Held on June 14, 2007 at the Skokie Doubletree Hotel & Conference Center.
The 23rd Annual Summer Workshop of the Northeastern Illinois Chapter of the American Statistical Association was given on the topic of:
Sample-Size Analysis for Study Planning
Dr. Ralph O'Brien, The Cleveland Clinic Foundation
Summary
Performing a conscientious sample-size analysis is often a cornerstone of effective study planning, because the research team (including the statistician) must delineate, critique, and tighten the specific aims, research questions, design, measurements, and analysis plans. This highly-interactive course uses realistic case studies to cover the scientific/statistical principles and general consulting pragmatics associated with such work. Issues of statistical science are stressed far more than mathematical and computational underpinnings. Although all analyses covered are traditional (frequentist), we show how a simple Bayesian notion ("The March of Science") helps to clarify and shape the planning process for investigators, statisticians, and peer reviewers. All examples employ SAS9 procedures and freeware, but participants also see how the same calculations are performed using PASS and nQuery Advisor along with a simple Excel spreadsheet developed/distributed by the instructor. Much of this course is based on O'Brien RG, Castelloe JM (2007), “Sample-Size Analysis for Traditional Hypothesis Testing: Concepts and Issues,” in Dmitrienko A, Chuang-Stein C, D’Agostino R (Ed.), Pharmaceutical Statistics Using SAS: A Practical Guide, Cary, NC: SAS Press, 237-271.
Biographical Background
Dr. Ralph O’Brien is currently a professor in the Department of Quantitative Health Sciences and the Lerner College of Medicine at the Cleveland Clinic Foundation. His career in statistics is broad, beginning in behavioral statistics (Psychology Dept., University of Virginia, 1975-82), then moving more to the mainstream (Statistics Dept., University of Tennessee, 1982-89), and then to biomedicine (Division of Biostatistics, Department of Statistics, University of Florida, 1989-94), where Dr. O’Brien was Director of Biostatistics. He is the creator of “O’Brien’s test” for variance differences, performed by both SAS® (PROC GLM) and JMP®. However, his main contributions have been in promoting more effective sample-size analyses for study planning. His freeware module, UnifyPow.sas, is used all over the world, and he now consults intensively with SAS personnel who are developing PROCs POWER and GLMPOWER and other such tools. He has given over 25 workshops on sample-size analysis to over 2000 people, and his continuing education courses at annual Joint Statistical Meetings earned him an “Excellence-in-CE” award from the American Statistical Association.