- Gebundene Ausgabe: 392 Seiten
- Verlag: Sage Publications Ltd.; Auflage: 1 (1. Oktober 2009)
- Sprache: Englisch
- ISBN-10: 1412953561
- ISBN-13: 978-1412953566
- Größe und/oder Gewicht: 23,1 x 15,5 x 2,3 cm
- Durchschnittliche Kundenbewertung: Schreiben Sie die erste Bewertung
- Amazon Bestseller-Rang: Nr. 67.033 in Fremdsprachige Bücher (Siehe Top 100 in Fremdsprachige Bücher)
- Komplettes Inhaltsverzeichnis ansehen
Propensity Score Analysis: Statistical Methods and Applications (Advanced Quantitative Techniques in the Social Sciences) (Englisch) Gebundene Ausgabe – 1. Oktober 2009
Dieses Buch gibt es in einer neuen Auflage:
Kunden, die diesen Artikel angesehen haben, haben auch angesehen
Welche anderen Artikel kaufen Kunden, nachdem sie diesen Artikel angesehen haben?
Es wird kein Kindle Gerät benötigt. Laden Sie eine der kostenlosen Kindle Apps herunter und beginnen Sie, Kindle-Bücher auf Ihrem Smartphone, Tablet und Computer zu lesen.
Geben Sie Ihre Mobiltelefonnummer ein, um die kostenfreie App zu beziehen.
Wenn Sie dieses Produkt verkaufen, möchten Sie über Seller Support Updates vorschlagen?
"The approach the authors take in writing this book is very effective for novices and experiences users...This balance between the practical and applied approach is a useful model for researchers to understand the process and interpretation of these analyses...[it] goes a long way in making propensity score analysis techniques more accessible, understandable, and useful to psychologists." -- Karl N. Kelley PsycCRITIQUES 20110706 "Guo and Fraser's book Propensity Score Analysis: Statistical Methods and Applications is the first comprehensive book that discusses and compares different PS techniques from theoretical and practical points of view. One of the book's strengths is its focus on the application of PS to real data. [T]his textbook gives a good introduction to PS matching techniques and some alternative approaches for estimating causal treatment effects. With its many examples in Stata, it may be useful for graduate students and applied researchers who have no or limited experience with PS methods but are familiar with basic regression methods and mathematical/statistical notation." -- Peter M. Steiner PSYCHOMETRIKA-VOL. 75, NO. 4, 775-777 20101208
Über den Autor und weitere Mitwirkende
Shenyang Guo, PhD, is the Kuralt Distinguished Professor at the School of Social Work, University of North Carolina. The author of numerous articles on statistical methods and research reports in child welfare, child mental health services, welfare, and health care, Guo has expertise in applying advanced statistical models to solving social welfare problems and has taught graduate courses on event history analysis, hierarchical linear modeling, growth curve modeling, and program evaluation. He has given many invited workshops on statistical methods-including event history analysis and propensity score matching-at the NIH Summer Institute, Children's Bureau, and at conferences of the Society of Social Work and Research. He led the data analysis planning for the National Survey of Child and Adolescent Well-Being (NSCAW) longitudinal analysis. Mark W. Fraser, PhD, holds the Tate Distinguished Professorship at the School of Social Work, University of North Carolina at Chapel Hill, where he serves as associate dean for research. He has written numerous chapters and articles on risk and resilience, child behavior, child and family services, and research methods. With colleagues, he is the co-author or editor of eight books, including Families in Crisis, Evaluating Family-Based Services, Risk and Resilience in Childhood, Making Choices, The Context of Youth Violence, and Intervention with Children and Adolescents. His award-winning text Social Policy for Children and Families reviews the bases for public policy in child welfare, juvenile justice, mental health, developmental disabilities, and health. His most recent book, Intervention Research: Developing Social Programs, describes a design perspective on the development of innovative social and health programs.
Die hilfreichsten Kundenrezensionen auf Amazon.com
Classic experimental design would say that the answers are unknowable. Without random assignment, one cannot determine a treatment effect. Without sample equivalence, one cannot compare groups. Without random sampling, assignment, control, and longitudinal measurement, one can say nothing about causation. However, over the past four decades, several groups of researchers in public policy, statistics, and econometrics have developed a family of similar methods to address these issues.
It turns out that random assignment is not always required, that one can say a lot about treatment effects given naturally occurring variation. And similarly, even when samples are drawn from distinct subpopulations, one can tease apart at least some of the treatment effect from the contributions of sample differences (bias). The key is to use other information that is available to help control and reduce bias, and to make the contribution of treatment effect as unbiased as possible.
So, what about the book? In a nutshell, it shows how to do just those kinds of models. It provides an outstanding overview of the theoretical issues and general structure of methods from Heckman, Rubin, Pearl, and others. It provides particular depth on the Heckman approach in econometrics, the Rubin causal model that is perhaps more approachable for social researchers, and the key differences between the two. The illustrative examples are extremely well-chosen, and at several points the basic concepts are illustrated with brilliantly clear, small data sets that can be appreciated in simple presentation in the book itself. In each case, it discusses both the mathematics behind the method (which can be comfortably skipped by those so inclined) and available software to estimate it.
It is more readable than 90% of statistics texts (I would estimate), and I expect it could be appreciated by most folks with a PhD related to any social or economics research field.
The main drawback to it, in my opinion, is that the software discussion primarily focuses on Stata. There is somewhat less coverage of R, which is a shame because (IMHO) R is quickly becoming the platform of choice for advanced researchers who use emerging methods. However, there is enough discussion of R analyses to get one started, and the Stata analyses are syntactically quite similar anyway. So this is a limitation but is not severe.
If you are a researcher interested in causal models or measurement of effects in non-experimental settings, and want an applied book rather than a math tome, this is the right book to get started. Thank you to the authors!
The book is very very useful. It provides a good framework about PSM and how to actually DO IT!
The examples are very illustrative and helpful. Also, the author responds to emails :D Very nice!
The only drawback that I see now is that the book did not discuss much about the shortcomings of PSM.
Excellent book! I highly recommend to those who need to use PSM.
Ähnliche Artikel finden