Capture-Recapture Methods for the Social and Medical Sciences

"This is a timely, important book on the use of capture-recapture methods for social and medical data. Several books have been written on capture-recapture methods for ecology, over many years, and one focussing on social and medical applications has been long overdue. This book illustrates the power of appropriate capture-recapture analyses in areas other than ecology. Several of the book chapters describe new methods, and suggest avenues for future research. The relevance of the methods described is evident, with applications to studies of the prevalence of scrapie, and estimating numbers of injecting drug users, of immigrants, and of victims of domestic violence, etc. Time and again we see the power of Statistics in providing answers to really important questions. I enjoyed reading this book enormously. A great attraction is the wide range of motivating examples, complete with data, which include several from ecology. The way that methods are regularly illustrated on both real and simulated data is engrossing. Models are clearly described and accessible. The book should be required reading, for years to come, for any university course on Applied Statistical Modeling, as well as being a vital reference for research. I am sure that this book will be much read, and make a major impact."From the Foreword by Byron J. T. Morgan "Capture-Recapture Methods for the Social and Medical Sciences is a much-needed text focussing on social science applications and methods of capturerecapture modelling. Theory within the field of analysis of capturerecapture data has developed in a disparate fashion between areas of applications, and many texts have focused on the presentation of ecological applications. This book does an excellent job of approaching the field from a different perspective, whilst still retaining how ecological applications fit into the developments. The book has been written by over 40 contributors, each an expert in their own area, which means that the theory presented is cutting-edge but has been presented in an accessible style...I will certainly be recommending this book to my collaborators and research students, as it provides a unique perspective, which is lacking in other capture-recapture books currently available on the market. Although this is a text that most likely will be dipped in and out of for reference to a particular topic, the book could be read cover-to-cover and anyone completing this can be sure of being up to date on the latest modelling approaches for capturerecapture data in the social sciences."-Rachel S. McCrea, Univeristy of Kent "Capture-Recapture Methods for the Social and Medical Sciences, edited by Böhning, van der Heijden, and Bunge is the first book explicitly centered on applications of capture-recapture methods to human populations and medical problems...The most interesting aspect of this book is the example applications. Most of them are taken from outside the classical setting of estimating animal abundance, and illustrate well the applicability of CR methods to diverse domains in social science and medicine...Capture-Recapture Methods for the Social and Medical Sciences deserves praise for being the first of its class. It is a good catalog of single-list CR methods, and contains some limited information about multi-list methods. This book can be useful for raising awareness among applied researchers of the applicability of capture-recapture methods for estimating human populations."-Daniel Manrique-Vallier, The American Statistician, Volume 34, 2020

Auteur

Dankmar Böhning is Professor of Medical Statistics and Director of the Southampton Statistical Sciences Research Institute at the University of Southampton. His interests are in capture-recapture modelling, meta-analysis and research synthesis as well as mixed modelling.

John Bunge is Professor of Statistics in the Department of Statistical Science of Cornell University. His interests are capture-recapture modelling, microbiome statistics, and nonclassical probability distribution theory.

Peter. G.M. van der Heijden is Professor of Social Statistics at the University of Utrecht and at the University of Southampton. His interests are capture-recapture modelling for the Social Sciences and Official Statistics.

Texte du rabat

Capture-recapture methods have recently become popular in the social and medical sciences to estimate the size of elusive populations such as illicit drug users or people with a drinking problem. This book brings together important developments which allow the application of these methods with contributions from more than 40 researchers.

Contenu

I Introductory Part

Basic concepts of capture-recapture

II Ratio Regression Models

Ratio regression and capture-recapture

The Conway-Maxwell-Poisson distribution and capture-recapture count data

The geometric distribution, the ratio plot under the null and the burden of Dengue Fever in Chiang Mai province

A ratio regression approach to estimate the size of the Salmonella infected flock population using validation information

III Meta-Analysis in Capture-Recapture

On meta-analysis in capture-recapture

A case study on maritime accidents using meta-analysis in capture-recapture

A meta-analytic generalization of the Lincoln-Petersen-estimator for mark-and-resight studies

IV Extensions of Single Source Models

Estimating the population size via the empirical probability generating function

Convex estimation

Non-parametric estimation of the population size using the empirical probability generating function

Extending the truncated Poisson regression model to a time-at-risk model

Extensions of the Chao-estimator for covariate information: Poisson case

Population size estimation for one-inflated count data based upon the geometric distribution

V Multiple Sources

Dual and multiple system estimation: fully observed and incomplete covariates

Population size estimation in CRC Models with continuous covariates

Trimmed dual system estimation

Estimation of non-registered usual residents in the Netherlands

VI Latent Variable Models

Population size estimation using a categorical latent variable

Latent class - Rasch models and marginal extensions

Performance of hierarchical log-linear models for a heterogeneous population with three lists

A multidimensional Rasch model for multiple system estimation where the number of lists changes over time

Extending the Lincoln-Petersen estimator when both sources are counts

VII Bayesian Approaches

Objective Bayes estimation of the population size using Kemp distributions

Bayesian population size estimation with censored counts

VIII Miscellaneous Topics

Uncertainty assessment in capture-recapture studies and the choice of sampling effort