This book provides a comprehensive treatment of the logic behind hypothesis testing. Readers will learn to understand statistical hypothesis testing and how to interpret P-values under a variety of conditions including a single hypothesis test, a collection of hypothesis tests, and tests performed on accumulating data. The author explains how a hypothesis test can be interpreted to draw conclusions, and descriptions of the logic behind frequentist (classical) and Bayesian approaches to interpret the results of a statistical hypothesis test are provided. Both approaches have their own strengths and challenges, and a special challenge presents itself when hypothesis tests are repeatedly performed on accumulating data. Possible pitfalls and methods to interpret hypothesis tests when accumulating data are also analyzed. This book will be of interest to researchers, graduate students, and anyone who has to interpret the results of statistical analyses. . Presents the classical and Bayesian interpretations of hypothesis tests under various conditions. Describes how to take samples and the process for planning to take samples. Explains how to use Excel® to obtain P-values for several common statistical tests and how to interpret them

Autorentext

Dr. Hirsch retired as Professor of Epidemiology and Biostatistics and Adjunct Professor of Statistics at The George Washington University in 2010. He started at GWU in 1987, with the principal task of helping to develop a public health program. That program is now the School of Public Health and Health Services with more than 300 graduate students enrolled. One of Dr. Hirsch's contributions to this program was the development of courses in public health methodology. These courses continue to be an important part of the curriculum. They include courses in statistics, study design, evaluation of health research literature, questionnaire development, sampling, meta-analysis, and theoretical epidemiology. Dr. Hirsch is probably best known for his textbooks and other publications that are intended to help people without a mathematics aptitude to understand quantitative issues that are integral to health research and practice. His ability to communicate technical information has been an important asset in his teaching and consulting work. His interest in relating statistics to real life problems prompted him to spend a sabbatical at the Center for Devices and Radiologic Health in the Food and Drug Administration. The purpose of this sabbatical was to understand how health methodology can best contribute to governmental regulatory issues. The result of this experience has also given Dr. Hirsch the ability to serve as an interface between industry and the government as well as train graduate students in real-world applications.

Inhalt

Statistical Hypotheses.- Frequentist Approach.- Bayesian Approach.- Testing Accumulating Data.