Practical Methods for Bounding Type I Error Rate with an Internal Pilot Design

New analytic forms for distributions at the heart of internal pilot theory solve many problems inherent to current techniques for linear models with Gaussian errors. Internal pilot designs use a fraction of the data to re-estimate the error variance and modify the final sample size. Too small or too large a sample size caused by an incorrect planning variance can be avoided. However, the usual hypothesis test may need adjustment to control the Type I error rate. A bounding test achieves control of Type I error rate while providing most of the advantages of the unadjusted test. Unfortunately, the presence of both a doubly truncated and an untruncated chi-square random variable complicates the theory and computations. An expression for the density of the sum of the two chi-squares gives a simple form for the test statistic density. Examples illustrate that the new results make the bounding test practical by providing very stable, convergent, and much more accurate computations. Furthermore, the new computational methods are effectively never slower and usually much faster. All results apply to any univariate linear model with fixed predictors and Gaussian errors, with the t-test a special case.     

Religion and Trends in Euthanasia Attitudes among U.S. Adults, 1977–2004

Though many studies address the role of religion in predicting social attitudes over time, none has examined this relationship specifically for euthanasia. Using a large, nationally representative data source, this study seeks to address this void. Our findings indicate that considerable differences exist among religious denominations regarding the legalization of euthanasia. Specifically, we note a liberalizing trend for all included denominations. We also demonstrate substantial differences in the rates of liberalization, particularly in comparison to conservative Protestants. We conclude with an assessment of our findings relative to previous studies on religion and public opinion.

Getting affairs in order: influences of social support and religious coping on end-of-life planning among open-heart surgery patients

Despite growing interest in end-of-life issues, little research has been done concerning end-of-life plans made by patients with cardiovascular disease (CVD), a disease that constitutes the highest mortality in the United States. Even less information is available on patients who undergo major cardiac surgery, a life-altering procedure with some risk of death, in terms of their engagement in end-of-life planning (EOLP). This prospective study enabled the contributors to explore EOLP among 309 middle-aged and older open-heart surgery patients, using survey data from three sequential interviews. A hierarchical logistic regression model shows that older age, higher education, greater social support, and negative religious coping were positively related to the likelihood of engaging in EOLP. Minority race was inversely related to EOLP. Findings suggest the potential role for social workers in helping cardiac patients to access social spiritual resources, and to consider engagement in EOLP.     

The Future of Nuclear Power: Value Orientations and Risk Perception

Since the turn of the 21st century, there has been a revival of interest in nuclear power. Two decades ago, the expansion of nuclear power in the United States was halted by widespread public opposition as well as rising costs and less than projected increases in demand for electricity. Can the renewed enthusiasm for nuclear power overcome its history of public resistance that has persisted for decades? We propose that attitudes toward nuclear power are a function of perceived risk, and that both attitudes and risk perceptions are a function of values, beliefs, and trust in the institutions that influence nuclear policy. Applying structural equation models to data from a U.S. national survey, we find that increased trust in the nuclear governance institutions reduces perceived risk of nuclear power and together higher trust and lower risk perceptions predict positive attitudes toward nuclear power. Trust in environmental institutions and perceived risks from global environmental problems do not predict attitudes toward nuclear power. Values do predict attitudes: individuals with traditional values have greater support for, while those with altruistic values have greater opposition to, nuclear power. Nuclear attitudes do not vary by gender, age, education, income, or political orientation, though nonwhites are more supportive than whites. These findings are consistent with, and provide an explanation for, a long series of public opinion polls showing public ambivalence toward nuclear power that persists even in the face of renewed interest for nuclear power in policy circles.     

Sexual minority youth in romantic relationships: Associations with youth well-being

This study investigated differences in depressive symptoms, loneliness, and self-esteem for monosexual (lesbian, gay) and plurisexual (bisexual, pansexual, queer) sexual minority youth (SMY) by relationship status (single, partnered) and relationship configuration (same-gender partner, different-gender partner). Participants included 338 SMY (M_age = 19.10 years) who reported on their relationship status, partner's gender identity, well-being, and ability to confide in partner about LGBTQ issues. Results indicated that for plurisexual youth, single status was associated with greater loneliness; plurisexual youth with same-gender partners reported fewer depressive symptoms and marginally greater ability to confide in their partner about LGBTQ issues than those with different-gender partners. Findings reveal similarities across SMY while also highlighting some unique challenges among plurisexual youth with different-gender partners.     

Queue hurdle Coxian phase-type model for two-stage process of population-based cancer screening

Statistical Methods & Applications - The quality assurance of two-stage population-based cancer screening program is determined by arrival rate (attending screening), positive rate (determined...     

Evaluation of error across natural gas pipeline incidents

This paper demonstrates how qualitative analysis can be a novel means of investigating theories of error and causation in natural gas pipeline incidents. Qualitative analysis offers unique opportunities to understand process, interactions, and the role of context in identifying active error and latent conditions in incident causation. Through the coding of text from 24 onshore natural gas pipeline incident reports on leaks and explosions in the United States and Canada, our findings reveal a proportion of active and latent errors consistent with other hazardous infrastructure contexts (roughly 3:1 latent-active ratio across 817 coded errors). These findings underscore the robustness of extant error theory and support the argument for documenting multiple, connected causes of disaster in aggregate. Conclusions highlight the utility of in-depth case analyses and critique present pipeline incident database aggregation. Our interpretation provides a means to convey complex causation in aggregate form thus enabling more nuanced future qualitative and qualitative analyses.     

Predictors of tree cover in residential open space: a multi-scale analysis of suburban Philadelphia

Urban Ecosystems - Tree cover is an important aspect of environmental quality and quality of life, but is declining in many metropolitan regions. This study examines tree cover and its relationship...     

Chinese immigrant child and maternal reactions to disappointment: Cultural fit impacts the bidirectional associations

Culture provides a context in which emotion socialization is embedded, and the bidirectional effects between parents’ emotion socialization and children's emotional behaviors may work differently across cultures. To understand how emotion socialization may be shaped by the cultural context, we examined the moderating role of Asian cultural values in bidirectional associations between maternal emotion socialization practices and child anger and sadness. Seventy-four U.S. Chinese immigrant mothers (M_age = 40.71 years, SD = 3.61) completed measures assessing their Asian cultural values and parenting style. Children experienced a disappointment task in the lab (Cole, 1986), and mothers and their children (M_age = 6.73 years, SD = .95; 55% female) were observed at two different time intervals. Mothers’ socialization practices (emotion dismissing, emotion coaching, and moral and behavioral socialization) and children's anger and sadness responses at both intervals were coded. Mothers’ greater Asian cultural values buffered the negative effects of their emotion dismissing practices on children's anger and sadness. However, Asian cultural values did not impact the effects of children's anger and sadness on mothers’ emotion dismissing practices. When mothers endorsed fewer Asian values, their emotion coaching practices reduced children's anger and sadness. Children's anger and sadness evoked more emotion coaching practices when mothers endorsed lower levels of Asian cultural values. In addition, children's anger and sadness evoked greater moral and behavioral responses from their mothers when mothers endorsed more Asian values. Overall, findings underscored the importance of cultural values in the interplay between mothers’ emotion socialization practices and children's emotions.     

Does Observed Conflict Recovery Play a Role in Adolescent Dating Aggression?

In a study of conflict recovery and adolescent dating aggression, 14- to 18-year-old couples (N = 209 dyads) participated in a 1-hr observational assessment. Negative behavior was observed during conflict-evoking “hot” tasks and in a “cooldown” task. Physical and psychological dating aggression were assessed via questionnaires. Negative behavior measured in the cooldown task was not associated with dating aggression after controlling for carryover effects of negativity from the hot to cooldown tasks. Moreover, cooldown negativity moderated the associations of hot task negativity and dating aggression. Actor and partner effects were disentangled via dyadic data analyses. Given the paucity of observational studies of dating aggression, our findings are an important contribution to the literature and in need of replication and extension.