Modelling daily multivariate pollutant data at multiple sites

Summary. This paper considers the spatiotemporal modelling of four pollutants measured daily at eight monitoring sites in London over a 4-year period. Such multiple-pollutant data sets measured over time at multiple sites within a region of interest are typical. Here, the modelling was carried out to provide the exposure for a study investigating the health effects of air pollution. Alternative objectives include the design problem of the positioning of a new monitoring site, or for regulatory purposes to determine whether environmental standards are being met. In general, analyses are hampered by missing data due, for example, to a particular pollutant not being measured at a site, a monitor being inactive by design (e.g. a 6-day monitoring schedule) or because of an unreliable or faulty monitor. Data of this type are modelled here within a dynamic linear modelling framework, in which the dependences across time, space and pollutants are exploited. Throughout the approach is Bayesian, with implementation via Markov chain Monte Carlo sampling.     

What the Practitioner Knows Versus What the Client is Told

This article revisits the question, pursued in an interchange between Thomas (1978) and Gambrill and Barth (1980), on whether conflicts exist between service goals and research goals in the use of single-system experimental designs (SSDs) for practice evaluation. The authors approach the issue through an analysis of how informed consent is handled in one widely used textbook promoting the SSD model of practice evaluation. They argue that the text’s suggestions for what to tell the client about reasons for using SSDs ignore knowledge acquisition goals emphasized elsewhere in the text, thus obscuring possible conflicts between the needs of the client and the requirements of SSD methodology. These failures of informed consent reveal largely neglected conflicts between practice and evaluation goals within the SSD model commonly taught to social work students.     

The Declining Significance of Race in Federal Civil Rights Law: The Social Structure of Employment Discrimination Claims

We examine changes in the nature and rate of complaints filed with the federal Equal Employment Opportunity Commission (EEOC) in the past 35 years. The EEOC's role has shifted over this period from ensuring job access for racial minorities to providing diverse protections for a much broader class of incumbent workers. We first describe trends in discrimination complaints, most notably the shift from racial discrimination to other bases of discrimination, and develop a conceptual model of choice among socially structured alternatives to account for them. We then test the model with a time series analysis of changes in the complaint rate among different worker groups to evaluate the relative importance of legal, political, and socioeconomic determinants of civil rights complaints. Net of changes in the political climate, benefit compensation, inequality, and education levels, we find that legal changes and group-specific unemployment rates are the strongest and most consistent determinants of the rate of race, sex, and total discrimination complaints. Our results suggest that people will bear the costs of filing a complaint when legal options are relatively attractive and when employment options on the external labor market are unattractive.     

Gibbs sampling for Bayesian non-conjugate and hierarchical models by using auxiliary variables

We demonstrate the use of auxiliary (or latent) variables for sampling non-standard densities which arise in the context of the Bayesian analysis of non-conjugate and hierarchical models by using a Gibbs sampler. Their strategic use can result in a Gibbs sampler having easily sampled full conditionals. We propose such a procedure to simplify or speed up the Markov chain Monte Carlo algorithm. The strength of this approach lies in its generality and its ease of implementation. The aim of the paper, therefore, is to provide an alternative sampling algorithm to rejection-based methods and other sampling approaches such as the Metropolis–Hastings algorithm.     

Efficient generation of random variates via the ratio-of-uniforms method

Improvements to the conventional ratio-of-uniforms method for random variate generation are proposed. A generalized radio-of-uniforms method is introduced, and it is demonstrated that relocation of the required density via the mode can greatly improve the computational efficiency of the method. We describe a multivariate version of the basic method and summarize a general strategy for efficient ratio-of-uniforms generation. Illustrative examples are given.