What If There Were No Significance Tests?: Classic Edition

The overriding theme of this book is that significance testing may still be useful if supplemented with all or some other methods such as Bayesian logic, confidence intervals, effect sizes and power, goodness of approximation measures, replication and meta-analysis, and theory appraisal and corroboration.An overview of significance testing is provided along with debates as to whether or not significance testing should be rejected or retained and methods that may supplement current significance testing procedures including confidence intervals. This book is ideal for advanced students and researchers in the behavioral and social sciences.