Statistical Methods for Psychology

For advanced undergraduates and graduate students, Howell (psychology, U. of Vermont) surveys statistical techniques commonly used in the behavioral and social sciences, especially psychology and education. He explains material on an intuitive, rather than technical, level and emphasizes looking at the data before jumping in with a hypothesis test and the importance of the relationship between the statistical test to be used and the theoretical questions being posed by the experiment. Students do not need any previous knowledge of statistics or math beyond high-school algebra. This edition includes more on effect size and magnitude of effect, new material on graphical displays, expanded discussion of resampling, a narrower range of tests in the chapter on multiple comparison techniques and inclusion of Benjamini and Hochberg's False Discovery Rate. It extends discussion of mixed models and replaces the chapter on log-linear models with material on meta-analysis and single subject designs. The older versions of the chapters on multiple comparison tests or log-linear models are available on the author's website, which also includes programs or selections of R code, additional examples and topics, more data, demonstrations, and other features. Annotation c2012 Book News, Inc., Portland, OR (booknews.com)