Hierarchical Generalized Linear Models and Frailty Models with Bayesian Nonparametric Mixing

This paper proposes Bayesian nonparametric mixing for some well-known and popular models. The distribution of the observations is assumed to contain an unknown mixed effects term which includes a fixed effects term, a function of the observed covariates, and an additive or multiplicative random effects term. Typically these random effects are assumed to be independent of the observed covariates and independent and identically distributed from a distribution from some known parametric family. This assumption may be suspect if either there is interaction between observed covariates and unobserved covariates or the fixed effects predictor of observed covariates is misspecified. Another cause for concern might be simply that the covariates affect more than just the location of the mixed effects distribution. As a consequence the distribution of the random effects could be highly irregular in modality and skewness leaving parametric families unable to model the distribution adequately. This paper therefore proposes a Bayesian nonparametric prior for the random effects to capture possible deviances in modality and skewness and to explore the observed covariates' effect on the distribution of the mixed effects.     

Preaching to the converted: conversion language and the constitution of the TV evangelical community

A paradox is noted in the presentation of conversion in TV evangelism; this is that although the necessity of being converted is a common theme it is clear that the evangelists are aware that their audience is already converted. An attempt is made to resolve the paradox by showing how TV evangelists use conversion language as rhetoric to legitimate their role and the rationality of the organisation which they represent. It is further shown how this legitimation is itself made possible through another use of conversion language which creates a sense of moral community embracing evangelists, organisation and audience as copartners in the shared project of conversion. Finally, it is shown how the presentation of TV evangelism as the central ritual of a moral community generates a further conversion rhetoric which depicts the community as both exclusive of and yet open to non-members; a contradiction which is resolved through the establishment of distinctive roles within the community related to the activity of winning converts.     

Staples,links, and poles in the construction of regional development theories

The developmental trajectories of extractive economies differ systematically from those of industrial economies. Analysis of these differences is difficult, because the specific characteristics and location in space of particular extracted resources distinguish extractive economies from each other far more than commodities produced and location distinguish industrial economies. The peculiar distoritions of Harold Innis's studies of particular staples as these were incorporated into general statements about regional economic development illustrate some of the tensions between ideographic and nomothetic goals in the analysis of regional economies and the dangers of resolving these tensions by collapsing the particular into the general. Subsequent use of these regional economic development models to obscure problems of particular resource extractive projects illuminates some of the unintended practical consequences of theoretical errors.     

A Full Bayesian Non-parametric Analysis Involving a Neutral to the Right Process

Implementation of a full Bayesian non-parametric analysis involving neutral to the right processes (apart from the special case of the Dirichlet process) has been difficult for two reasons: first, the posterior distributions are complex and therefore only Bayes estimates (posterior expectations) have previously been presented; secondly, it is difficult to obtain an interpretation for the parameters of a neutral to the right process. In this paper we extend Ferguson & Phadia (1979) by presenting a general method for specifying the prior mean and variance of a neutral to the right process, providing the interpretation of the parameters. Additionally, we provide the basis for a full Bayesian analysis, via simulation, from the posterior process using a hybrid of new algorithms that is applicable to a large class of neutral to the right processes (Ferguson & Phadia only provide posterior means). The ideas are exemplified through illustrative analyses.     

Naming the Beast: Defining and Classifying Service Innovations in Social Policy

Over the last decade, the search for innovationhas become a core component of much social policyformulation and implementation. This search, however,has often been undertaken with little attempt to define the actual nature of innovation or to classifyits different modes. This paper reports recent researchwhich has developed a typologyof organizational changewithin which to situate and analyze innovation in social policy. It argues that thistypologyis an essential tool both for socialpolicymakers and for those concerned with itsanalysis.     

The environment as a determinant of child mortality among migrants in frontier areas of Pará and RondÔnia, Brazil, 1980

This article examines child mortality as an indicator of the quality of life among migrants living in Brazilian Amazonia in 1980. I focus on migrants in the frontier states of Pará and RondÔnia, which experienced rapid settlement during the 1970s. The key question here is the effect of settlement location on child mortality rates. While RondÔnia had lower ratios of population per public health establishments and personnel than Pará, RondÔnia nonetheless exhibited a higher malaria prevalence in 1980. I therefore attribute locational differences in child mortality to environmental factors important to malaria transmission rather than to health care infrastructure. The findings from multivariate regression analysis show that net of the effects of human capital, migration history and migrant living standards, settling in Pará rather than RondÔnia resulted in significantly lower rates of child mortality. These findings suggest that environmental factors coinciding with location of frontier settlement had important consequences for the living standards of migrants in the Brazilian Amazon.     

New balanced sampling plans excluding adjacent units

Hedayat et al. [1988a. Sampling plans excluding contiguous units. J. Statist. Plann. Inference 19, 159–170; 1988b. Designs in survey sampling avoiding contiguous units. In: Krishnaiah, P.R., Rao, C.R. (Eds.). Handbook of Statistics, vol. 6. Elsevier, Amsterdam, pp. 575–583] first introduced balanced sampling designs for the exclusion of contiguous units. Sampling plans that excluded the selection of contiguous units within a given sample, while maintaining a constant second-order inclusion probability for non-contiguous units, were investigated for finite populations of N units arranged in a circular, one-dimensional ordering. There remain many open questions about the existence of such plans and their extension to plans excluding adjacent units. We present new generation techniques and new balanced sampling plans for the exclusion of adjacent units under finite, one-dimensional, circularly and linearly ordered populations.     

Disability and identity: the challenge of epilepsy

Through examining the case of people with epilepsy (which, as we demonstrate, has an ambiguous status in relation to both popular and academic conceptions of disability) we explore the fluid, negotiable and contingent nature of identity and, in particular, the identification as ‘disabled’. Disability, we argue, cannot be reduced to either biology or social oppression, or even primarily to biological or social factors: it is the outcome of a complex interaction between a multiplicity of factors – biological, environmental, social, psychological, cultural and political – which will interact and be experienced differently by different people, at different times and in different situations. Rather than conceiving of disability in ‘all or nothing’ terms and of differing explanations as competing and mutually exclusive, it would be more productive to see them as partial and potentially complementary contributions to the better understanding of a complex and multifarious reality.