status was associated with lack of insrumental support. The association between work absenteeism and social support at work was confirmed in multivariate analysis with control for confounding factors. In this analysis, high work strain and low social support were associated with a lower quality of life, which in turn was associated with increased absenteeism. Although there are well-known difficulties in drawing causal conclusions from cross-sectional studies, it seems reasonable to conclude that social support at work is an important factor to take into consideration when analysing illness and absenteeism from work.     

The use of performance measurement as a basis for project control of offshore modification oil and gas projects

The methods presented in this article are based on the qualitative and quantitative analysis of close-out data drawn from a portfolio of 38 modification projects. The primary premise is that modification projects are intrinsically subject to quantity and complexity growth during the course of detailed engineering as a consequence of the indeterminate interface with the existing facility. The normative project control routines need to be supplemented by material take-offs-based weight monitoring as a basis for re-estimation and re-calibration of the project baselines as detailed engineering proceeds in order to mitigate this uncertainty as early as possible. Similarly, estimating needs to be based on norms drawn from performance measurement of modification work rather than extrapolations from greenfield projects, particularly the work performed on the platforms. This article presents a portfolio of project control methods based on the performance measurement principle covering the short term of single projects as well as the longer term multi-project perspective.