Addressing potential prior‐data conflict when using informative priors in proof‐of‐concept studies

Bayesian methods are increasingly used in proof‐of‐concept studies. An important benefit of these methods is the potential to use informative priors, thereby reducing sample size. This is particularly relevant for treatment arms where there is a substantial amount of historical information such as placebo and active comparators. One issue with using an informative prior is the possibility of a mismatch between the informative prior and the observed data, referred to as prior‐data conflict. We focus on two methods for dealing with this: a testing approach and a mixture prior approach. The testing approach assesses prior‐data conflict by comparing the observed data to the prior predictive distribution and resorting to a non‐informative prior if prior‐data conflict is declared. The mixture prior approach uses a prior with a precise and diffuse component. We assess these approaches for the normal case via simulation and show they have some attractive features as compared with the standard one‐component informative prior. For example, when the discrepancy between the prior and the data is sufficiently marked, and intuitively, one feels less certain about the results, both the testing and mixture approaches typically yield wider posterior‐credible intervals than when there is no discrepancy. In contrast, when there is no discrepancy, the results of these approaches are typically similar to the standard approach. Whilst for any specific study, the operating characteristics of any selected approach should be assessed and agreed at the design stage; we believe these two approaches are each worthy of consideration. Copyright © 2015 John Wiley & Sons, Ltd.     

The Research Environment Norm Inventory (RENI): a study of integrity in research administrative systems

University research administrators have been generally ignored in basic studies of research integrity. Hensley noted that research administrators are "essential... to the achievement of the specific missions of postsecondary institutions... and to science and the academic infrastructure". The following study sought to extend the scope of research on research integrity to research administrative structures with a new instrument called the Research Environment Norm Inventory or RENI. University research administrators and their professional association were targeted for data collection. Evidence suggested that research administration in the United States supports integrity in the research environment through: (1) respect for community; (2) respect for institutional boundaries; (3) professionalism; (4) respect for authority structures; (5) sensitivity to system conflicts. The study suggested that integrity structures are dictated largely by the institutional settings and environments.     

Foreign-born emigration: A new approach and estimates based on matched CPS files

The utility of postcensal population estimates depends on the adequate measurement of four major components of demographic change: fertility, mortality, immigration, and emigration. Of the four components, emigration, especially of the foreign-born, has proved the most difficult to gauge. Without "direct" methods (i.e., methods identifying who emigrates and when), demographers have relied on indirect approaches, such as residual methods. Residual estimates, however are sensitive to inaccuracies in their constituent parts and are particularly ill-suited for measuring the emigration of recent arrivals. Here we introduce a new method for estimating foreign-born emigration that takes advantage of the sample design of the Current Population Survey (CPS): repeated interviews of persons in the same housing units over a period of 16 months. Individuals appearing in a first March Supplement to the CPS but not the next include those who died in the intervening year, those who moved within the country, and those who emigrated. We use statistical methods to estimate the proportion of emigrants among those not present in the follow-up interview. Our method produces emigration estimates that are comparable to those from residual methods in the case of longer-term residents (immigrants who arrived more than 10 years ago), but yields higher--and what appear to be more accurate--estimates for recent arrivals. Although somewhat constrained by sample size, we also generate estimates by age, sex, region of birth, and duration of residence in the United States.     

Modeling Demand For Unionization With Nontraditional Data Analysis Methods

Upon reviewing the extant literature on determinants of unionism, it becomes clear that many areas that have had a plethora of research attention do not converge upon singularly directional findings. This study explores a potential cause of such an apparent anomaly: nonlinearity of data. An exploratory examination of correlation coefficients among typical union determinant variables seems to show different patterns of relationships at different levels of union demand. Thus, a break from traditional linear data analysis techniques is explored in the interest of explaining more variance with typical, theoretically derived variables by using neural network analysis. Results of analyses on industry level data reveal that using neural network analysis to model union demand explained over four times as much variance as multiple regression analysis.     

The Impact of Learning and Labor Attrition on Worker Flexibility in Dual Resource Constrained Job Shops

This study investigates the impact of worker learning, worker flexibility, and labor attrition on the system performance of a dual resource constrained (DRC) job-shop. The effects of learning and labor attrition have not been previously addressed in DRC literature. Results from the study, consistent with previous literature, show that the greatest benefits are achieved when inter-departmental worker flexibility is incrementally introduced into the system. In addition, the learning environment, which depends on the initial processing time of jobs and the learning rates of workers, is shown to impact the acquisition of flexibility. The study also shows that the impact of labor attrition on system performance under certain shop conditions may be significant.     

Evaluating social investment in disability policy

Disability policy in European countries is displaying a shift towards social investment: increasing human capital and access to the labour market. The reasoning that underlies this transition is that disabled persons would benefit from mainstream employment, but are impeded in traditional policy by deficiencies in labour supply and demand. However, the shift towards more activating policies in many countries is accompanied by a decline in social protection. It is unclear whether social investment may effectively promote the employment chances of disabled persons within this context. The present research examines this question through a quantitative, cross-sectional, multilevel analysis on microdata from 22 EU countries. Our findings suggest greater activation to predict lower employment chances, while reducing passive support shows mixed effects. Conversely, measures for facilitation in daily life predict greater employment chances, as do measures for sheltered work. These findings raise questions over the value of social investment for disabled persons—and underline the need to overcome broader barriers in the labour market and in society.     

Capturing temporal dynamics of fear behaviors on a moment‐to‐moment basis

Identifying patterns of fearful behaviors early and accurately is essential to identify children who may be at increased risk for psychopathology. Previous work focused on the total amount of fear by using composites across time. However, considering the temporal dynamics of fear expression might offer novel insights into the identification of children at risk. One hundred and twenty‐five toddlers participated in high‐ and low‐fear tasks. Data were modeled using a novel two‐step approach. First, a hidden Markov model estimated latent fear states and transitions across states over time. Results revealed children's behavior was best represented by six behavioral states. Next, these states were analyzed using sequence clustering to identify groups of children with similar dynamic trajectories through the states. A four‐cluster solution found groups of children varied in fear response and regulation process: “external regulators” (using the caregiver as a regulation tool), “low reactive” (low reaction to stimulus), “fearful explorers” (managing their own internal state with minimal assistance from the caregiver), and “high fear” (fearful/at‐caregiver state regardless of task). The combination of analytic tools enabled fine‐grained examination of the processes of fearful temperament. These insights may help prevention programs target behaviors that perpetuate anxious behavior in the moment.     

Sensitivity Analysis in Linear Programming Models with Common Inputs

The importance of sensitivity analysis information in linear programming has been stressed in the management science literature for some time. Indeed, Gal [3] has devoted an entire text to just this issue. Linear programs with common inputs (cost coefficients or right-hand-side values) present a problem in that classical sensitivity analysis does not allow for the simultaneous changes required to determine the sensitivity of these models to common inputs. We first survey the approaches previously developed for simultaneous-change sensitivity analysis and cast them in the framework of the special common input case. These general techniques are compared to a simple aggregate variable technique that has not received attention in the literature.     

The Popular Kids Don't Matter: Centrality and Influence on Adolescent Behavior

This study examines the widely held belief that socially central individuals are disproportionately influential in their networks. Using data from the National Longitudinal Study of Adolescent Health, it models the association between four behavioral outcomes and two distinct specifications of the behaviors' relationship to network prominence. This study finds little evidence that sociometrically central individuals are more influential than randomly chosen peers from the same network when predicting drinking, smoking, and sports participation. Students resemble their peers in systematic ways, but it is unlikely that this is because central students serve as a reference for the group or because students adjust their actions based on the social rewards that they observe given their position in a social network.