Alferes (psychology, U. of Coimbra, Portugal) presents a guide to randomization in experimental design, "the devices required for an unbiased and efficient estimation of treatment causal effects." The text is written for undergraduates, graduates and professionals in the social, educational, behavioral and health science. It attends more to "internal validity and some specific aspects of statistical conclusion validity and only tangentially to data analysis or statistical testing." The author introduces the nature and structure of randomized experiments, examines the limits imposed on random assignment, and reviews the structural properties of within-subject designs. A final section also offers considerations for planning, monitoring, analyzing and reporting a randomized experiment. There are appendices on random-number tables, permutations and Latin squares. Annotation ©2012 Book News, Inc., Portland, OR (booknews.com)