Read more
Informationen zum Autor Peter Davis is Director of the COMPASS Research Centre and Professor of Sociology at the University of Auckland! with cross-appointments in the School of Population Health and in the Department of Statistics! also at the University of Auckland. Previously he served as Professor of Public Health at the University of Otago's Christchurch School of Medicine. Davis specialises in medical sociology! and has achieved international recognition in his field! having worked as a consultant for the World Health Organisation. His main interests are in research methods! social structures! and policy! particularly health policy and health services. He has collaborated with colleagues in health research and in social statistics on a number of major surveys since the 1970s. He was Senior Editor (Health Policy) at the international journal! Social Science and Medicine! until 2012. Klappentext Most social research is carried out in observational settings; that is, most social researchers collect information in the "real world" trying to do as little possible to alter the circumstances of study. However, there is a fundamental problem with this kind of research, in that it is very hard to draw "causal" conclusions, because of the complexity and obduracy of social reality. This is not just a problem for social scientists interested in policy or social action. It applies across the board more generally because it becomes difficult to know, without the conditions for credible inference, what conclusions can be drawn from any piece of empirical research that aspires to be anything more than descriptive of social phenomena. This four-volume set of readings introduces the reader to the advances that have been made in trying to help social researchers draw more credible inferences from investigations carried out in observational settings. Drawing from a variety of sources - from logicians and philosophers, to applied statisticians, computer scientists and econometricians, to epidemiologists and social researchers - this collection provides an invaluable resource for scholars in the field. Volume One: Background Volume Two: Analytical Techniques Volume Three: Temporal Relations Volume Four: Experimental Analogues In social science research, oftentimes, the researcher's ultimate goal is to be able to make causal inference statements about what would contribute to socially significant outcomes. Unfortunately, not being able to implement true experimental design in most social science research situations makes such causal inference risky and full of pitfalls, as it can become very difficult to rule out rival hypotheses or explanations. This collection of seminal papers on issues related to making causal inferences provides a "must read" for social science researchers, green hand or experienced alike, who desire to avoid numerous pitfalls in the process of making causal inferences in social science research. Zusammenfassung This four-volume set of readings introduces the reader to the advances that have been made in trying to help social researchers draw more credible inferences from investigations carried out in observational settings Inhaltsverzeichnis VOLUME ONE: BACKGROUND PART ONE: CAUSAL INFERENCE FROM OBSERVATIONAL DATA Estimating Causal Effects of Treatments in Randomized and Non-Randomized Studies - Donald Rubin Statistics and Causal Inference - Paul Holland Misunderstandings between Experimentalists and Observationalists about Causal Inference - Kosuke Imai et al The Estimation of Causal Effects from Observational Data - Christoper Winship and Stephen Morgan Causal Inferences in Sociological Research - Markus Gangl PART TWO: POTENTIAL OUTCOMES AND COUNTERFACTUALS On the Application of Probability Theory to Agricultural Experiments - Jerry Splawa-Neyman, ...
