The use of simple reparameterizations to improve the efficiency of Markov chain Monte Carlo estimation for multilevel models with applications to discrete time survival models

Summary.  We consider the application of Markov chain Monte Carlo (MCMC) estimation methods to random-effects models and in particular the family of discrete time survival models. Survival models can be used in many situations in the medical and social sciences and we illustrate their use through two examples that differ in terms of both substantive area and data structure. A multilevel discrete time survival analysis involves expanding the data set so that the model can be cast as a standard multilevel binary response model. For such models it has been shown that MCMC methods have advantages in terms of reducing estimate bias. However, the data expansion results in very large data sets for which MCMC estimation is often slow and can produce chains that exhibit poor mixing. Any way of improving the mixing will result in both speeding up the methods and more confidence in the estimates that are produced. The MCMC methodological literature is full of alternative algorithms designed to improve mixing of chains and we describe three reparameterization techniques that are easy to implement in available software. We consider two examples of multilevel survival analysis: incidence of mastitis in dairy cattle and contraceptive use dynamics in Indonesia. For each application we show where the reparameterization techniques can be used and assess their performance.     

A Cross‐Ethnic Study of Adolescents' Depressed Mood and the Erosion of Parental and Peer Warmth During the Transition to Young Adulthood

This 1‐year longitudinal study investigated the effects of adolescents' depressed mood on perceived parental and peer warmth during the transition to young adulthood. We hypothesized that ethnicity would moderate such effects. As part of a larger study, 511 adolescents (154 European, 205 Hispanic, and 152 Asian Americans) participated in this research. They were surveyed shortly before their high school graduation and again 1 year later. Analyses based on 2‐wave cross‐lagged models showed that higher levels of initial depressive symptoms predicted lower levels of subsequent perceived parental and peer warmth for European Americans. For Asian Americans, higher initial depressed mood was significantly associated with lower levels of perceived peer warmth and was marginally associated with lower levels of parental warmth. In contrast, the erosion of parental and peer warmth was not evident for Hispanic Americans. The role of culture in the erosion of parental and peer warmth during life transitions was discussed.     

Evaluating the Efficacy of the Prodigy Prevention Program across Urban and Rural Locales

Comorbid juvenile offenders are an overlooked, vulnerable population which tend to present with a more diagnostically serious picture. Their arrest rates have dramatically increased over the last 10 years, yet there is a dearth of prevention and intervention programs available. This study examined the effects of the Prodigy Cultural Arts Program on at-risk and adjudicated youth in a rural and an urban locale. Results suggest a significant improvement in family functioning overall as well as statistically significant changes in mental health symptoms including Depression/Anxiety, Somatic, and Suicidal symptoms for both males and females at urban and rural locales. Females appeared to especially benefit from the program. The findings here extend our knowledge regarding viable alternatives for juvenile offenders with mental health symptoms, particularly females.     

An Ecological Economics Perspective on Sustainability and Prospects for Ending Poverty

Despite our pretensions to science, modern industrial society is as myth-bound and mystical as any that has preceded it. Our prevailing cultural myth includes a dangerous vision of global sustainability and poverty reduction centered on unlimited economic expansion, free trade and technological fixes. This paper dissects the modern myth, exposing its conceptual flaws and practical failings. It then proposes an alternative conceptual framework for development derived from ecological economics and ecological footprint analysis. The new framework recognizes that the human enterprise is a subsystem of the ecosphere whose growth is constrained by biophysical limits. If humanity is to seize control of its destiny it must arise above wishful thinking and tribal instinct. Global society needs a new cultural myth rooted in humanity's unique claim to intelligence and self-awareness in the face of danger. Human security depends on equitable development—not growth—within the means of nature. Sustainability with social justice can be achieved only through an unprecedented level of international cooperation rooted in a sense of compassion for both other peoples and other species.     

The political ecology of disaster: An analysis of factors influencing U.S. tornado fatalities and injuries, 1998–2000

This study examines casualties from tornadoes in the United States between the years 1998 and 2000. A political model of human ecology (POET) was used to explore how the environment, technology, and social inequality influence rates of fatalities and injuries in two models. Data were drawn from four sources: John Hart's Severe Plot v2.0, National Weather Service (NWS) Warning Verification data, Storm Prediction Center (SPC) watch data, and tract-level census data. Negative binomial regression was used to analyze the causes of tornado fatalities and injuries. Independent variables (following POET) are classified in the following manner: population, organization, environment, and technology. Rural population, population density, and household size correspond to population; racial minorities and deprivation represent social organization; tornado area represents environment; and tornado watches and warnings, as well as mobile homes, correspond to technology. Findings suggest a strong relationship between the size of a tornado path and both fatalities and injuries, whereas other measures related to technology, population, and organization produce significant yet mixed results. Census tracts having larger populations of rural residents was, of the nonenvironmental factors, the most conclusive regarding its effects across the two models. The outcomes of analysis, although not entirely supportive of the model presented in this study, suggest to some degree that demographic and social factors play a role in vulnerability to tornadoes.     

How much does income matter in neighborhood choice?

There is a substantial literature on the residential mobility process itself and a smaller contribution on how households make neighborhood choices, especially with respect to racial composition. We extend that literature by evaluating the role of income and socioeconomic status in the neighborhood choice process for minorities. We use individual household data from the Los Angeles Family and Neighborhood Study to investigate the comparative choices of white and Hispanic households in the Los Angeles metropolitan area. We show that income and education are important explanations for the likelihood of choosing neighborhoods. But at the same time, own race preferences clearly play a role. While whites with more income choose more white neighborhoods, Hispanics with more income choose less Hispanic neighborhoods. One interpretation is that both groups are translating resources, such as income and education, into residence in whiter and ostensibly, higher status neighborhoods.

REORDER-POINT MODELS WITH DISCRETE PROBABILITY DISTRIBUTIONS

This paper is concerned with the determination of reorder points and safety-stock levels for situations where both the demand and lead-time distributions are discrete and independent. Formulas are developed for computing exact probabilities for the lead-time demand distribution on the basis of empirical distributions. Practitioners can use the model to obtain exact probabilities for situations where the computational burden is justified, while academicians can use the model to evaluate heuristic approaches to the reorder-point problem.     

Concepts,Theory, and Techniques: FACTORS AFFECTING THE PERCEIVED UTILIZATION OF COMPUTER-BASED DECISION SUPPORT SYSTEMS IN THE OIL INDUSTRY

This study investigates factors that may affect the use of decision support systems. The independent variables investigated included: characteristics of the decision makers such as age, educational level, experience, and cognitive style; characteristics of the decision support system such as response time, accuracy, timeliness, relevancy of output, format, and the distance traveled to interact with the system; and characteristics of the implementation process such as user involvement, user training, and top management support. The study included 8 systems and 64 subjects from the oil industry. The results indicated that the most important variables affecting decision support system usage were accuracy of output, user training, relevancy of output, and the decision maker's experience. Subject Areas: Decision Support Systems, Information Management, and Organizational Behavior.     

Ranking inequality: Applications of multivariate subset selection

Inequality measures are often presented in the form of a rank ordering to highlight their relative magnitudes. However, a rank ordering may produce misleading inference, because the inequality measures themselves are statistical estimators with different standard errors, and because a rank ordering necessarily implies multiple comparisons across all measures. Within this setting, if differences between several inequality measures are simultaneously and statistically insignificant, the interpretation of the ranking is changed. This study uses a multivariate subset selection procedure to make simultaneous distinctions across inequality measures at a pre-specified confidence level. Three applications of this procedure are explored using country-level data from the Luxembourg Income Study. The findings show that simultaneous precision plays an important role in relative inequality comparisons and should not be ignored. An erratum to this article can be found at