home *** CD-ROM | disk | FTP | other *** search
/ NetNews Usenet Archive 1992 #16 / NN_1992_16.iso / spool / bit / listserv / statl / 1185 < prev    next >
Encoding:
Text File  |  1992-07-22  |  910 b   |  26 lines
  1. Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
  2. Path: sparky!uunet!paladin.american.edu!auvm!UNHH.UNH.EDU!W_URBAN
  3. X-Vmsmail-To: @STAT-L
  4. Message-ID: <920722083615.202116e5@UNHH.UNH.EDU>
  5. Newsgroups: bit.listserv.stat-l
  6. Date:         Wed, 22 Jul 1992 08:36:15 -0400
  7. Sender:       "STATISTICAL CONSULTING" <STAT-L@MCGILL1.BITNET>
  8. From:         W_URBAN@UNHH.UNH.EDU
  9. Subject:      Multiple comparisons w/ repeated measures or split plot
  10.               structures
  11. X-To:         STAT-L@VM1.MCGILL.CA
  12. Lines: 12
  13.  
  14.     Under the new PROC MIXED in SAS, there is a /CL switch (confidence
  15.     limits) for the LSMEANS statement.  It would be interesting for
  16.     someone to compare that output w/ that from SPSS.
  17.  
  18.     n.b.  With unequal numbers of subjects per treatment, no two
  19.     methods/packages that I have used have given the same F ratios.
  20.  
  21.     Willard Urban
  22.     Biometrics
  23.     University of New Hampshire
  24.  
  25.     will@unh.edu
  26.