Why would SPSS CROSSTABS not do exact or Monte Carlo versions of
McNemar-Bowker test? (vers. 22).
McNemar-Bowker test is to test for symmetry in kxk frequency
cross-table of pre- vs post-treatment results. Crosstabs always
returnes asymptotic test only even if exact version was requested.
I've heard that the exact version of the test, suitable for small
counts, exists:
http://stats.stackexchange.com/a/65552/3277. Why is
it missing in SPSS.
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
