Hello.
Some numbers (observed):
10 482
15 1974
I've a problem with whether to use chi-square or Fisher's exact test.
in a 2x2 contingency table (e.g.above), when, ONE of the expected values is
below 5, making 25% of the observed values less than 5 - so according to
Cochran (1954), the Chi-square value is then biased therefore not legitimate.
BUT, if I use Fisher's exact test (actually one of the plug in no's and
calculate versions on the web: http://depts.vassar.edu/~lowry/fisher.html),
given that some of my observed values exceed 2000, the function returns an
error message that my sample size is too large, and recommends that I try
chi-square!
Does anyone have any suggestions about this one?
thanks very much
David Oatway
----------------------------------------------------------------------------
David Oatway
Centre for Land Use and Water Resources Research
Porter Building
University of Newcastle
Newcastle upon Tyne
NE1 7RU
United Kingdom
Tel: +44 191 222 5956
Fax: +44 191 222 6563
URL: http://www.cluwrr.ncl.ac.uk
----------------------------------------------------------------------------
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|