Dear all,
Circular statistics are stats for angles, for seasons, or for time of
day - anything that is circular.
OUR PROBLEMS:
(1) We have a number of patients, each of whom have a series of 5 tests
and we wish to compare the ANGLE that they set off at, between the tests
(this is like a 2 way ANOVA, with person id, and type of test, but we
want a version that works with ANGLES, i.e. taking account of the fact
that 0 and 360 are the same thing).
(2) We would also like to compare the variability in angle between 2
sets of measurements on the SAME patients, i.e again on paired data
(CIRCULAR stats version of Pitmans test)
Books for those interested which solve other problems in circular
stats:
Topics in circular statistics. Jammalamadaka and SenGupta, World Sci
Publishing Co Pte Ltd, Singapore 2001.
Circular statistics in biology. Edward Batschelet. Academic Press Inc
(London), subsid of Harcourt Brace Jovanovich publishing, 1981
Many thanks in advance for your help
best wishes
Hilary
Hilary Watt
Lecturer in Statistics
Medical Statistics Unit
London School of Hygiene and Tropical Medicine
Keppel Street, London, WC1E 7HT
Tel: 020 7927 2169
Fax: 020 7637 2853
E-mail: [log in to unmask]
HONORARY LECTURER at the INSTITUTE OF NEUROLOGY
Part Funded by the Institute of Neurology to provide
Statistical consultancy members of the Institute
My ION intranet web pages for my teaching material and courses, links
to
other courses and other teaching material, including material on SPSS.
http://www.ion.ucl.ac.uk/intranet/admin_services/stats_teaching.htm
|