The Ethical Practice of Statistics for the Perplexed: A Graduate and Undergraduate Course in the Social and Behavioral Sciences

• Activity Number: 55
• Type: Invited
• Date/Time: Sunday, August 4, 2013 : 4:00 PM to 5:50 PM
• Sponsor: Committee on Professional Ethics
• Author(s): Lawrence James Hubert*+
• Companies: University of Illinois
• Keywords: ethical reasoning ; probability of causation ; probabilistic reasoning ; Bayes rule ; inferring causality ; meta-analysis

Abstract

A course will be described that is devoted to the ethical practice of statistics, along with the reasons for developing it, the challenges encountered while developing it, and solutions to the challenges. The ethical practice of statistics is defined as being in accord with the accepted rules and standards for right conduct that govern the discipline of statistics and its many areas of application. An emphasis is placed on the use of statistical and probabilistic reasoning in the social, behavioral, and biomedical sciences, and in relation to law and the judiciary. Topics include, among others, the use of Bayes Rule in screening for rare events; the importance of base rates; the probability of causation; inferring causality; meta-analysis; ethics in data collection; the Federal Rules of Evidence; experimental design and data interpretation; Simpson's paradox; statistical sleuthing and explanation; observational and controlled studies; prediction and correlation. The course is entitled "Statistics, Ethics, and the Social and Behavioral Sciences"; the primary text is "A Statistical Guide for the Ethically Perplexed" by Lawrence Hubert and Howard Wainer, Chapman & Hall, 2012.