this paper I (1) examine three levels of inferential strength supported by typical social science data-gathering methods, and call for a greater degree of explicitness, when HMs and other models are applied, in identifying which level is appropriate; (2) reconsider the use of HMs in school effectiveness studies and meta-analysis from the perspective of causal inference; and (3) recommend the increased use of Gibbs sampling and other Markov-chain Monte Carlo (MCMC) methods in the application of HMs in the social sciences, so that comparisons between MCMC and better-established fitting methods---including full or restricted maximum likelihood estimation based on the EM algorithm, Fisher scoring or iterative generalized least squares---may be more fully informed by empirical practice.

Consider the conditionally independent hierarchical model (CIHM) where observations y i are independently distributed from f(y i j` i ), the parameters ` i are independently distributed from distributions g(`j), and the hyperparameters are distributed according to a distribution h(). The posterior distribution of all parameters of the CIHM can be efficiently simulated by Monte Carlo Markov Chain (MCMC) algorithms. Although these simulation algorithms have facilitated the application of CIHM's, they generally have not addressed the problem of computing quantities useful in model selection. This paper explores how MCMC simulation algorithms and other related computational algorithms can be used to compute Bayes factors that are useful in criticizing a particular CIHM. In the case where the CIHM models a belief that the parameters are exchangeable or lie on a regression surface, the Bayes factor can measure the consistency of the data with the structural prior belief. Bayes factors can ...

This paper addresses the use of weight functions to model publication bias in meta-analysis. Since this bias is hard to gauge, we introduce a non-parametric "-contamination class of weight functions. We then illustrate how to explore sensitivity of conclusions to the specification of the weight function by examining the range of results for the entire class. We find lower bounds on the coverage of confidence intervals. If no publication bias is present, results are robust even when considered over the entire "-contamination class. However, if publication bias is present, then the coverage provided by the usual interval estimator is not robust. In this case, an alternative interval estimator is suggested. We also illustrate how both upper and lower bounds on posterior quantities of interest may be found for the case in which prior information is available. Some key words: Weight functions; Selection bias; Meta-analysis. 1 Introduction This paper addresses the use of weight functions t...

This paper surveys technological changes that affect the collection, organization, analysis and presentation of data. It considers changes or improvements that ought to influence the research process and direct the use of technology. It explores implications for graduate research training. The insights of Evidence-Based Medicine are widely relevant across many different research areas. Its insights provide a helpful context within which to discuss the use of technological change to improve the research process. Systematic data-based overview has to date received inadequate attention, both in research and in research training. Sharing of research data once results are published would both assist systematic overview and allow further scrutiny where published analyses seem deficient. Deficiencies in data collection and published data analysis are surprisingly common. Technologies that offer new perspectives on data collection and analysis include data warehousing, data mining, new approaches to data visualization and a variety of computing technologies that are in the tradition of knowledge engineering and machine learning. There is a large overlap of interest with statistics. Statistics is itself changing dramatically as a result of the interplay between theoretical development and the power of new computational tools. I comment briefly on other developing mathematical science application areas -- notably molecular biology. The internet offers new possibilities for cooperation across institutional boundaries, for exchange of information between researchers, and for dissemination of research results. Research training ought to equip students both to use their research skills in areas different from those in which they have been immediately trained, and to respond to the ch...

Consider the exchangeable Bayesian hierarchical model where observations y i are independently distributed from sampling densities with unknown means, the means ¯ i are a random sample from a distribution g, and the parameters of g are assigned a known distribution h. A simple algorithm is presented for summarizing the posterior distribution based on Gibbs sampling and the Metropolis algorithm. The software program Matlab is used to implement the algorithm and provide a graphical output analysis. An binomial example is used to illustrate the flexibility of modeling possible using this algorithm. Methods of model checking and extensions to hierarchical regression modeling are discussed.

The Model Quality Report in Business Statistics project was set up to develop a detailed description of the methods for assessing the quality of surveys, with particular application in the context of business surveys, and then to apply these methods in some example surveys to evaluate their quality. The work was specified and initiated by Eurostat following on from the Working Group on Quality of Business Statsitics. It was funded by Eurostat under SUP-COM 1997, lot 6, and has been undertaken by a consortium of the UK Office for National Statistics, Statistics Sweden, the University of Southampton and the University of Bath, with the Office for National Statistics managing the contract. The report is divided into four volumes, of which this is the first. This volume deals with the theory and methods for assessing quality in business surveys in nine chapters following the survey process through its various stages in order. These fall into three parts, one dealing with sampling errors, one with a variety of non-sampling errors, and one covering coherence and comparability of statistics. Other volumes of the report contain: • a comparison of the software methods and packages available for variance estimation in sample surveys (volume II); • example assessments of quality for an annual and a monthly business survey from Sweden and the UK (volume III); • guidelines for and experiences of implementing the methods (volume IV). An outline of the chapters in the report is given on the following page. Acknowledgements Apart from the authors, several other people have made large contributions without which this report would not have reached its current form. In particular we would like to mention

INTRODUCTION: COMBINING ENVIRONMENTAL INFORMATION An important area of statistics concerns the combination of information from diverse sources relating to a common endpoint. Rich applications of data combination have been seen in ecotoxicology (Mastala et al., 1992; Warren-Hicks and Wolpert, 1994), water quality testing (Dominici et al., 1997), cancer epidemiology (Morris, 1994), and bio-clinical settings (Chalmers, 1991; Hasselblad et al., 1992). A common rubric for combining the results of independent epidemiological and clinical studies is meta-analysis (Hedges and Olkin, 1985). The goal of the methodology is to bring together results of different studies, re-analyze the disparate results within the context of their common endpoints, increase the sensitivity of the analysis to detect the presence of exposure effects, and provide a quantitative analysis of the phenomenon of interest based on the combined data. In th

this paper we investigate interval estimation of both the mean and heterogeneity variance

The wildlife component of DecAID is based on a thorough review, analysis, and synthesis of the empirical literature on wildlife-dead wood relations. We developed the wildlife component by compiling data on snag and log size, snag density, and amounts of down wood related to individual species or groups of wildlife species as presented in the literature, for various habitats and types of wildlife use (breeding, feeding, roosting). The wildlife use data are arranged in three cumulative species richness curves representing means and plus or minus one standard error (or equivalent variant). The curves can be consulted to determine which species or groups are provided for snag or down wood at three statistical levels, and the amounts and sizes of snags and down wood needed to achieve a specified wildlife objective of providing for specified species, or some percent of species, at a specified statistical level. Other components of the DecAID model can then be consulted to determine hazards or mitigation for risks of fire and contribution of insects and disease to the dead wood component, and to provide for fungi and non-pest invertebrates associated with snags and down wood.

Main entry under title: Statistical methods for adaptive management studies (Land management handbook; 42) Includes bibliographical references: p.