KEELE UNIVERSITY
CENTRE FOR MEDICAL STATISTICS
SEMINAR SERIES on BIOSTATISTICS 2001/2002
NO 11: Wednesday, March 27th 2002, at 2:30 pm
"Randomize or else: what a Tesco trolley can teach us about statistics"
Professor J B Copas
Department of Statistics, University of Warwick
Selection bias, what happens if we don't randomize, is endemic to most
applications of statistics. Missing data, hidden confounding,
publication bias, non-ignorable censoring, non-compliance, are all
familiar examples. Usually we ignore the problem by making tacit and
untestable assumptions like "missing at random". But what happens when
such assumptions are wrong? The problem of selection bias cannot be
"solved", but at least we can think about it and try to develop good
ways of doing sensitivity analysis.
All welcome!!
Venue:
Room 2.22, Third Floor
MacKay Building
Keele University
http://www.keele.ac.uk
http://www.keele.ac.uk/depts/ma/seminars/medstats.html
http://www.keele.ac.uk/university/campus/maps/
*****************************
Janet Drewery
Centre for Medical Statistics
MacKay Building
Keele University
Keele, Staffs ST5 5BG
Tel: (01782) 583269
Fax: (01782) 583269/584268
E-mail: [log in to unmask]
*****************************
|