The magic number: what should be the standard for statistical significance?
David Weakliem
Professor, Dept. of Sociology, University of Connecticut
February 23rd 11:30-1:00
Gentry Room 144
A t-ratio of approximately two (P=.05) is almost universally used as the standard for “statistical significance.” However, it has few defenders and many critics. Some critics hold that it is not strict enough, and that we should adopt a standard of about 3; others hold that it should differ depending on some features of the individual problem. In this talk, I discuss the choice of a standard for statistical significance. I consider the Neyman-Pearson approach of balancing the “cost” of Type I and Type II errors. I then consider the Bayesian approach, which requires more precise specification of the alternative hypothesis. I conclude by discussing implications for the practice of research.
David L. Weakliem, PhD, is Professor of Sociology at the University of Connecticut. He has been a fellow at the Center for Advanced Study in the Behavioral Sciences at Stanford University and at the Australian National University. Dr. Weakliem is Editor-in-Chief of Comparative Sociology and a past Deputy Editor of the American Sociological Review. Dr. Weakliem is also the author of Hypothesis Testing and Model Selection in the Social Sciences.
For more information, contact: Betsy at betsy@uconn.edu