Restricted most powerful Bayesian tests for linear models

Uniformly most powerful Bayesian tests (UMPBTs) are a new class of Bayesian tests in which null hypotheses are rejected if their Bayes factor exceeds a specified threshold. The alternative hypotheses in UMPBTs are defined to maximize the probability that the null hypothesis is rejected. Here, we generalize the notion of UMPBTs by restricting the class of alternative hypotheses over which this maximization is performed, resulting in restricted most powerful Bayesian tests (RMPBTs). We then derive RMPBTs for linear models by restricting alternative hypotheses to g priors. For linear models, the rejection regions of RMPBTs coincide with those of usual frequentist F‐tests, provided that the evidence thresholds for the RMPBTs are appropriately matched to the size of the classical tests. This correspondence supplies default Bayes factors for many common tests of linear hypotheses. We illustrate the use of RMPBTs for ANOVA tests and t‐tests and compare their performance in numerical studies.     

Estimating unmet need for contraception by district within Ghana: an application of small-area estimation techniques

The importance of meeting the unmet need for contraception is nowhere more urgent than in the countries of sub-Saharan Africa, where the fertility decline is stalling and total unmet need exceeds 30 per cent among married women. In Ghana, where fertility levels vary considerably, demographic information at sub-national level is essential for building effective family planning programmes. We used small-area estimation techniques, linking data from the 2003 Ghana Demographic and Health Survey to the 2000 Ghana Population and Housing Census, to derive district-level estimates of contraceptive use and unmet need for contraception. The results show considerable variation between districts in contraceptive use and unmet need. The prevalence of contraceptive use varies from 4.1 to 41.7 per cent, while that of the use of modern methods varies from 4.0 to 34.8 per cent. The findings identify districts where family planning programmes need to be strengthened.     

More nonparametric Bayesian inference in applications

Discussion of “Nonparametric Bayesian Inference in Applications” by Peter Mueller, Fernando A. Quintana, Garritt Page: More Nonparametric Bayesian Inference in Applications.     

Care-giving Coping Strategies Among Mothers with Chronically Ill Children

ABSTRACT

The purpose of the study is to identify the types of stresses experienced and the coping strategies used by mothers who are responsible for providing care to their chronically ill and disabled children. A qualitative approach was selected to systematically document the experiences and highlight individual coping strategies. Purposive sampling was used in this qualitative pilot study, fifteen (15) mothers who were known to the researcher as well as two schools that specifically cater for children with disabilities were selected to participate in the study. Thirty questionnaires were handed out of which fifteen were returned. Data was collected using the coping Health Inventory for Parents (CHIP) to measure the extent to which parents perceived a number of coping strategies to be helpful in coping with the chronic illness of their child (ren). The scale comprised 45 items with 3 subscales. The results highlighted the types of stresses that were experienced by the mothers and the coping strategies practiced by these mothers when caring for their disabled children. Mothers who care for chronically ill children are faced with daily stress that relates to care as well as their personal life.     

The A+ of a practice doctorate

In light of the growing trend of developing practice doctoral degree programmes in social work, the author uses the first Doctor of Social Work (DSW) programme in Hong Kong, which is a unique programme in China, as a case study to examine the pedagogy and constraints of doctoral programmes in social work. The programme serves as an alternative to a typical PhD programme and is designed particularly for seasoned practitioners in the field who strive to achieve a balance between a research and practice orientation and between an academic and a pragmatic focus. This paper emphasises that, to stand out from the crowd, practice doctorates in Hong Kong should take part in practice-based research and publish in refereed journals.     

Change IS Possible: Reducing High-Risk Drinking Using a Collaborative Improvement Model

Abstract

Objective: To describe the adoption of public health and improvement methodologies to address college students’ high-risk drinking behaviors and to aid in prevention efforts. Participants: Members of 32 colleges and universities, content experts, and staff members of the National College Health Improvement Program (NCHIP). Methods: A 2-year learning collaborative developed by NCHIP trained individuals from 32 different college and universities in using the Plan–Do–Study–Act cycle as a method to create and implement initiatives aimed at reducing students’ high-risk drinking behaviors and related harms. Results: Participants experienced success ranging from noteworthy increases in type and amount of interventions directed at reducing high-risk drinking, to creating collaboratives across campus, the local community, and stakeholders. Challenges related to data collection and creating lasting cultural change remain. Conclusions: The use of quality improvement methodologies and creation of a national collaborative successfully effected meaningful change in high-risk drinking behaviors on college campuses.     

Pilot Testing a Cognitive-Behavioral Protocol on Psychosocial Predictors of Exercise,Nutrition, Weight,and Body Satisfaction Changes in a College-Level Health-Related Fitness Course

Abstract

Objective: Small-scale pilot testing of supplementing a required college health-related fitness course with a cognitive-behavioral exercise-support protocol (The Coach Approach). Participants: Three classes were randomly assigned to Usual processes (n = 32), Coach Approach–supplemented: Mid-size Groups (n = 32), and Coach Approach–supplemented: Small Groups (n = 34) conditions. Methods: Repeated-measures analyses of variance (ANOVAs) assessed overall and between-class changes in the behavioral/physiological factors of exercise, fruit/vegetable intake, and body mass index (BMI); and the psychosocial factors of self-regulation, exercise self-efficacy, mood, and body satisfaction. Dependent t tests evaluated within-class changes. Multiple regression analyses tested prediction of exercise by changes in self-regulation, self-efficacy, and mood. Results: Significant improvements in self-regulation and fruit/vegetable intake were found in all classes. The Coach Approach–supplemented classes demonstrated significant increases in exercise. Significant improvements in BMI, self-efficacy, and body satisfaction were found in only The Coach Approach–supplemented: Small Groups class. Psychosocial changes predicted increased exercise. Self-regulation was the strongest contributor. Conclusions: Overall, results were positive and warrant more comprehensive testing.     

Teaching Family Policy Through a Policy Practice Framework

This article reports on the development of an innovative family policy practice course for direct practice students that emphasizes experiential learning to develop policy practice skills. The course uses a policy practice framework to understand the creation and implementation of major family legislation. A semester-long project provides students with opportunities to develop skills in advocacy, policy development, and planning by working for social change in the community. The article discusses examples of projects that have been implemented by students, discusses challenges to teaching experiential methods for skill building, and provides critical evaluation of the projects and their relevance for macro practice research.