Orthogonal Stiefel manifold optimization for eigen-decomposed covariance parameter estimation in mixture models

Within the mixture model-based clustering literature, parsimonious models with eigen-decomposed component covariance matrices have dominated for over a decade. Although originally introduced as a fourteen-member family of models, the current state-of-the-art is to utilize just ten of these models; the rationale for not using the other four models usually centers around parameter estimation difficulties. Following close examination of these four models, we find that two are actually easily implemented using existing algorithms but that two benefit from a novel approach. We present and implement algorithms that use an accelerated line search for optimization on the orthogonal Stiefel manifold. Furthermore, we show that the ‘extra’ models that these decompositions facilitate outperform the current state-of-the art when applied to two benchmark data sets.     

Graduate education in an age of decline

In the 1980s, sociology departments in a number of universities have been caught up in a deteriorating educational environment. Many have experienced reduced numbers of faculty and students, often becoming little more than service departments. This article looks at this process as it occurred at a midwestern university noted for its strong theoretical tradition in graduate education. How this happened and the effects it had on graduate students is the focus of this work. Barbara Ryan graduated from Washington University in St. Louis in 1986 and presently is teaching in the Department of Sociology at Northern Illinois University. Her specialty area is social movements and her dissertation was an analysis of change in the contemporary women’s movement.     

A new approach to assessing ethical conduct in scientific work

The intent of the current article is to describe the development of a new approach to the study of ethical conduct in scientific research settings. The approach presented in this article has two main components. The first component entails the development of a taxonomy of ethical events as they occur across a broad range of scientific disciplines. The second involves the identification of proximate criteria that will allow systematic and objective evaluation of ethical behaviors through low-fidelity performance simulations. Two proposed measures based on the new approach are intended to identify and measure variations in the scientific environment that might predispose certain individuals to make unethical decisions.     

Latent Variable Models for Mixed Discrete and Continuous Outcomes

We propose a latent variable model for mixed discrete and continuous outcomes. The model accommodates any mixture of outcomes from an exponential family and allows for arbitrary covariate effects, as well as direct modelling of covariates on the latent variable. An EM algorithm is proposed for parameter estimation and estimates of the latent variables are produced as a by-product of the analysis. A generalized likelihood ratio test can be used to test the significance of covariates affecting the latent outcomes. This method is applied to birth defects data, where the outcomes of interest are continuous measures of size and binary indicators of minor physical anomalies. Infants who were exposed in utero to anticonvulsant medications are compared with controls.     

Inequality in Life Spans and a New Perspective on Mortality Convergence Across Industrialized Countries

The second half of the twentieth century witnessed substantial convergence in life expectancy around the world. We examine differences in the age pattern of mortality in industrialized countries over time to show that inequality in adult life spans, which we measure with the standard deviation of life table ages at death above age 10 years, S₁₀, is increasingly responsible for the remaining divergence in mortality. We report striking differences in level and trend of S₁₀ across industrialized countries since 1960, which cannot be explained by aggregate socioeconomic inequality or differential external‐cause mortality. Rather, S₁₀ reflects both within‐ and between‐group inequalities in life spans and conveys new information about their combined magnitudes and trends. These findings suggest that the challenge for health policies in this century is to reduce inequality, not just lengthen life.     

The Social Context of Young Children's Peer Victimization

The goal of this study was to investigate differences in the social context of peer victimization for preschoolers and kindergarteners. Data were collected from 168 children. For preschoolers, neither social acceptance nor friendships were significantly related to peer victimization. Instead, playing with peers and exposure to aggressive peers were associated with higher rates of peer victimization. For kindergarteners, exposure to aggressive peers also contributed to the risk for peer victimization, but being liked by peers and having friends were inversely related to victimization, thereby providing a buffering effect. The developmental implications of these findings are discussed.     

Creating more trauma-informed services for children using assessment-focused tools

This article promotes integrating assessment and evidence-based practice in the treatment of traumatized children through a review of two newly developed trauma assessment tools: (1) the Child Welfare Trauma Referral Tool (CWT), and (2) Assessment-Based Treatment for Traumatized Children: A Trauma Assessment Pathway Model (TAP). These tools use pathways and algorithms to increase understanding of individual child trauma victims, and assist professionals working with children to make appropriate referral and treatment decisions within both child welfare and mental health contexts.     

College students use cell phones while driving more frequently than found in government study

OBJECTIVE: Cell phone use while driving is hazardous; it quadruples the risk of a collision and multiplies the risk of a fatality nine-fold. The National Highway Traffic Safety Administration estimates that 8% of young drivers and 5% of all drivers use cell phones while driving. PARTICIPANTS AND METHODS: The authors trained graduate student volunteers to observe the daytime cell phone use of 3,650 drivers leaving the student exits of college parking structures at a large university. RESULTS: The student observers recorded a cell phone usage rate of 11.1%, which was significantly higher than that seen in the National Occupant Protection Use Survey (p<.00001). Female drivers were 1.51 times more likely to be using a cell phone while driving than were men (12.9% vs 8.6%, p<.001), and drivers with passengers were 0.15 times more likely to drive while telephoning than were solo drivers (1.8% vs 12.1%, p<.001). CONCLUSIONS: The authors offer suggestions for possible interventions to reduce this hazardous behavior.