Hi Glenys
GL> 1. Should it be used in r x c tables that are not 2 x 2 but where cell
GL> counts are less than 5?
The extension of Fisher's exact test to tables bigger than 2x2 was
difficult to compute by hand, tha's why it wasn't routinely used when
Exp cells were lower than 5. If you have SPSS EXACT TESTS MODULE
installed in your computer, then you can (and should) use it in that
situation. Anyway, Cochran's rule says that if less than 20% of the
cells have frequencies less than 5, but over 1, then the asymptotic
significance provided by chi-square test is reliable.
GL> 2. Should it be used if minimum expected count is less than 1?
Yes, since Cochran's rule doesn't allow the use of the chi-square if
there are expected counts under 1.
GL> 3. Does SPSS perform Fisher's permutation test?
To my knowledge, no.
--
Regards,
Dr. Marta García-Granero,PhD mailto:
[hidden email]
Statistician
---
"It is unwise to use a statistical procedure whose use one does
not understand. SPSS syntax guide cannot supply this knowledge, and it
is certainly no substitute for the basic understanding of statistics
and statistical thinking that is essential for the wise choice of
methods and the correct interpretation of their results".
(Adapted from WinPepi manual - I'm sure Joe Abrahmson will not mind)
Locked
