Statisticians in the social, biological, medical, and physical sciences. A week-long course on Common Errors in Statistics (and How to Avoid Them) sponsored by the Australian Consortium for Social and Political Research will be held in Brisbane at the University of Queensland, St Lucia Campus Monday 28 June - Friday 2 July 2004. See http://acspri.anu.edu.au/courses/winter/wp2004/index.html.
The course, based on the text of the same name by Phillip Good and James Hardin, attempts to fill in the practical details commonly omitted from introductory statistics courses on data collection, experimental design, estimation, hypothesis testing, the reporting of results, graphics, and modeling.
Class discussion is encouraged and many points raised in the course are sure to inspire it: For example,
o Do not report p-values for hypotheses suggested by the data
o Do not use the same data both to formulate models and to test them.
o Forget maximum likelihood; think minimum loss.
o Report bootstrap intervals rather than standard errors.
o Report confidence intervals rather than p-values.
The course will include demonstrations of CART, R, Resampling Stats, XLMiner, StatXact, and other programs designed to get the most from your data.
A complete course outline may be viewed at http://acspri.anu.edu.au/courses/winter/wp2004/wp2004CourseBooklet.pdf.
http.ms//www.statistician.usa
"Never trust anything that can think for itself if you can't see where it keeps its brain." JKR
Do you Yahoo!?
Win a $20,000 Career Makeover at Yahoo! HotJobs
