Hi. A reply from one of our statisticians:
"
I believe our asymptotic z-statistic is based on the positive ranks. Our table has a clear footnote on this, unless we are doing something wrong. Besides, conducting the test on the negative ranks will end up with the same statistical inference."

------------------------------
Rick Marcantonio
Quality Assurance
IBM
------------------------------
Original Message:
Sent: Wed October 19, 2022 01:09 PM
From: Clifton Hamilton
Subject: z score for Wilcoxon Signed Rank Test
Can anyone explain why (and how) SPSS bases the z score on negative ranks when conducting a Wilcoxon Signed Rank Test using ordinal data?
#SPSSStatistics