There is something called the Fligner-Killeen test, which has a bit of
messy arithmetic but is nevertheless straightforward. It and numerous other
tests for differences in variation (actually relative variation) are
documented and carefully compared w.r.t. Type I and Type II error in:
Steven M. Donnelly and Andrew Kramer. 1999. "Testing for multiple species
in fossil samples: An evaluation and comparison of tests for equal relative
variation." American Journal of Physical Anthropology. 108: 507-529.
=-=-=-=-=-=-=-=-=-=-=-=-=
Mike Lacy
Fort Collins CO USA
(970) 491-6721 office
