Kline (psychology, Concordia U., Montreal) reviews the literature about shortcomings in significance testing, and argues that its shows that major changes are needed in how researchers analyze data and report findings. Then he helps researchers and students in psychology and other behavioral sciences who do not have strong quantitative background acquire new skills for estimating interval and effect size, and reviews alternative ways to test hypotheses, including Bayesian estimation. The first edition was published in 2004, and this second adds robust statistical methods for estimating parameter, effect size, and interval. The literature review is also brought up to date. Annotation ©2013 Book News, Inc., Portland, OR (booknews.com)