Statistical Models for Causal Analysis

Simplifies the treatment of statistical inference focusing on how to specify and interpret models in the context of testing causal theories. Simple bivariate regression, multiple regression, multiple classification analysis, path analysis, logit regression, multinomial logit regression and survival models are among the subjects covered. Features an appendix of computer programs (for major statistical packages) that are used to generate illustrative examples contained in the chapters.

Free of unwieldy mathematics, Statistical Models for Causal Analysis provides a lucid introduction to statistical models used in the social and biomedical sciences, particularly those models used in the causal analysis of nonexperimental data. Featuring an approach that focuses on model specification and interpretation, this innovative work-designed specifically for students and professionals in need of a working knowledge of the subject - is a practice-oriented guide to learning how to use these models in analytical work. Based on a highly successful classroom course, Statistical Models for Causal Analysis includes computer programs implementable on either mainframe computers or microcomputers as well as examples taken from an actual population study. The book provides not only a clear understanding of principles of model construction but also a working knowledge of how to implement these models using real data. Topics covered are bivariate linear regression, multiple regression, multiple classification analysis, path analysis, logit regression, multinomial logit regression, survival models (including proportional hazard models and hazard models with time dependence). While omitting a good deal of difficult mathematics, such as derivations of sampling distributions and standard errors, the book nonetheless provides a rigorous and focused examination of model specification and interpretation, illustrating their application to the kinds of research that social and biomedical scientists undertake. Supported by numerous tables and graphs, using real survey data, as well as an appendix of computer programs for the statistical packages SAS, BMDP, and LIMDEP, the book is an ideal primer forunderstanding and using statistical models in analytical work. Eminently clear and highly practical, Statistical Models for Causal Analysis is essential for social science and biomedical professionals wishing to upgrade their methodological skills and students in need of a challenging, yet simplified treatment, of these useful, versatile models that have become essential tools for the modern researcher in these fields.

Includes bibliographical references (p. 251-253) and index.

Bivariate Linear Regression.

Multiple Regression.

Multiple Classification Analysis.

Path Analysis.

Logit Regression.

Multinomial Logit Regression.

Survival Models 1: Life Tables.

Survival Models 2: Proportional Hazard Models.

Survival Models 3: Hazard Models with Time Dependence.

Appendices.

References.

Index.