“Statisticians Behaving Badly:
Common Statistical Pitfalls & How to Avoid Them"
Discussion will focus on commonly observed statistical errors and what can be done to correct them. Topics will include the over-reliance on null hypothesis testing, the problem of multiplicity, the inappropriate use of stepwise variable selection, the misuse of multivariate analysis of variance, the overreliance on linear models, disrespecting continuous variables, use of non-descriptive statistics, inappropriate use of change scores, and abuse of graphics. Participants are encouraged to bring in their own examples of statistical pitfalls.
Scott R. Millis, PhD, MEd is Professor in the Department of Physical Medicine & Rehabilitation at Wayne State University School of Medicine and the Director of Research at the Rehabilitation Institute of Michigan. Dr Millis is a Chartered Statistician (CStat) with the British Royal Statistical Society and Chartered Scientist (CSci) with the British Science Council. He is also board certified in three psychological specialties by the American Board of Professional Psychology: Clinical Neuropsychology, Clinical Psychology, and Rehabilitation Psychology.