Date:         Wed, 7 Jul 1999 14:09:38 -0500
Reply-To:     "Nichols, David" <nichols@SPSS.COM>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         "Nichols, David" <nichols@SPSS.COM>
Subject:      Re: Wilcoxon Matched Pairs Question

Yes, any case missing on either pre or post will be deleted from the
analysis. The Wilcoxon test is based on rankings of difference scores, and
thus any case for which a difference score cannot be computed because one or
both of pre and post is or are missing will be deleted.

David Nichols
Principal Support Statistician and
Manager of Statistical Support
SPSS Inc.

> -----Original Message-----
> From: Randy Richter [mailto:richterrr@SLU.EDU]
> Sent: Thursday, July 01, 1999 1:19 PM
> To: SPSSX-L@LISTSERV.UGA.EDU
> Subject: Wilcoxon Matched Pairs Question
>
>
> A friend is comparing survey responses pre to post clinical
> internship.
> Unfortunately some data is missing from both time periods (people did
> not answer all questions).  Am I correct in assuming that any results
> from the Wilcoxon will be calculated using only  cases where there is
> data for both the pre and post time periods?  The missing
> values is set
> to Exclude cases case-by-case.  I am using SPSS ver 8.02.
>
> Randy R. Richter
> Saint Louis University
>