Experimental and Quasi-Experimental Designs for Generalized Causal Inference

Experimental and Quasi-Experimental Designs for Generalized Causal Inference, by W. R. Shadish, T. D. Cook, and D. T. Campbell (2001). Boston: Houghton Mifflin.

Research methods, including research design, is one of the key areas for the training of graduate students in industrial and organizational psychology, organizational behavior, management, and a host of other fields. In addition, a solid understanding of research methods is vital to the design and conduct of basic and/or applied research. In view of this, comprehensive and authoritative books on research methods are an invaluable resource to both students and professionals.

Although a large number of books on research methods have been published, only a handful approach the overall quality of Shadish, Cook, and Campbell's (2001 ) book. It is the third of a series of works that have long been viewed as among the best books on research design, the first two being Campbell and Stanley's (1963) Experimental and Quasi-Experimental Designs for Research and Cook and Campbell's (1979) Quasi-Experimentalion: Design and Analysis Issues for Field Settings. Cook and Campbell's (1979) revision of Campbell and Stanley's (1963) book represented a major overhaul of it. Following in that tradition, Shadish et al.'s (2001 ) book represents a substantial revision of Cook and Campbell (1979).

The main focus of Shadish et al. (2001) is nonexperimental, quasi-experimental, and experimental strategies for conducting research. These are covered in 14 chapters, the first of which is concerned with the value of experiments for generalized causal inference. The next two deal with inferences about four types of validity (i.e., statistical conclusion, internal, construct, and external) that are affected by the design and conduct of research. The following four chapters are devoted to several classes of quasi-experimental designs (i.e., those without pretests or control groups, those that use both control groups and pretests, those that involve interrupted time-series designs, and those of the regression discontinuity variety). The book then offers three chapters on randomized experimental designs and a host of issues (e.g., ethics, participant recruitment, treatment implementation, and attrition) associated with research using such designs. The next three chapters consider a set of issues related to generalized causal inference (i.e., a grounded theory for generalized causal inference, causal inference strategies for single studies, and generalized causal inference strategies for...