Measuring the complexity of generalized linear hierarchical models

Measuring a statistical model's complexity is important for model criticism and comparison. However, it is unclear how to do this for hierarchical models due to uncertainty about how to count the random effects. The authors develop a complexity measure for generalized linear hierarchical models based on linear model theory. They demonstrate the new measure for binomial and Poisson observables modeled using various hierarchical structures, including a longitudinal model and an areal‐data model having both spatial clustering and pure heterogeneity random effects. They compare their new measure to a Bayesian index of model complexity, the effective number p_D of parameters (Spiegelhalter, Best, Carlin & van der Linde 2002); the comparisons are made in the binomial and Poisson cases via simulation and two real data examples. The two measures are usually close, but differ markedly in some instances where p_D is arguably inappropriate. Finally, the authors show how the new measure can be used to approach the difficult task of specifying prior distributions for variance components, and in the process cast further doubt on the commonly‐used vague inverse gamma prior.     

Informing the Public About Uncertain Risks

Many problems arise in connection with the communication of risk information. In this article the content of the information communicated is taken as a starting-point for analyzing the risk communication process. We studied the way in which authorities communicated health risks to local residents in Dutch soil pollution situations and found that communication problems were characterized more by misunderstanding and conflict between parties with different views and interests than by a lack of understanding. Therefore, it is important to look more closely at the way the communicators of information (in our cases, officials) select risk information from risk assessments, and to study the effects that the information selected has on the receivers (in our cases, residents). Both the process of selection by the officials and the process of interpretation by the residents will be shown to be influenced by the different institutional backgrounds. The article presents a new approach to risk communication. Both the risk information presented by the authorities and the public reactions to this risk information are considered to reflect the institutional background of authorities and residents, and can be analyzed in these terms. Such an approach has consequences for the study of risk communication and the manner in which it is practiced.     

Ascertaining the effect of planned organizational change on the work environment

In a dynamic environment, organizations often are required to effect major changes in operations. The success of such changes invariably depends upon numerous factors one of which is how the change affects the work environment. Therefore, in planning organizational change it is important to consider the work environment and to objectively evaluate changes in the work environment in a timely fashion so that corrective actions may be formulated if needed. Such framing and monitoring of the organizational change facilitates successful implementation of the organizational change. This paper suggests a measure which may be used in evaluating an organization's work environment. A case is discussed.     

A Stochastic Assessment of the Public Health Risks of the Use of Macrolide Antibiotics in Food Animals

Campylobacteriosis is an important food-borne illness with more than a million U.S. cases annually. Antibiotic treatment is usually not required. However, erythromycin, a macrolide antibiotic, is recommended for the treatment of severe cases. Therefore, it is considered a critically important antibiotic and given special attention as to the risk that food animal use will lead to resistant infections and compromised human treatment. To assess this risk, we used a retrospective approach; estimating the number of campylobacteriosis cases caused by specific meat consumption utilizing the preventable fraction. We then determined the number of cases with macrolide resistance Campylobacter spp. based on a linear model relating the resistance fraction to on-farm macrolide use. In this article, we considered the uncertainties in the parameter estimates, utilized a more elaborate model of resistance development and separated C. coli and C. jejuni . There are no published data for the probability of compromised treatment outcome due to macrolide resistance. Therefore, our estimates of compromised treatment outcome were based on data for fluoroquinolone-resistant infections. The conservative results show the human health risks are extremely low. For example, the predicted risk of suboptimal human treatment of infection with C. coli from swine is only 1 in 82 million; with a 95% chance it could be as high as 1 in 49 million. Risks from C. jejuni in poultry or beef are even less. Reduced antibiotic use can adversely impact animal health. These low human risks should be weighed against the alternative risks.     

Improving the Efficiency of the Nelson–Aalen Estimator: the Naive Local Constant Estimator

Abstract.  The Nelson–Aalen estimator is well known to be an asymptotically efficient estimator of the cumulative hazard function, see Andersen et al. ( Statistical models based on counting processes , Springer-Verlag, New York, 1993) among many others. In this paper, we show that the efficiency of the Nelson–Aalen estimator can be considerably improved by using more information in the estimation process than the traditional Nelson–Aalen estimator uses. While our approach results in a biased estimator, the variance improvement is substantial. By optimizing the balance between the bias loss and the variance improvement, we obtain results on the efficiency gain. Several examples for known failure time distributions are used to illustrate these ideas.     

Perspectives on Regional and Enterprise Marginality: Dairying in Michigan's North Country1

Abstract Dairy farmers in Michigan's Upper Peninsula, an economically marginal rural region, are encumbered by contextual constraints that are not similarly faced by their downstate counterparts. However, this does not mean that they are any less ambitious or less efficient in dealing with the resources available to them. Indeed, dairying and similar small-scale economic enterprises in this and other marginal contexts may be quite competitive and sustainable if appropriate sociopolitical supports are instituted and, in the case of dairying, if marketing orders and regulatory policies are formulated and implemented in light of their impact upon the economic growth and viability of the wider region. To explore this issue, we consider the situations, basic characteristics, and restructuring trends over time of dairy farms in three upstate localities as compared with that of a downstate dairy farming community.     

The Development of Urban Neighbourhoods and the Benefit of Indication Systems

The last few years much use is being made ofso-called indication systems to map the development ofneighbourhoods. It is hoped that these systems quicklybring possible livability problems to light, so thatpeople can intervene in accordance with policy.In this article the advantages and pitfalls of such anindication system are gone into. Such a system has totake into account the fact that the development ofneighbourhoods does not have a fixed pattern.Furthermore the much-used objective indicators oftendo not mesh with the subjective perception ofresidents and the appraisal processes ofprofessionals. Also much more consideration will haveto be given to social characteristics, which in ourday have become increasingly more determining for thelivability of neighbourhoods. It is also importantthat the action of managers be part of acharacterization of neighbourhoods. It is being arguedthat an indication system cannot be confined toregistering indicators and other data, for it has todirectly contribute to policy interventions. That iswhy a good indication system takes both facts and creation of an image into account, as well aslivability characteristics and theconsiderations of residents and managers. In a fewbrief concluding observations, the different functionsan indication system can have as a policy instrumentare gone into.     

Propensity to adopt innovative housing: Development of knowledge indexes

The process by which energy efficient housing alternatives is diffused through society in relation to the information flow is relatively unknown. The major purpose of this study was to validate a continuum of the propensity to adopt alternative housing (passive and active solar, earth sheltered and retrofitted) based on knowledge level. Findings indicate knowledge indices are valid predictors of consumer acceptance of energy efficient housing alternatives. A greater level of knowledge was correlated with respondents willing to adopt an alternative housing type.